TOWARD A THEORY OF WEB-MEDIATED KNOWLEDGE SYNTHESIS: HOW
ADVANCED LEARNERS USED THE WEB TO CONSTRUCT KNOWLEDGE ABOUT
CLIMATE CHANGE BEHAVIOR
By
Michael DeSchryver

A DISSERTATION
Submitted to
Michigan State University
in partial fulfillment of the requirements
for the degree of
DOCTOR OF PHILOSOPHY
Educational Psychology and Educational Technology
2012

ABSTRACT
TOWARD A THEORY OF WEB-MEDIATED KNOWLEDGE SYNTHESIS: HOW
ADVANCED LEARNERS USED THE WEB TO CONSTRUCT KNOWLEDGE ABOUT
CLIMATE CHANGE BEHAVIOR
By
Michael DeSchryver

This dissertation utilized a multiple case study design to explore how advanced learners
synthesize information about ill-structured topics when reading-to-learn and reading-to-do on the
Web. Eight graduate students provided data in the form of think-alouds, interviews, screen
video, digital trails, and task artifacts. Data analysis was based on abductive coding, first
examining synthesis through the theoretical lenses of reading comprehension, Cognitive
Flexibility Theory on the Web, and creativity, followed by a constant comparative exploration of
emergent phenomenon in the data. The empirical findings from this study provided the
foundation for a Theory of Web-mediated Synthesis comprised of interdependent elements divergent keyword search phrases, synthesis for meaning, in-the-moment insights, repurposing,
reinforcement and note-taking - which together lead to creative syntheses. Illustration and
elaboration of these elements are provided in the context of two in-depth case studies. In doing
so, this dissertation provides a post-comprehension lens through which to better explore and
understand generative reading and learning activities on the Web.

Copyright by
MICHAEL DESCHRYVER
2012

ACKNOWLEDGMENTS

I would like to first thank my dissertation committee – Ellen Cushman, Doug Hartman,
Punya Mishra, and Rand Spiro – for their time, attention, insight, and support. In particular,
Thank you Punya for your guidance and mentorship as both advisor and dissertation committee
director. Aside from keeping me on task and providing thoughtful encouragement since my very
first day in your proseminar, you taught me the true nature of creative thinking, and have
changed the way I see the world. Special thanks also to Rand for sharing your ideas with me.
Your wisdom both challenged and inspired me throughout the program, and was invaluable to
my development as a scholar.
I would like to express my gratitude to the Spencer Foundation/National Association of
Education and the Michigan State Graduate School. The financial support provided by these
agencies allowed the ideas in this dissertation to incubate. The results of this project would have
been much different without this support.
Finally, I want to thank my family. To Elyse and Audrey, you may never know the
critically important role you played in my life as I progressed through this doctoral program and
dissertation. Thank you for sharing your laughter and unconditional love. You provided me with
inspiration when inspiration was lacking, and balance when it was hard to find. To Jamie, I think
you do know that I never could have survived the last seven years without your unconditional
love and support. Thank you for trusting in me. I will spend the rest of my life repaying you in
every way that I can.

iv

TABLE OF CONTENTS

LIST OF TABLES .............................................................................................................. viii
LIST OF FIGURES .............................................................................................................. ix
INTRODUCTION ................................................................................................................. 1
CHAPTER 1
THEORETICAL FRAMEWORK ......................................................................................... 6
Reading Comprehension Research ............................................................................ 6
Prior knowledge ............................................................................................. 7
Informational text comprehension ................................................................. 7
Inferencing ..................................................................................................... 9
Summarizing ................................................................................................ 10
Multiple text comprehension ....................................................................... 11
Hypertext and Web-based Reading Research .......................................................... 13
Hypertext reading......................................................................................... 13
Web-based reading....................................................................................... 15
The new literacies of online reading comprehension ...................... 16
Constructive responsive reading in the Internet............................... 18
Other considerations for Web-based reading ................................... 20
Cognitive Flexibility Theory.................................................................................... 26
Creativity………...................................................................................................... 30
(Re)Defining Synthesis ............................................................................................ 33
Synthesis for meaning .................................................................................. 36
Generative synthesis .................................................................................... 37
Virtual conceptual simultaneity (VCS) synthesis ........................... 37
Creative synthesis ............................................................................ 38
Chapter Summary .................................................................................................... 39
CHAPTER 2
METHOD…………….. ...................................................................................................... 42
Chapter Introduction ................................................................................................ 42
Design …………...................................................................................................... 42
Participant Selection Procedures.............................................................................. 42
Tasks…………………. ........................................................................................... 45
Data Collection ........................................................................................................ 47
The Research Sessions ............................................................................................. 48
Data Analysis…… ................................................................................................... 53
Subjectivity in Qualitative Research ........................................................................ 56
Chapter Summary .................................................................................................... 60

v

CHAPTER 3
FINDINGS ACROSS ALL EIGHT PARTICIPANTS
Chapter Introduction ................................................................................................ 61
Overview………… .................................................................................................. 62
Divergent keyword search phrases .............................................................. 62
Synthesis for meaning .................................................................................. 64
In-the-moment insights ................................................................................ 65
Repurposing synthesis ................................................................................. 66
Reinforcement synthesis .............................................................................. 67
Creative synthesis ........................................................................................ 68
Background knowledge ............................................................................... 68
Soft resources ............................................................................................... 70
Note-taking .................................................................................................. 71
Chapter Summary .................................................................................................... 72
CHAPTER 4
IN-DEPTH CASE STUDY #1 – OLIVIA ........................................................................... 73
Chapter Introduction ................................................................................................ 73
Olivia…………………............................................................................................ 74
Divergent keyword search phrases .............................................................. 75
Synthesis for meaning .................................................................................. 80
In-the-moment insights ................................................................................ 82
Repurposing synthesis ................................................................................. 85
Reinforcement synthesis .............................................................................. 86
Synthesis in multiples .................................................................................. 86
Creative synthesis ........................................................................................ 88
Conclusions – Case Study #1 ................................................................................... 98
Chapter Summary .................................................................................................. 107
CHAPTER 5
IN-DEPTH CASE STUDY #2 – BETH ............................................................................ 109
Chapter Introduction .............................................................................................. 109
Beth…………… .................................................................................................... 110
Divergent keyword search phrases ............................................................ 111
Synthesis for meaning ................................................................................ 116
In-the-moment insights .............................................................................. 117
Repurposing synthesis ............................................................................... 119
Reinforcement synthesis ............................................................................ 120
Creative synthesis ...................................................................................... 124
Conclusions – Case Study #2 ................................................................................. 129
Chapter Summary .................................................................................................. 134
CHAPTER 6
TOWARD A THEORY OF WEB-MEDIATED KNOWLEDGE SYNTHESIS.............. 136
Chapter Introduction .............................................................................................. 136
Background……….. .............................................................................................. 136

vi

Divergent Keyword Search Phrases....................................................................... 138
Synthesis for Meaning ........................................................................................... 141
In-the-Moment Insights ......................................................................................... 143
Repurposing Synthesis ........................................................................................... 145
Reinforcement Synthesis ....................................................................................... 148
Note-Taking….. ..................................................................................................... 151
Creative Synthesis .................................................................................................. 155
Web-Mediated Knowledge Synthesis .................................................................... 156
CHAPTER 7
SUMMARY……………… ............................................................................................... 158
Introduction…….. .................................................................................................. 158
Synthesis for Meaning and Generative Synthesis .................................................. 158
Summary of Findings from Research Questions ................................................... 159
RQ1…… ..................................................................................................... 159
RQ2…… ..................................................................................................... 160
RQ3…… ..................................................................................................... 162
RQ4…… ..................................................................................................... 163
RQ5…… ..................................................................................................... 164
The Theory of Web-Mediated Knowledge Synthesis ............................................ 165
Implications of the Theory ..................................................................................... 167
Descriptive................................................................................................... 167
Inferential .................................................................................................... 169
Application .................................................................................................. 169
Limitations…….. ................................................................................................... 171
APPENDICES………. ...................................................................................................... 172
Appendix A: Profile for Olivia .............................................................................. 173
Appendix B: Olivia’s Synthesis for Meaning Example ......................................... 176
Appendix C: Olivia’s Synthesis for Meaning – Multiple Perspectives from a Soft
Resource ...................................................................................................... 177
Appendix D: Olivia’s Path to Synthesis for Meaning based on Verbatim Notes .. 179
Appendix E: Olivia’s Insights Related to the Text(s) Read ................................... 181
Appendix F: Olivia’s Repurposing Examples ....................................................... 183
Appendix G: Olivia’s Reinforcement Examples ................................................... 187
Appendix H: Olivia’s Creative Synthesis #1 ......................................................... 192
Appendix I: Olivia’s Creative Synthesis #2........................................................... 195
Appendix J: Olivia’s Creative Synthesis #3 .......................................................... 199
Appendix K: Profile for Beth ................................................................................. 207
Appendix L: Beth’s Synthesis for Meaning Examples .......................................... 210
Appendix M: Beth’s Insight Examples .................................................................. 212
Appendix N: Beth’s Reinforcement Examples ...................................................... 216
Appendix O: Beth’s Creative Synthesis #1 ........................................................... 218
Appendix P: Beth’s Creative Synthesis #2 ............................................................ 222
REFERENCES………. ..................................................................................................... 224

vii

LIST OF TABLES

Table 1: Descriptive Data of Synthesis Elements, All Participants ......................................62
Table 2: Descriptive Data from Keyword Search Phrases Used, All Participants ...............63
Table 3: Average Length of Keyword Search Phrases, All Participants ..............................64
Table 4: Descriptive Data from Web Sites Visited, All Participants ....................................70
Table 5: Descriptive Data for Olivia, By Task .....................................................................74
Table 6: Keyword Search Phrases Olivia, Reading-to-Learn ...............................................75
Table 7: Keyword Search Phrases Olivia, Reading-to-Do ...................................................79
Table 8: Descriptive Data for Olivia, Synthesis Elements per Minute ...............................100
Table 9: Selected Keyword Search Phrases, Olivia ............................................................102
Table 10: Descriptive Data for Beth, By Task ....................................................................110
Table 11: Keyword Search Phrases, Beth, Reading-to-Learn ............................................112
Table 12: Keyword Search Phrases, Beth, Reading-to-Do .................................................116

viii

LIST OF FIGURES

Figure 1: Image: Use of multiple tabs in the browser ............................................................51
Figure 2: Image: Computer based versus written notes .........................................................52
Figure 3: Image: Text from Five Reasons Why We Don’t Care About Climate Change ......87
Figure 4: Image: Bulleted paragraphs in Web text ..............................................................122
Figure 5: Image: Bulleted lists in Web text .........................................................................123
Figure 6: Image: Computer based versus written notes .......................................................153
Figure 7: Image: The Theory of Web-Mediated Knowledge Synthesis ..............................157

ix

INTRODUCTION

This project demonstrates how eight advanced learners used the Web as they explored
and applied ideas about an ill-structured topic. A post-comprehension perspective guided the
inquiry, focusing on the generative reading experiences of participants. That is, the study
examined how these advanced learners constructed knowledge that was neither explicit nor
implied in the texts they visited, through a multifaceted process of Web-mediated synthesis.
The study provides five key contributions to the understanding of knowledge synthesis on
the Web. First, based on a thorough review of existing literature, it re-conceptualizes the notion
of synthesis by identifying and differentiating a distinction that to date has not been made clear
in the literature, that of synthesis for meaning versus generative synthesis. Second, based on this
re-conceptualization, the data provided across eight participants, and in-depth case studies of two
of these participants, it proposes a theory for Web-mediated knowledge synthesis. This theory
accounts for isolated incidents of knowledge synthesis in online environments and also explains
how elaborate, iterative, and layered activities may ultimately lead readers to creatively
synthesize knowledge that results in solutions to complex and ill-structured problems. Third,
this study demonstrates that Web-mediated synthesis is itself complex, multiplicitous,
unpredictable, and ill structured. Fourth, it explores the role of background knowledge in the
synthesis process, differentiating it from idea-play and insight. Finally, this study presents how
note-taking is integral to the process of Web-mediated synthesis, serving important functions that
support all its various elements.

1

Several trends motivated this study. The first was the ongoing rapid technological
change all around us, particularly as it relates to the Web. Given the ubiquity of the Web in
modern culture, it is now a mainstay for educational, professional, and personal learning. At
school, 90% of undergraduates use the Internet to look for information about their studies
(Selwyn, 2008) easily preferring the Web to library resources (Griffiths & Brophy, 2005; Van
Scoyoc, 2006). At work, 75% of professionals, executives, managers, and government workers
use the Internet several times a day to complete job-related activities (Madden & Jones, 2008).
And, at home, the Internet is used more commonly to explore common problems than advice
from professionals, family, friends, newspapers, magazines, television or the library (Estabrook,
Witt, & Rainie, 2007). This environment provides the foundation for a new ecology of
information, of reading, of knowledge, of teaching, of learning, and of thinking.
This trend has far-reaching implications. Web users can take advantage of its ubiquity,
the well directedness of search, and the “ambient findability” (Morville, 2005) of a seemingly
unlimited scope of information, to harness an unprecedented adjunct to human memory. They
can devote less time and cognitive energy to remembering what the texts they encounter say and
revisit the meaning of those texts at any given moment. This can happen through the original
online text itself, inline annotations or markups, or online database collections of user selected
text segments. As such, use of the Web may allow for more time to engage in higher order
thinking, such as creativity, analysis, integration, and problem solving (Anderson & Rainie,
2010).
The second trend motivating this study was school related. Technology “integration”
initiatives still largely result in the widespread perception that technology is merely a “tool”
(e.g., Schrum, 2005; Thieman, 2008) or an “information resource” (Kuiper & Volman,

2

Handbook, 2008) to supplement the traditional learning process. In fact, Edyburn (2003)
suggested that many educators have a bias for the knowledge contained in one’s head over that
which can be facilitated with external devices, indicting a preference for educational
performance through “naked independence.” This has resulted in a disconnect between how
these tools are used in schools and in the workforce. For instance, while students do not have
access to the Web during high-stakes testing in the United States, they will have access to such
technology resources in nearly every professional context for the rest of their lives. Policy
makers in Denmark are starting to realize these issues and have made accommodations. Pilot
initiatives in Denmark include laptops connected to the Web during high-stakes testing and the
test questions do not ask students to regurgitate facts and figures (Hobson, 2009). Consequently,
the present study was designed to begin to inform educators in the United States, and elsewhere,
as to the foundational and potentially revolutionary implications this new ecology is effecting on
thinking, reading, and learning, opening up more forward thinking educational environments like
those in Denmark.
The third trend was also cultural. We live at a time in history of unprecedented
complexity, which increasingly values integrated decision-making and creative solutions. From
nutritional choices (Gunther, 2010), to health care options (Thaler & Sunstein, 2008), to
retirement savings decision-making (Blaufus & Ortlieb, 2009), to prescription drug plan choices
(Hanoch, Wood, Barnes, Liu, & Rice, 2011); individuals need to know how to harness the power
of Web-based content to make informed choices in their lives. At work, fields from accounting
(Friedman, 2004) to genetic engineering (Hollander, 2004) to teaching with technology (Mishra
& Koehler, 2007) require creative solutions to complex, ill-structured, or “wicked,” problems. A
number of these issues are facing our citizens today on a broad scale, such as multi-party

3

regional water issues (Adams, 2001); genetic engineering (Hollander, 2004); cyber-terrorism
(Mitroff, Alpasian & Green, 2004) and international economics (Schwab, 2001). Just knowing
how to find or even understand information on the Web will not help develop solutions to any of
this. We no longer live in an “information economy” but in an “age of complexity” (Schwab,
2011) for which new ways of thinking about thinking are required.
The fourth motivating trend emanated from the field of literacy. A variety of
perspectives within this field explore the new, multi, or informational literacies of online reading
and learning (e.g., see Hartman, 2000; Leu, Kinzer, Coiro, & Cammack, 2004; Moje, 2009; New
London Group, 1996), many of which focus on the skills and strategies needed to understand
and make meaning of the multiple texts provided on the Web (Afflerbach & Cho, 2009;
Warshcauer & Ware, Handbook, 2008). Several of these perspectives also focus on the literacy
skills and strategies needed to produce a variety of media elements (text, video, audio,
multimedia) based on the near instant read/write capabilities of the Web. Literature on both
media understanding and media production often incorporate the idea of synthesis, as defined by
bringing elements together from various and often disparate resources. However, little attention
is given to the potential for generative thinking (e.g., DeSchryver & Spiro, 2008; Coiro, 2009;
Spiro, 2006a, b, c, d, e, f) when learners “act with the meaning” (Cho, 2011). Such Webmediated generative thinking may bridge online media comprehension and media production by
supporting the construction and flow of new ideas, resulting in a broader conception of synthesis,
where synthesis becomes a creative activity.
The purpose of this study is to provide a theory from which to explore Web-mediated
synthesis in greater depth, using the following broad question as a starting point: How do
advanced learners synthesize new ideas in a Web-mediated learning environment? In this first

4

chapter, I explore the four theoretical lenses that guided my original understanding of synthesis
on the Web. I then integrate these four perspectives and operationalize the term “synthesis” as
starting point for the study, and conclude with the specific research questions that emerged from
this literature review. Thereafter, Chapter 2 outlines the method utilized in this study, by
describing the design, the tasks, and data collection and analysis procedures. Chapter 3 presents
general findings across all eight participants. Chapters 4 and 5 report additional findings based
on in-depth case studies of two individual participants. Finally, Chapter 6 summarizes and
discusses the implications of these findings.

5

CHAPTER 1
THEORETICAL FRAMEWORK
Synthesis is a term used widely in both education and educational research. This means
it has also been defined in a variety of ways. Therefore, at the outset of this study, I explored four
separate theoretical perspectives to provide a foundation from which to operationalize synthesis:
(1) reading comprehension research (2) hypertext and Web-based reading research, (3) Cognitive
Flexibility Theory, and (4) creativity.
Reading Comprehension Research
Given that online reading utilizes many of the processes that are critical to offline reading
(Leu, et al., 2008), I first explored the general literature on reading comprehension. Theories
explaining offline reading indicate that extraction and construction of meaning from text (i.e.
comprehension) is the result of the myriad possible transactions among various reader
characteristics, text attributes, purpose/task, and context (Pearson, 2009; Rand, 2002). The
phrases “reading-to-learn” (Sticht, 1979) and “reading-to-comprehend” (Kintsch, 1988;
Mosenthal & Kirsch, 1991) describe reading activity where the primary purpose is to understand
the text. However, once the meaning is understood, readers may also establish a “deep
understanding of the subject matter, so that information acquired can be used productively in
novel environments” (Kintsch, 1994, p. 294), or for a particular purpose or situational need
(National, 2008). This latter conception of reading is often considered “reading-to-do” (e.g.,
Mikulecky & Drew, 1991; Sticht, 1975).
For the purposes of this study, the reading environment comprised of advanced learners
reading multiple informational texts in both “reading-to-learn” and “reading-to-do” contexts. The
most important contributions from offline reading comprehension research to the development of

6

this study derived from five areas: (1) prior knowledge; (2) informational text comprehension;
(3) inferencing; (4) summarizing; (5) and, multiple text comprehension.
Prior knowledge. One of the primary activities documented by proficient readers is the
ability to access their prior knowledge and connect or relate it to the text they are reading
(Afflerbach, 1990; Anderson & Pearson, 1984; Clymer, 1968; Kintsch, 1998; Kintsch &
Rawson, 2005). Prior knowledge includes general world knowledge, subject matter knowledge,
knowledge of text structure (Coiro, 2007), and epistemic knowledge (Kardash & Howell, 2000).
All contribute to successful understanding of text (Afflerbach, 1986; Weaver & Kintsch, 1991).
Kintsch (1994) described this interaction between the text and the reader’s background
knowledge and experience as a construction-integration model whereby the information provided
in a text (the internal meaning of the text, or text base) must be elaborated from the prior
knowledge and then integrated with it to form the readers interpretation of the text (i.e., the
situation model). Prior knowledge facilitates inferencing (Graesser, Singer & Trbass, 1984;
Kintsch, 1990; Tarchi 2010), the process of determining the main idea when it’s not explicitly
stated (Afflerbach, 1990), text search success (Symons & Pressley, 1993), prediction strategy
usage (Afflerbach, 1990), and the ability to link across texts (Strums & Braden, 2002). It is
important to note, however, that prior knowledge can be both accurate and inaccurate (e.g.,
Kendeou & van den Broek, 2005).
Informational text comprehension. Significant differences appear in the literature
when considering narrative versus informational text, including structure and purpose. While
narrative texts primarily communicate a story, informational texts communicate information
about the natural or social world, persuade people to think or do something, give directions, or
tell a true story (Duke, 2010). Traditionally, informational texts include difficult and complex

7

topics, specific organizational structures (Chambliss, 1995), and specialized vocabulary, which
make the text more difficult to understand and use (Coiro, 2007). The purpose of informational
text is often to get the gist of, locate, or remember particular information (Brown, 2003; Dreher,
1993; Guthrie & Kirsch, 1987).
There are four categories of informational texts: (1) exposition, (e.g., textbooks, news
stories and trade books); (2) argumentation and persuasive text (e.g., speeches, editorials,
advertisements); (3) procedural text and documents (e.g., manuals, directions) (National, 2008);
and, (4) nonfiction narrative (Duke, 2010). The arrangement of these varies, but they commonly
follow one of five structures. A descriptive text provides attributes, specifics or setting
information on a topic. A sequential text provides ideas grouped in order or time. A causal text
provides cause-and-effect relationships between different ideas. A problem/solution text is
divided into two parts of question and answer. Finally, a comparison text presents similarities
and differences in a way that facilitates comparison, through contrasting or alternative
perspectives. Successful readers of informational texts are able to make inferences about these
different text structures to aid in their comprehension. They are able to identify connections
between the text context and their background knowledge, and summarize text to express
important ideas (Coiro, 2007; Kintsch & Kintsch, 2005; Weaver & Kintsch, 1981). Background
knowledge of these structures and how to use them for comprehension is presumed to come from
school-based training (Pearson & Fieldsing, 1991); however, school children typically get very
little training in how to successfully read informational text (Duke, 2000).
In order to comprehend informational text, the reader must be able to seek and locate text
information effectively. Several components have been noted to comprise this ability: (1)
formulating a goal; (2) category inspection; (3) information extraction; (4) integration with

8

previous knowledge; and, (5) recycling until the goal is achieved (Guthrie & Kirsch, 1987;
Guthrie and Mosenthal, 1987). In many cases, informational text comprehension is supported by
the use of key search terms. Readers find information more easily when given these terms
(Dreher, 1992), and single terms are easier to use than complex multi-word terms (Brown, 2001).
This sort of information extraction is strategic and selective and is not considered reading to
learn (Brown, 2003). Text characteristics in traditional offline texts that aid in information
seeking behavior include: navigational tools (e.g., glossary, table of contents), titles, headings,
typographic schemes (e.g. bolding, color, or punctuation), and graphic design (Brown, 2003).
Inferencing. Making inferences about text content is one of the nine basic reading skills
(Davis, 1944) and is at the heart of the comprehension process (Dole, Duffy, Roehler & Pearson,
1991). It is a higher order activity (Cho, 2011; Greasser & Kreuz, 1993; Greasser, Mill, &
Zwaan, 1997) based on the interactions among the text and the previous knowledge of the reader.
Inferencing fills in details that are not explicitly stated in the text and allows the reader to
elaborate on what is read (Dole, et al., 1991). Inferences can occur in the moment or when
prompted after reading and are often a conscious act. Knowledge-based inferences result from
activating generic and specific prior knowledge, the latter typically taking longer than the former
(Graesser, Swamer, Baggett & Sell, 1996). Novel knowledge-based inferences build up a
referential situation model of understanding (i.e. mental model) only after iterative interactions
between prior knowledge and accumulating knowledge from multiple text sources (Just &
Carpenter, 1992; Graesser & Britton, 1996). The result of inferencing is coherence at both local
and global levels of the text.
Kintsch (1998) indicated that some inferences come from simple retrieval, while others
result from active memory processes generating new information (Rouet, 2006). Automatic

9

retrieval inferences result from topic-specific associations between information in working
memory and prior knowledge. Controlled retrieval inferences require a more conscious and
extended search of memory for specific links based on cues in short term memory. Automatic
generative inferences result from general world knowledge about the evolving mental
representation of source material. Additionally, controlled generative inferences occur when
deductive reasoning and logic help to fill in gaps in the text, which relies heavily on prior
knowledge (Kintsch, 1998; Rouet, 2006). Together, these devices allow readers to “properly
understand a text” (Kintsch, 1998, p. 189).
Summarizing. The ability to summarize text is also important to successful reading
comprehension (Dole, Duffy, Roehler & Pearson, 1991). Good readers “sift through large units
of text, differentiate important from unimportant ideas, and then synthesize those ideas and
create a new coherent text that stands for, by substantive criteria, the original” (p. 244). These
summaries can be “writer-based” or “reader-based.” Writer-based summaries typically occur
from text that is largely unfamiliar, and they are best written when the text is being read. The
summary writer is not accountable for the amount of material summarized or the quality and
mechanics of the summary. Writer-based summaries are, in essence, summary notes taken by
the reader. Reader-based summaries are typically assigned tasks summarizing longer sections of
text (e.g., an entire book or article) with more strict parameters on the length, grammar, and
cohesiveness. As such, reader-based summaries change the goal of the task from comprehension
of text to composition. Research suggests that the process of making of either type may facilitate
learning (Brown, Campione & Day, 1981).
Hidi and Anderson (1986) described the process of text summarization in great detail,
which includes comprehension, evaluation, condensation, and transformation of ideas in the

10

original text. During summarization, some sections of the text are deleted, while others are
purposefully included in the summary. Condensation occurs by substituting higher level, more
general concepts for lower level concepts with more detail. Finally, integration, combination,
and transformation of text occur to provide the most accurate representation of the original text.
According to Hidi and Anderson, “the major concerns of the summary writer, therefore, are not
how to plan and generate new content…” (p. 472, emphasis added), but how to understand and
encapsulate meaning that already exists within the text.
Multiple text comprehension. The comprehension of multiple texts is a complex task
based on comparing, contrasting, relating, and differentiating information across various texts
(Afflerbach and Cho, 2009). This can involve building intertextual links (Afflerbach &
VanSledright, 2001; Hartman, 1995; Stromso, Braten, & Samuelstuen, 2003; Wolfe & Goldman,
2005), and requires constant revision of one’s own understanding of and reaction to previous text
(Hartman, 2000). Cho (2011) describes this process succinctly
readers learn the content conveyed across different texts, relating the currently read text
to previous texts, cross-referencing and extracting related information, assembling
different ideas into globally coherent meaning, and continuously elaborating a crosstextual mental model—synthesis strategies are highly required in the construction of
cross-textual meaning (p. 64, emphasis added)
Not only does comprehension occur for each document, readers must integrate the information
across sources, and remember the source for new information (Rouet, 2006). There are three
differences from single text comprehension. First multiple text comprehension focuses on
source information. Second, readers of multiple texts must make distinctions between and
corroborate information across documents. In doing so, multiple text comprehension may lead to

11

the updating of previous knowledge or beliefs (Johnson & Seifert, 1999). Finally, the multiple
documents may complement each other and “fill in the gaps” of other documents by confirming
or disconfirming inferences the reader may have made about them. Because each document
typically only contributes to the overall understanding of the topic or situation, the reader is
responsible for recognizing global relationships that can be integrated into a coherent whole.
These relationships manifest in several different ways and are understood through various
strategies. The Theory of Documents Representation describes how these relationships emerge
between and among sources. Based on this theory, content from multiple sources may include
links that (1) support/oppose one another; (2) agree/disagree with one another; (3) show how
texts are based on each other; (4) indicate a temporal relationship; (5) demonstrate the spirit of
one another; and, (6) show how they may be relevant to each other (Braten, Stromso & Britt,
2009; Perfetti, Rouet & Britt, 1999). Hartman (1995) demonstrated that these connections may
result from linking the same text (primary endogenous), another text in the current context
(secondary endogenous), or texts outside of the current context (exogenous). Afflerbach and
Cho (2009) identified a variety of multi-text linking strategies, including recursive reading and
theme identification across texts, prediction of text content based on understanding of previous
texts, identification of tentative meanings using other sources to reduce ambiguities, and building
understanding through re-reading two or more texts to find links regarded previously as
unrelated. These strategies are used to both deconstruct and reconstruct the linkages among texts
(Hartman, 1994), and are critical to understanding across the landscape of texts available
(Afflerbach & Cho, 2009).
Stromso, Braten, and Samuelstuen (2003) suggested that these relationships facilitate
memorization, organization, elaboration, and monitoring, all of which can lead to greater textual

12

understanding. Memorization involves the selection and rehearsal of text information without
moving beyond the content, whereas elaboration requires the reader to make intertextual links
that make the content more meaningful, such as by offering analogies, providing examples, or
describing personal experiences. Organization is when readers relate, group or order ideas from
the texts, and monitoring is when readers assess or regulate their comprehension of the text.
Integrating this information within the developing situation model results in a document model
of coherent understanding that reflects not just one situation but a range of situations reflected by
the documents considered together (Braten, Stromso & Britt, 2009; Perfetti, Rouet &Britt, 1999).
Given the possibility of an “intertextual loop” (Hartman, 1994, p. 618) as readers revisit
documents multiple times and create a mental web of meaning (Stromso & Braten, 2002), the
reader’s comprehension of the text can be continually updated.
Hypertext and Web-based Reading Research
Hypertext reading. The advent of hypertext systems for reading and learning brought
about new considerations for reading comprehension (Bolter, 1998; Landow, 2006; Reinking,
McKenna, Labbo, & Keiffer, 1998). Landow described hypertext as “an information medium
that links verbal and non-verbal information…thereby creat[ing] text that is experienced as nonlinear, or more properly, as multilinear or multisequential” (p. 3). This information includes text,
video, sounds, animations, images, maps, and diagrams and there is not a specific beginning or
endpoint. Bolter (1998) indicated that the primary feature of hypertext is its fluidity; each
hypertext experience is a “set of different potential texts awaiting realization” (p. 5), which leads
to an unstable and unpredictable environment. Activating links among the pages controls
movement from page to page and readers determine their experience both by their own decisions
and by the author’s placement of embedded connections (Bolter, Burbules, & Callister, 2000;

13

Rouet & Levonen, 1996). Finally, readers of hypertext tend to focus on the macro-level of
processing information, as compared to a micro-level for those with linear offline text
(Alexander, Kilikowich, & Jetton, 1994).
Hypertext provides complications and requires skills and abilities beyond traditional
offline text (RAND, 2002). Coiro & Dobler (2004) outlined the four primary differences between
offline text and hypertext. First, because of the non-linear and multi-sequential nature of
hypertext, readers must take a more active role in quality and coherence building. Second,
because of the “hidden” navigations elements of hypertext, and the lack of semantic clarity and
surrounding cues with hyperlinks, inference demands are higher. Third, the addition of
hyperlinked images and icons requires readers to be able to decode and interpret more visual
elements. Finally, the explicit intertextual representations create a more highly complex reading
environment, both mentally and physically (e.g., as it relates to orientation within the texts).
Similar to offline text, the reader’s experience with hypertext is primarily one of meaning
construction (Bolter, 1998). However, while offline meaning construction is primarily invisible
and internal, hypertext reading demonstrates more external manifestations of meaning making
through the choices of links followed. Because of the increased complexity provided by the
hyperlinking structure, certain skills identified for successful offline comprehension may be even
more important when reading hypertext. For instance, “abstract link labels and incoherent
transitions” (Coiro, 2007, p. 34) require increased use of inferential reasoning (Wenger & Payne,
1996). Based on such inferencing, Kim and Kamil (1999) demonstrated how successful
hypertext readers were able to make effective predictions about what information specific links
might provide.
The role of prior knowledge in hypertext comprehension is less clear than with offline

14

text. Some hypertext readers with low prior knowledge comprehension seem to have benefitted
from highly organized and coherent hypertext (e.g. see Balyctiene, 1999; McNamara & Kintsch
1996; McNamara et al, 1996; Potelle & Rouet, 2003; Salermon, et al, 2005). Others
demonstrated no benefits from structured overviews (Brinkerhoff, Klein, & Koroghlanian, 2001;
De Jong & van der Hulst, 2002; Hofman & van Oostendorp, 1999; Mueller-Kalthoff & Moeller,
2003; Puntambekar, Stylianou, & Hübscher, 2003). This suggests that the “unstable” structure
of hypertext may have an interaction with prior knowledge levels.
Finally, the, multiplicity of hypertext is important to comprehension. Salermon, et al.,
2005) demonstrated that the more different texts read by a reader, the better textbase
comprehension that emerged, especially for low prior knowledge readers. Salermon further
suggested that this finding might indicate that high prior knowledge readers can fill in gaps from
unread resources. Alternatively this may also suggest that the multiplicity and sheer scope of
hypertext systems may allow for low prior knowledge readers to fill in their own gaps through
well-directed and rapid movement around the knowledge landscape (e.g., see DeSchryver &
Spiro, 2008).
Web-based reading. When readers move from more “closed” hypertext systems to an
open online environment (i.e., the Web), the reading environment increases in complexity yet
again (Coiro & Dobler, 2004; Eagleton & Dobler, 2007; Hartman, et al., 2010; Zhang & Duke,
2008).

The scope of the information available on the Web as compared to closed systems

provides qualitative differences in the effects of information multiplicity. The structure, form
and context can change from day-to-day. Advertisements are common distractors to the reader’s
purpose. Web texts are more likely reflect hidden social, economic, and political goals (Coiro &
Dobler, 2004). Modern Web-based search tools (e.g., Google) provide a wider range of services

15

and options than closed system searches, utilize sophisticated artificial intelligence algorithms,
are “traceable,” and can produce results in various different forms (e.g., visual, timeline). The
rapid evolution of computing speed, server space, compression algorithms, and Web page design
technology has facilitated increased multi-modal integration of more polished elements with
greater attention to layout, visual, and graphic design. These modes of information delivery
include newly available elements such as video with text comments embedded and “mash-ups”
of multiple modes of data (e.g., maps with text, image and video data embedded). The Web is
also a more interactive read-write medium employing a variety of technologies to incorporate
“soft” resources (e.g. blogs) and real-time reader feedback on both soft and hard resources, such
as through comments.
Several frameworks for successful reading on the Web explain these complexities in
detail. The two that informed the development of this study were: (1) the new literacies of online
reading comprehension (Leu et al., 2004; Leu, Zawilinski, et al., 2007) and, (2) Constructive
Responsive Reading on the Internet (Afflerbach & Cho, 2009).
The new literacies of online reading comprehension. Support for the idea that new
information and communication technologies (ICTs, including the Web) require New Literacies
is strong and ever growing (e.g., Bruce, 2003; Coiro, Knobel, Lankshear & Leu, 2008; Gee,
2003; Karchmer, 2001; Leu, Kinzer, Coiro, & Cammack, 2004; Kuiper, Volman & Terwel,
2009; McKenna, Labbo, Keiffer, & Reinking, 2006). The notion of New Literacies applies to
multiple fields of inquiry, from computer-mediated communication to media literacy to
educational technology (Coiro, Knobel, Lankshear & Leu, 2008). A recent review of scholarship
in this area proposed a set of shared characteristics across the various ways of operationalizing
New Literacies. The broader New Literacies perspective is therefore identified by the following

16

assumptions: (1) the Web and other new digital technologies require new skills, strategies, social
practices and dispositions for effective use; (2) new literacies are critical to taking part in the
global ecology of digital information; (3) new literacies are rapidly changing; and, (4) new
literacies are underscored by multiplicity (i.e., multimodal, multifaceted, multiperspectival) and
may call for analysis from multiple disciplinary perspectives (Coiro, Knobel, Lankshear & Leu,
2008).
One of the most detailed frameworks available in this area is from the University of
Connecticut New Literacies Research Team, specific to online reading comprehension (e.g., see
Coiro & Dobler, 2007; Leu, Reinking, et al., 2007). It defines and delineates online reading
comprehension “as a problem-based inquiry process involving new skills, strategies, and
dispositions on the Internet to generate important questions, and then locate, critically evaluate,
synthesize, and communicate possible solutions to those problems online” (Leu, Coiro, Castek,
Hartman, Henry & Reinking, 2008, p. 323). Castek et al. (2011) provided further details on this
approach. Their perspective assumes that Internet reading almost always starts with a question,
and has the purpose of solving problems. Within that environment, generating effective keyword
search strategies, inferring website usefulness from search engine results, and scanning for
relevant material comprise the primary locational skills and strategies. These are referred to as
the “circuit breaker” for online reading comprehension (Leu, Mcverry et al. 2009). That is, if
you cannot locate information, you cannot read it. Critical evaluation of resources is considered
more important than with offline texts, given the scope and diversity of resources encountered.
Similarly, due to the multiplicity of resources, synthesis is crucial, since the reader must integrate
separate and unique ideas. Another unique feature of the online environment is the close
connection between reading and writing skills. For instance, online reading comprehension

17

includes using email, text messaging, chats, blogs, wikis, and discussion boards or video
conferencing to solve problems.
Reading online utilizes some of the traditional elements of reading comprehension, such
as locating main ideas, summarizing, inferencing, and evaluating (Coiro, 2003). However, this
process is not considered isomorphic with offline comprehension (Coiro & Dobler, 2007;
Educational Testing Services, 2005; International Reading Association, 2009; Leu, et al., 2005;
RAND, 2002; Sutherland-Smith, 2002). For example, it involves self-directed construction of
text and knowledge (Coiro & Dobler, 2007), while “readers navigate their own paths through an
infinite informational space to construct their own versions of the online texts they will read”
(Leu, et al., 2008, p. 323). Online reading comprehension also manifests as a collaborative and
social practice, whereby the unique affordances of the media lead to gains in depth of
understanding (Leu, et al, 2011). Some of the new skills needed to do this successfully include
using a search engine, reading search engine results, scanning Web pages for relevant links, and
knowing how to determine who authored information posted on a Web site. The Taxonomy of
Internet Comprehension Skills and Strategies (Taxonomy) (“Taxonomy,” n.d.) is a detailed
outline of these skills derived mostly from the study of skilled K-12 students (Leu, Reinking, et
al., 2007).
Constructive responsive reading on the Internet. Constructive Responsive Reading
(Pressley & Afflerbach, 1995) explained basic offline reading comprehension through
identifying content, content evaluation, and monitoring. Afflerbach and Cho (2008) extended
this theory to explore fundamental changes in the architecture of reading fostered by the Internet.
As they suggest, readers must address “novel or hybrid strategies that new reading situations
create” (p. 85). These include a group of strategies considered unique to online reading that

18

address the various unknowns of online reading (e.g., possible links, texts, decisions, and
interactions). They describe a process in which the reader is constantly “managing a shifting
problem space” (Afflerbach & Cho, 2009, p. 212), which they claim does not have a counterpart
in traditional reading, either in practice or theory. Specifically, readers must locate, identify and
select appropriate texts and determine the order by which they are read.
Within this framework, Afflerbach and Cho (2008) outlined the constructive reading
strategies found across multiple studies of Internet reading. These strategies are used in a
“dynamic, recursive interplay” with those identified for traditional reading. In order to realize
and construct potential texts, the reader must: (1) search for relevant Web sites; (2) focus
searches by generating keyword phrases; (3) conduct complementary searches with revised
keywords; and, (4) scrutinize, predict, and make inferences about the relevance and utility of
hyperlinks. There are many skills required to identify and learn important Web-based
information, including the ability to use search engines to construct meaning, search for
information related to already established meaning, link to related sites beyond the original goal,
use multilayer inferences across the three-dimensional space of the Web, revisit pages to revise
constructed meaning, and combine disparate forms of information to construct meaning. Taking
into consideration these “new or hybrid” strategies, several key points emerged. First, the role of
inferencing expands compared to traditional texts and includes the need to make educated
guesses about these unknowns. Inferences are now about the path as well as the content.
Second, there is an increased emphasis on intertextual reading strategies for meaning
construction, such that they become the “central aspects of the activities for constructing
meaning” (Cho, 2011, p. 86). Third, search and keyword construction are central to the process
of Web-based reading. Fourth, the goal of these strategies is to help the reader construct

19

meaning for the multiple Web texts explored. It is worth noting that while Afflerbach and Cho
did not explicitly list synthesis in their framework, their description of the act of combining
disparate forms of information from multiple websites and resources to construct meaning would
represent a synthesis-like activity.
Other considerations for Web-based reading. Many consider the role of prior
knowledge in Web-based reading environments to be equally central to the offline
comprehension process (e.g. Calisir & Gruel, 2003; Potelle & Rouet, 2003). However, this is not
always clear. For instance, Desjarlais and Willoughby (2007) found that access to the Web
facilitated learning as measured in essay writing in similar ways for both high and low domain
knowledge participants. In addition, Coiro (2011) noted, “prior knowledge of the topic may be
somehow less important when reading on the Internet” (p. 360). This may be a result of the
“new forms” of background knowledge available on the Web for in-the-moment schema
construction. In particular, this may be due to the nearly unlimited scope of resources, speed of
access, and increasingly well-directed searches available on the Web (Spiro & DeSchryver,
2010, LRA poster). At the same time, while the value of prior knowledge of the topic may be
less important, the importance of prior knowledge of online text structure, website organization,
and search engines may be even greater, increasing the complexity of Web-based reading
(Eagleton & Dobler, 2007).
Another consideration in Web-based reading context is the role that note taking plays.
Offline note taking enjoys an abundance of research. Kiewra (1989) outlined both the encoding
and storage benefits of student note-taking in lectures. When reading, though the intention and
function of note taking can be quite different than during a lecture, both the processing and
review of notes also serves to deepen mental representation and comprehension (Slotte & Lonka,

20

1999). In several studies, processing notes alone was enough to increase comprehension. This
was most often attributed to the depth of processing (Craik & Lockhart, 1972) attained. Learning
gains from offline notes are also based in large part on the quality of the notes taken. For
instance, summary notes enabled deep learning better than those taken verbatim, and
organization, restructuring, storage, and integration of notes were key features for successful
note-taking (Makany, Kemp & Dror, 2009). Verbatim notes helped the learner remember what
the text said, but summary notes better facilitated understanding the connections between the
new information and analogous situations (Kintsch & Kintsch, 1996). Makany, Kemp and Dror
(2009) also found that non-linear note-takers significantly outperformed linear note-takes in both
quantity and quality of learning. Most studies in this area assumed a model whereby learners take
notes during a lecture or while reading, review those notes prior to a test or transfer task, and
then undertake a test or transfer task without their notes. However, Slotte and Lonka (1999)
demonstrated that the availability of notes during a transfer task resulted in better performance.
Though well researched in offline contexts, attention to the use of either traditional notes
or new forms of note-taking available on the Web is minimal (e.g., see Stanford Faculty Senate,
n.d.). For instance, although the Taxonomy (“Taxonomy,” n.d.), identified note taking as an
aspect of online reading comprehension, it did so as a component of communication, with very
little detail. Though note taking on the Web does support similar intentions and functions to
those taken offline, it also offers additional affordances. Online clipping systems facilitate both
verbatim and summary notes. The availability of tags and keyword searching enhances the
learner’s ability to organize, recall, and integrate notes taken on the Web. Annotations are easily
accommodated at the time notes are taken, or during revisitation and review. And, notes taken

21

online are also available in a variety of contexts and representations, including on the original
Web page resource itself.
Equally important to how online notes can be taken or revisited is when they can be used.
The ubiquitous nature of the Web underscores learning environments where both the Web itself
and the personal Web one creates with notes and annotations are available while learning and
during transfer tasks related to that learning. Government officials in Denmark piloted a
program where its college students had access to the Web during their final year exams (Hobson,
2009). This is indicative of the new information ecology the Web is producing. The role of
online note taking and the adjunct systems to support it in this new ecology is, at present, unclear
and unstudied. As such, this study sought to identify whether note taking has a greater role in the
online reading process, in particular whether it is an important component of knowledge
synthesis for advanced learners on the Web.
Another important consideration is the potential significance of keyword selection and the
search process to Web-based reading (e.g., Eagleton & Guinee, 2002; Eagleton, Guinee &
Langalis, 2003; Guinee, Eagleton & Hall, 2003; Kuiper, Volman & Terwel, 2009). One of the
key differences between offline reading/hypertext systems and the Web is the dependence and
value of using keyword search phrases. For instance, Google is the dominant search engine on
the Web, and keyword search is the “default standard” for Web based information retrieval
(Wilson, 2009). Consequently, using a keyword search phrase is often the entry point for
reading on the Web. Henry (2006) indicated that searching may be the most important new
literacy skill for reading comprehension on the Internet, given that “all other decisions and
reading functions on the Internet emanate from the decisions that are made during the search
process” (p. 616). The results of Nachmias & Gilad’s work prompted them to claim, “Web

22

searching is a nontrivial complex skill” (2002, p. 481). Similarly, Moraveji et al. (2011) wrote,
“Formulating queries is a particularly complex task due to its reliance upon abstraction,
vocabulary, domain knowledge, and grammar. Despite this complexity, query formulation is
often the focus of explicit instruction because it is so central to Web search practice” (p. 2).
Likewise, the International ICT Literacy Panel (ETS, 2002) indicated that access and retrieval of
information is a key to ICT literacy. However, unlike those above who consider it a complex
skill, this panel listed the ability to access information (which includes searching) as requiring
the lowest cognitive complexity of the literacy process.
While the importance of this process is well documented, and both the Taxonomy and
Constructive Responsive Reading on the Internet frameworks include the use of searching and
keyword construction, neither is explicit about how this is done well. Research based on the
construction of keywords in closed systems, such as online catalogs, provides some insight into
this phenomenon. For instance, Bates (1979) outlined the use of “term tactics” in selecting and
revising terms for search formulation in manual and on-line systems; Fidel identified 18 different
“moves” in query formation for online searching; and Chen and Dhar (1990) explored query
refinement and the role of subject and system knowledge for an online catalog. However,
several considerations differentiate modern Web-based keyword search phrase development and
use from these older systems. First, in Web-based reading tasks, search phrase use is fully
integrated into the reading and learning process. Keyword search phrases guide Web-based
readers from one resource to another while they are reading them. In comparison, library catalog
search phrases used in the past were typically employed prior to the reading process. Second,
modern search engines are highly sophisticated in their interactions with users. For example,
keyword search phrase suggestions are now common in search engines like Google. Third, the

23

immediacy of results in Web searches allows for the results of one search to inform the next
iteration of search. This dynamic changes the cognitive process of keyword search development
(DeSchryver & Spiro, 2008, Spiro, 2006a, 2006b). Finally, the availability of “histories” on
keyword search phrase use provides the potential for more review and study of metacognitive
approaches to keyword search development.
Most Web searches are very simple. For example, Beitzel (2004) demonstrated that the
average query length is 2.2 terms, and that 81% of users view only one page. Nachmias and
Gilad (2002) also found that most of their undergraduate participants used trivial strategies for
keyword development (i.e., single keyword search). Unfortunately, Britt and Gabrys (2001)
demonstrated that many of these searches are unsuccessful. Studies comparing novice and expert
Web searching demonstrated that experts have more complex information seeking strategies.
Experts used more unique search terms, persisted longer, used varied strategies, visited more
diverse resources, integrated more synonyms, and changed their search terms when searches
were unsuccessful (Hembrooke, et al, 2005; Hsieh-Yee, 1993; Warwick, Rommer, Blandford,
Gow, & Buchanan, 2009; Vakkari, 2001). “Double experts” (i.e., experts in both search and
domain) selected search terms that produce better results in size and quality (Holscher and
Strube, 2000), and some experts used more Boolean phrases (Holscher & Strube) while others
did not (Aula, 2005). Collectively, this evidence may indicate that “expert” keyword phrases use
Boolean logic, and/or that experts select the correct terms and do not need Boolean functions to
be successful. Finally, many researchers have noted that prior conceptual knowledge influenced
the ability of experts to select appropriate search terms (Hembrooke, Granka, Gay, & Liddy,
2003; Vakkari, 2002; Vakkari, Pennanen, & Serola, 2003).
The process of developing keyword search phrases has many necessary features. Lau and

24

Horvitx (1999) outlined several of these activities, including: generalization, specialization,
reformulation, and interruption. Additionally, Rieh & Xie (2001) proposed that searchers must
make parallel, dynamic, and alternative reformulations of searches all in the context of content,
format and resources. Nachmias and Gilad (2002) further identified the use of general
knowledge in keyword phrase formulation, specifically describing the value of the search words
that were not mentioned in the search task bar.
The complexity of search terms is also linked to the type of task. When using a search
engine on the Web the difference between navigation and exploration purposes is particularly
important. White and Drucker (2007) found that “navigators” demonstrated few deviations in
their search paths, tacked problems sequentially, and were more likely to revisit domains. On the
other hand, the search paths of “explorers” branched frequently. Explorers submitted many
different keyword queries, and visited many new domains. The authors suggested that these
approaches to searching were related to both cognitive style and task. That is, navigators
exhibited a “serialist” cognitive style (Pask, 1976), using local learning strategies, examining one
idea at a time, and focusing on separate topics and the sequences that connect them together.
However, explorers reflected a “holist” style, which was cognitively complex, applied a global
approach to learning, and focused on the relationships between multiple ideas during the learning
process. In this way, navigation was more appropriate for fact-finding tasks, and the variable
behavior of exploration was more relevant to sense-making tasks.
Finally, it is possible that keyword selection requires expanding the role of inferencing
and prediction in similar, but even more complex ways than do hyperlinks. Just as selecting a
hyperlink involves inferencing about its relevance or utility (Afflerbach & Cho, 2008) and
predicting what information it will provide (Eagleton & Dobler, 2007), readers may apply the

25

same processes to keyword search phrase development and selection. However, since the reader
develops the keyword search phrases, the additional effort required may either represent
additional extraneous cognitive load (e.g., Sweller, van Merrienboer & Paas, 1998) or be
germane cognitive load. In the latter case, keyword development is an integral part of the
meaning making process, as it helps the reader create implicit interconnections among the
multiple resources encountered both before and after selecting that specific phrase (DeSchryver
& Spiro, 2008).
Cognitive Flexibility Theory
When considering the parameters of Web-mediated synthesis for this study, both
Cognitive Flexibility Theory (CFT) (Spiro, Coulson, Feltovich & Anderson, 2004) and Cognitive
Flexibility Theory on the Web (CFT-W) (Spiro & DeSchryver, 2010) provided useful insights.
CFT describes how to promote advanced knowledge acquisition of ill-structured topics, where
learners “attain a deeper understanding of content material, reason with it, and apply it flexibility
in diverse contexts” (Spiro, et al, 2004, p. 641). The foundation of this theory was based on
seven features: (1) avoidance of oversimplification and overregularization; (2) the importance of
multiple representations; (3) the necessity of cases; (4) the use of knowledge in practice rather
than the abstract; (5) the employment of in-the-moment schema assembly; (6) the importance of
multiple interconnectedness; and, (7) active participation and adjunct support for the
management of complexity. CFT is a constructivist, or generative, theory in which each
application of knowledge in use is different. It devalues storage or fixed knowledge and instead
promotes the “mobilization of potential knowledge” (p. 649). This is accomplished through the
flexible use of preexisting knowledge and through the acquisition of new knowledge in ways
amenable to flexible application (Spiro, Feltovich, Jacobson, & Coulson, 1992). This flexible

26

use of prior knowledge is not based on the recall of intact schematic representations, but the reassembly of multiple pieces of prior knowledge into a new “schema-of-the-moment.”
In its early applications, CFT utilized the affordances of closed hypertext systems to
promote these experiences. Using thematic coding systems, these hypertexts allowed for
virtually limitless text configurations, non-linear multiple criss-crossings of the knowledge
landscape, and conceptual combination and re-combination in a minimally guided exploration
environment. In these systems, multiple juxtapositions of content were provided in rearranged
instructional sequences, through revisiting the same materials at different times, in various
contexts, with different purposes, and from different conceptual perspectives (Spiro et al, 1992).
The time that it took to move from one resource to another was important in this environment.
That is, the speed with which conceptual variability could be demonstrated across cases was
essential to avoiding the long periods of time typical to more traditional linear learning
environments. Use of these systems also inculcated an epistemological belief structure that was
appropriate for learning about ill-structured ideas - a mindset that appreciated complexity and
irregularity.
CFT-W is a revisitation and extension of CFT for the open hypertext-learning
environment afforded by the Web (Spiro & DeSchryver, 2010). Several enabling affordances of
the Web were identified through CFT-W, which support its fundamental cognitive and metacognitive processes. These enabling affordances include: a virtually unlimited scope of
resources, near instant speed of access, extended opportunities for serendipity, availability of
multiple in-line adjunct cognitive aids, non-linear random access, and the well directedness of
search. Together, these affordances support several outcomes, including learner-initiated,
complex, reciprocally-adaptive searching techniques; open mindsets for learning; new forms of

27

background knowledge for in-the-moment schema construction; virtual conceptual simultaneity;
noticing unexpected connections; and transdisciplinary, interconnected boundary crossing. CFTW is presented in much greater detail with illustrative examples in DeSchryver & Spiro (in
preparation). Below, I highlight the primary extensions from CFT that guided and informed this
study.
One of the key differences between CFT and CFT-W was the emphasis on searching.
CFT-W also provided insights into the genesis of keyword search phrases. Spiro (2006e) wrote
of this process:
As you are going through this ongoing process (dynamic and unceasing), what it
is that you are “studying” changes, evolves. Your searches are changing partly
because ‘what’ you are searching for changes. The Web teaches you what the
‘subject’ is that you are trying to learn about. So you search differently, taking
off from where you end up at a given time to next places that are at conceptuallyoblique angles from the point of departure. Learners develop an organic
relationship with the material they are learning from: in rapid turnover time, what
they are trying to learn is affected by what they are finding, and that in turn
affects what they are looking for (and how they look for it), until an equilibrium is
reached – an accommodation between you (including what you already know and
what you are coming to know), your task, and the world, as dynamically mediated
by the Web (Spiro, 2006e, p. 5)
In doing so, the reader creates his or her own non-linear path through the Web. However, the
links established between and among information resources are “externally oriented” as opposed
to the internal guidance provided by the author-determined precompiled links.

28

The second feature of CFT-W that extended beyond the theoretical claims of CFT was
the enabling cognitive affordance of virtual conceptual simultaneity. This process involves
many ideas simultaneously being considered in the context of each other, across multiple
resources, from which “conceptual wholes greater than the sum of the parts can form” (Spiro,
2006e, p. 2). Based on the enabling affordances of speed, scope, and non-linear random access
on the Web and the associative cognitive process of spreading activation, ideas encountered by
the learner from multiple sources appear “simultaneously” in functioning cognitive space.
Experiencing the often-heterogeneous resources in this way may lead to the recognition of loose
interconnections among them (i.e., connections that might otherwise not appear to the reader),
including those not implied in the texts. In this way, virtual conceptual simultaneity may lead to
generative thinking of the kind targeted in this study.
CFT-W also provided new insights into the role of prior knowledge on the Web
(DeSchryver & Spiro, in preparation). As noted above, this role is more nuanced and less clear
in hypertext in Web-based environments than with traditional offline text. First, it outlines how,
when lacking adequate prior knowledge, a Web-based reader can use the virtually unlimited
scope of resources available, well-directedness of search, speed of access, and non-linear access
to fill in prior knowledge gaps that assist in assimilating and understanding the current text(s).
Filling in knowledge gaps in this way may also minimize the misuse of prior knowledge, such as
by helping to avoid the possible inaccurate recall of memories (Spiro, 1980). Second, given the
ability to fill in these gaps in previous knowledge, low previous knowledge readers may also be
able to make more inferences than they would otherwise offline (i.e., because inferences are
often based on prior knowledge). Third, any inferences made on the Web can be instantly tested
for accuracy and applicability to the task. Given the importance of inferences to understanding,

29

this ability to maximize the use of correct inferences and minimize the use of incorrect
inferences may lead to greater understanding. CFT-W also demonstrates how prior knowledge,
either traditional formats or that accumulated by filling in gaps in the moment, is used to
construct keyword search phrases and decide which links to follow in search results.
Another key difference between CFT and CFT-W was how revisitation is achieved. In
the closed hypertext CFT system, revisitation most often meant a reviewing of the same text.
However, while revisitation of the same text on the Web is accommodated, so is the revisitation
of the same idea, but from a different resource. In the latter case, an added dimension of
multiplicity allows for even greater appreciation for the irregularity and complexity of illstructured topics. As with CFT systems, however, this will likely only be accomplished once the
appropriate scaffolds for learning on the Web in this way are established, since this added
multiplicity may also lead to disorientation or cursory reading (Kuiper, Volman, & Terwel, 2005;
Hannafin, Hannafin & Gabbitas, 2009).
Finally, CFT-W indicated how the combination of an open mindset and the ability to
notice unexpected connections among the resources visited on the Web can lead to productive,
serendipitous stumbling. When reading about an ill-structured topic, the multiplicity of
resources available helps readers to understand that “it’s not that simple” by demonstrating the
irregularity of application. This can reduce temptations to over-generalize about ill-structured
ideas and lead the reader to be more open to recognizing the value of accidental, unexpected, and
otherwise ignored information. In generative reading-to-learn and reading-to-do tasks such as
those utilized in this study, these sorts of experiences may also have the important outcome of
facilitating synthesis of information.
Creativity

30

Of the higher order thought processes commonly associated with generative thinking,
creativity is arguably the most popular of late. In popular media, the likes of Friedman (2005),
Pink (2006), and Florida (2002) have all highlighted the importance of creative thinking in our
schools and for long-term economic success. The International Society of Technology in
Education has highlighted creativity in its first standard for both students and teachers (ISTE
2008 & 2009). As a result, promoting classroom creativity is often supported at the national
level, in both developed and developing countries (e.g., Gunseli, 2006; Loveless, Burton, &
Turvey, 2005; Park, Lee, Oliver, & Cramond, 2006; QCA, 2004; Williams, 2002).
Though creativity may be defined in a variety of ways, for the purposes of this study, it
will be explained by a composite of definitions from: (1) Plucker, Beghetto, and Dow (2004)
who defined creativity based on a survey of 90 creativity research articles; (2) O’Quin and
Besemer (2006) who designed the Creative Product Semantic Scale (CPSS) and, (3) Mishra &
Koehler (2008) who adapted the CPSS subset for elaboration and synthesis. For the purposes of
this study, creativity was defined as:
the interaction among aptitude, process, and environment by which an individual or
group produces a perceptible product or idea that is novel, useful, and whole within a
social context
The primary reason to examine creativity in the context of this study was to explore and
emphasize the processes that resulted in new or novel ideas when reading on the Web, which
aligns with the goals of generative thinking. However it is important to highlight the importance
of utility and value; that is, in order to avoid the misconception that creativity is unbounded,
creative ideas must be useful. Equally important is the proposal that creative ideas must be
“whole,” a term used by Mishra & Koehler to encapsulate the various subscales of the CPSS that

31

are intended to measure how organic, elegant, complex, understandable, and well-crafted
something is. Finally, this definition makes clear that creativity is not typically a rapid in-thehead insight, but a complex process mediated not only by cognitive ability but process and
environment (which in this study, refers to the Web).
Literature on the process of creativity provides insights into the generative thinking of
interest in this study. Wallas (1926) first proposed that creativity is based on preparation,
incubation, illumination, and verification. Osborn (1953) expanded these ideas to include
orientation, ideation, and synthesis. The latter two are of particular interest in this study, defined
as “piling up alternatives by way of ideas” and “putting the pieces together,” respectively. The
creative cognition explanation of creativity provides even more detail about how the generation,
exploration, and integration of ideas occurs (Ward, Smith, & Finke, 1999). Creative cognition
describes idea generation as a retrieval of existing ideas from memory, followed by associating
or combining those ideas to form new, synthesized forms. Individuals may also use analogical
transfer of ideas from one domain to another, and categorical reduction of ideas. During the
exploratory phase, individuals may search for novel or desirable attributes, metaphorical
implications, or functionality in an idea. They will evaluate the idea from different perspectives
or within different contexts, interpret the idea as a possible solution to a problem, and identify
practical or conceptual limitations to the idea.
The attention given to exploration and association in all theories of creativity makes clear
the importance of both. Einstein referred to this as combinatorial play that helps to connect
ideas and concepts that have rarely, if ever, been combined. Amabile, Hadley and Kramer
(2002) described this as “throwing a bunch of balls into cognitive space, [and] juggling them
around until they collide in interesting ways” (p. 52). Creativity of this type often requires that

32

this play occurs among disparate ideas for a long enough period of time to generate new idea
spaces (McWilliam, 2009) or move ideas from one state to another (Jackson, 2006). As such,
these descriptions comprise a perspective on creativity that values what will henceforth be
referred to as “idea-play.”
(Re)Defining Synthesis
As is apparent above, the concept of synthesis is introduced widely across fields of
scholarship, from reading research to the creative process. However, the meaning of this term
varies equally as widely. This complexity can be demonstrated by exploring one of the most
common manifestations of synthesis for learning, Bloom’s Taxonomy of Educational Objectives
for the cognitive domain (1956). Though conceptualized as a creative activity leading to new and
original wholes “with emphasis on creating new meaning or structure,” words such as combine,
compile, organize, rearrange, and summarize have all too often displaced the creative aspect of
synthesis. It is clear from Bloom’s original intent that combining, rearranging, and summarizing
were parts of the process that could help lead to synthesis of new and original wholes, but this
nuance is often difficult to parse. A reclassification of this Taxonomy (Bloom, 2001) helped to
alleviate this issue, when Bloom re-named synthesis as creativity. Another common use of the
term synthesis stems from the Big6 approach to ICT Literacy, which first describes synthesis
using lower-order terms (e.g., organize and communicate), and then uses some examples that
connote higher order thinking (e.g., create and generate), while using other examples that are
associated with lower-order words (e.g., classify and group) (Eisenberg, Johnson & Berkowitz,
2010).
This level of complexity is also noted in the reading literature. As described above, the
primary goal associated with synthesis in reading comprehension is to understand the meaning of

33

one or multiple texts. Cho (2011) described how synthesis strategies are required to achieve
cross-textual meaning. Dole et al. (1991) use the word “synthesis” to describe summarizing,
where readers synthesize ideas to create a new coherent text that stands for the original. Other
theories of reading discuss “moving beyond the text,” but stop short of describing how this leads
to new ideas or knowledge. Instead, “beyond the text” in these cases connotes inferencing,
which is tied to the author’s implied meaning (Dole, et al; Kintsch, 1994; Scharere, Pinnell,
Lyons & Fountas, 2005), to the reader’s interpretation of content, to the author’s intent in
relation to the reading context (VanSledright & Kelly, 1998; Wineburg, 1991a, 1991b; Wolfe &
Goldman, 2005), or to using ideas from the text in new settings (Clymer, 1968). For online
reading comprehension, synthesis has been defined by the ability to sort relevant information
from irrelevant information, organize and understand the texts available (“Taxonomy,” n.d.) and
assimilate multiple sources across multiple modes to construct meaning (Glister, 2000). In this
way, the Taxonomy (“Taxonomy,” n.d.), defines synthesis as “integrating information from
multiple resources” across multiple texts and media types. Furthermore, the theory of
Constructed Response Reading on the Internet accounts for this through a process of “combining
disparate forms of information to construct meaning.” None of these interpretations of synthesis
connotes a generative process that moves the reader substantially beyond what is explicit or
implicit in the text.
Cho (2011) extends the concept of meaning making from text in a quasi-generative way.
Based on the Programme for International Student Assessment (PISA) 2009 Framework
(Organisation for Economic Co-operation and Development, 2009; International Association for
the Evaluation of Educational Achievement, 2009), he notes that the meaning making online “is
most often a midpoint in the act of reading, where the meaning construction is followed by the

34

use of the meaning that is constructed from a variety of print and digital texts” (p. 39). He does
so while emphasizing the situated nature of reading and comprehension. That is, “people read to
use and act with the meaning they construct through the reading of the wide range of texts in
different situations” (p. 39, emphasis added). However, acting with meaning in this way, may or
may not lead to generative knowledge constructions.
One way to interpret the complexity of synthesis is to think of it in different levels. Lowlevel synthesis may involve combining, organizing, and summarizing that leads the reader to
better understand and make meaning of text. Whereas, high-level synthesis results in creativity
and knowledge generation on the part of the reader. Consistent with this idea, Eagleton and
Dobler (2007) describe how strong readers synthesize ideas to understand what is being read by
using inferencing and summarizing as they bridge ideas together from separate locations on the
Web. This is considered low-level synthesis. They further assert that:
At the highest level, synthesizing goes beyond determining meaning and moves
into the development of a new perspective or viewpoint, also known as
transformation. A reader’s thinking changes based on her transaction with texts,
as she comes to “own” her new ideas. (p. 38, emphasis added)
They go on to point out the complexity of the various elements of synthesis, for example, that “a
synthesis takes a summary one step further by including the reader’s personal response to the
text in the form of connections to herself, other texts, and the world” (p. 201). Jenkins similarly
described this concept of high-level synthesis as a process by which “multiple resources are
combined to produce new knowledge” (2006, p. 94, emphasis added). It is this sort of high-level
new knowledge synthesis that creativity describes, as well. Osborn’s (1956) notion of synthesis
included “putting the pieces together.” However, as opposed the goal of meaning making in

35

reading comprehension theories, Osborn conceived of putting pieces together as part of a process
to create novel and useful products or ideas. Ward et al. (1999) similarly described synthesis as
the process that results in new ideas.
Given the importance of higher-order thinking and deeper learning on the Web (e.g.,
Kuiper, Volman & Terwel, 2005) the field is in need of greater attention to how learners
integrate and understand resources being visited across multiple texts (Hartman, 1995, 2000;
Leu, Zawilinski, et al., 2007) (i.e., how they engage in low-level synthesis), and then, how they
solve problems with the Web by moving beyond information consumption to knowledge
generation (i.e., how they engage in high-level synthesis). Coiro (2009) summed this up by
stating that we need more research that explores “how to synthesize information (e.g. isolated
data) with unique personal experiences and perspectives to generate new knowledge that can be
acted upon in ways that help others” (p. 2, emphasis added). The focus on new knowledge that
moves beyond meaning construction and understanding is why the current study proposes that
high-level synthesis should be considered a “post-comprehension” perspective of reading online.
Based on these ideas, prior to meeting with participants, I operationalized the term
synthesis in ways that were specific to the online learning environment. Since low-level
synthesis resulted in meaning making, I defined this as “synthesis for meaning.” High-level
synthesis resulted in new knowledge construction, so I referred to this as “generative synthesis.”
Synthesis for meaning. Synthesis for meaning occurs when readers on the Web
organize, combine, compose, rearrange, rewrite, compile, and structure information in a way that
facilitates their understanding of the multiple text(s) encountered (Duke & Pearson, 2002; Dole
et al, 1991; Glister, 2000; Hartman, 1995; Leu, Zawilinski, et al., 2007; Mayer, 2000; Rouet,
2006; Stromso & Braten, 2002). Key components involve sorting relevant from irrelevant

36

information and organizing it effectively (“Taxonomy,” n.d.), summarizing information (Dole et
al, 1991; Hidi & Anderson, 1986), making inferences about the implied author meaning or intent
(Dole, et al; Kintsch, 1994; Scharere, Pinnell, Lyons, and Fountas, 2005), and relating the text to
background knowledge (Dole, et al., 1991). Skills that support synthesis for meaning include
note taking, bookmarking, clipping, highlighting, and other forms of offline and online text
management. The texts to be synthesized are available in multiple media formats (e.g., written
text, audio, images, video, table, graphs), thus online readers must be proficient at summarizing
information from multiple sources. They engage in a low-level synthesis by “…actively
construct[ing] the text that they read through the choices they make about which sites to visit”
and “putting together an understanding of what they have read” (Leu, Zawilinski, et al., 2007, p.
48). The result of synthesizing in this way is to make meaning of the text.
Generative synthesis. Two forms of generative synthesis were defined for the purposes
of this study, Virtual Conceptual Simultaneity Synthesis and Creativity Synthesis.
Virtual conceptual simultaneity (VCS) synthesis. The first form of generative synthesis is
called VCS synthesis, based on Spiro’s enabling cognitive affordance of virtual conceptual
simultaneity (VCS). It is a form of creative synthesis that happens “in-the-moment.” While
learning on the Web, multiple, heterogeneous resources are simultaneously “being considered in
the context of each other” from which “conceptual wholes greater than the sum of the parts can
form” (Spiro, 2006e, p. 2). This latter form of synthesis is more prevalent on the Web than with
traditional offline resources given the possibility of the virtual conceptual simultaneity of
resources. Because the online reader can rapidly click from one resource or idea to another, they
may experience the associated cognitive process of spreading activation. Ideas accessed from
multiple sources may be subject to “simultaneous” processing within the same functioning

37

cognitive space, and this can lead to conceptual breakthroughs. VCS synthesis is also likely to be
task agnostic, whereby the conceptual whole that emerges will be related to the resources
recently visited more so than the primary task or purpose for reading on the Web.
Creative synthesis. The second form of generative synthesis is called creative synthesis.
In combination with the methods of synthesis described above and several other critical
processes theorized to undergird creativity, such as incubation and iteration (Bruner, 1962;
Csikszentmihalyi, 1997; Osborn, 1953; Sternberg & Lubart, 1999; Ward, Smith, & Finke, 1999),
ideas may occur that serve as novel, effective, and holistic solutions (Mishra & DeSchryver,
2009) to a task. The creative cognition approach (Ward, Smith, & Finke, 1999) is particularly
well suited to exploring and identifying Web-mediated reading and thought processes given how
well most of these processes can be performed using keyword searches and other affordances of
the Web. For instance, retrieval of ideas and associating or combining those ideas is supported
by keyword search phrase use, especially when using compound keyword phrases. The Web
also easily accommodates transformation of ideas to new forms, where readers can search for
phrases in the main Google page and then switch to a Google Image search. And there are
myriad opportunities for serendipitous encounters with information on the Web (DeSchryver &
Spiro, 2008). Given the longer periods of time often associated with creative solutions, the
results of this form of synthesis are typically task specific; the solutions that emerge will be more
likely to be related to the primary task or purpose for reading on the Web than the resources just
recently visited.
It is important to note two things about these three definitions of synthesis. First, they are
not meant to imply that a synthesis activity is necessarily defined by any one of these processes.
For this study, synthesis is considered a complex phenomenon, more appropriately represented

38

by a multiplicity of activities. Any single instance of synthesis could be represented by a
combination of keyword construction, prior knowledge, summary, inference, and note taking. At
the same time, creative synthesis might manifest as a combination of VCS and synthesis for
meaning. Second, given the amorphous nature of synthesis, this study was also designed to
allow for emergent forms of synthesis to occur that were not described by the definitions above.
In this way, the study might provide some insight into this complex process.
Research that focuses on synthesis for learning on the Web has, to date, been primarily
framed through the synthesis for meaning lens noted above, and has proven difficult to observe
(e.g., Leu, Zawilinski, et al., 2007). This study expanded beyond that framework, and sought to
provide evidence of all three types of synthesis in advanced learners. In doing so, it addressed
the challenges of studying synthesis in four ways. First, by working with advanced learners, it
was expected that synthesis of all types would occur more often than with K-12 learners and that
advanced learners would be better able to articulate their synthesis experiences through thinkalouds and interviews. Second, the tasks provided to participants in this study were specifically
designed to promote synthesis, especially during the second, reading-to-do task. Third, the
coding system for data analysis allowed for multiple codes to be assigned to any given reading
activity, which provided a level of appreciation for the complexity of online reading. Finally,
this study used an expanded the definition of synthesis which allowed new forms of synthesis to
emerge. This provided both a wider and more specific net with which to categorize related
online activity.
Chapter Summary
This chapter reviewed the essential components of four different theoretical frameworks
that informed a reconceptualization of synthesis when reading online. The reading

39

comprehension literature contributed important ideas about prior knowledge, informational text,
inferencing, summarizing, and multiple text comprehension. The hypertext and Web-based
reading literature highlighted how readers are required to take more active roles in complex, nonlinear, multi-modal environments and the specific skills and strategies required for successful
comprehension therein. The CFT and CFT-W literature offered further support for the
importance of searching, virtual conceptual simultaneity, and prior knowledge on the Web.
Finally, the creativity literature contributed specific processes that undergird generative thinking.
Collectively, these four areas of scholarship helped re-define synthesis. First, the reading
comprehension, hypertext, and Web-based reading literatures offered traditional conceptions of
synthesis. In these traditions, synthesis leads to understanding both explicit and implied ideas in
the texts being read. I categorized all synthesis activity related to this as synthesis for meaning.
The CFT, CFT-W, and creativity literatures helped extend this notion of synthesis with two
forms of generative synthesis. CFT-W offered the idea that virtual conceptual simultaneity on
the Web can lead to “new” ideas on the part of the reader, and creativity scholarship helped
outline the idea the creative synthesis is an integrated reading activity leading to novel, useful,
and whole solutions.
Based on this re-conceptualization of synthesis, I conducted a qualitative study of the
synthesis skills and strategies used by advanced learners on the Web to answer the following
research questions:
1. How do synthesis for meaning, VCS synthesis, and creative synthesis manifest for
advanced learners in Web-mediated reading-to-learn and reading-to-do
environments?
2. What other forms of generative synthesis emerge from the data, and how do they

40

manifest, for advanced learners in Web-mediated reading-to-learn and reading-todo environments?
3. How do synthesis for meaning activities and generative synthesis activities
interact for advanced learners in Web-mediated reading-to-learn and reading-todo environments?
4. What role does background knowledge play to support the knowledge synthesis
of advanced learners in Web-mediated reading-to-learn and reading-to-do
environments?
5. What role does note taking play to support the knowledge synthesis of advanced
learners in Web-mediated reading-to-learn and reading-to-do environments?

41

CHAPTER 2
METHOD
Chapter Introduction
This chapter presents the qualitative methodologies used in this study. It details the
design, participant selection procedures, task development, data collection, research session
details, and data analysis procedures used to investigate the five main research questions.
Design
This study utilized a multiple-case study design and multiple methods of data collection
and analysis to triangulate findings.
Participant Selection Procedures
This study utilized purposive sampling (Patton, 1990) to identify eight highly literate
advanced learners (i.e. doctoral, graduate business, and law students), who demonstrated using
the Web in sophisticated ways for learning, indicated a comfort with thinking out loud, and
represented a diversity of backgrounds and experiences. Several factors guided the choice to
explore advanced learners. First, studying experts has historically provided important
contributions to the learning sciences (e.g., Berliner, 1986; Chi, Feltovich & Glaser, 2005;
Ertmer & Newby, 1996; Pressley & Afflerbach, 1995; Wyatt, Pressley, El-Dinary, 1993).
Second, the advanced learners utilized in this study all successfully passed validated measures of
high offline reading in their entrance to a nationally recognized graduate school (GRE, GMAT or
LSAT). Third, the use of advanced learners increased the likelihood of synthesis interactions on
the Web as compared to the K-12 students most often used in studies of Web-based reading (e.g.,

42

Coiro & Dobler, 2007; Leu et at, 2004), where synthesis has been very difficulty to identify.
Finally, recent work by DeSchryver & Spiro (2008; in preparation) provided illustrations of
several skills that may be unique to advanced learners on the Web. For instance, they indicated
that advanced learners may use sophisticated methods for locating information on the Web that
involve an organic and reciprocal relationship between the learner and the knowledge landscape
of the Web. They also proposed that advanced learners rely on an integrated method of
“triangulation” for resource evaluation that becomes part of the text comprehension and learning
process (DeSchryver & Spiro, 2008). Finally, they suggested that advanced learners often
benefit from utilizing Web learning skills and cognitive processes in concert, and not in isolation.
Ensuring a diversity of experiences and backgrounds in the final sample was important to
maximize exposure to different ways of thinking. That is, as Zerubavel (1995) noted, university
learning can create mental “walls” and “moats” that lead to mental rigidity within disciplines. In
addition, Coiro & Dobler (2007) noted how different types of readers might demonstrate
different strategic knowledge and ability to describe this knowledge. A diversity of participants
thus maximized the potential for this study to document a variety of different synthesis skills and
strategies in advanced learners.
The selection process started with an online survey, administered to identify advanced
learners who demonstrated the potential to use the Web in sophisticated ways for learning. The
survey was largely researcher generated based on the researchers knowledge of these constructs,
a strategy described by Netemeyer, Bearden & Dharma (2003). It also used questions adapted
from the Survey of Internet Usage and Online Reading (“Survey”, 2003). Three experts in the
field of Web-based reading and learning established content validity by judging how well the
survey reflected the intended constructs (Netemeyer, Bearden & Dharma, 2003). Based on their

43

feedback, I modified the survey, and then re-administered it to ten education graduate students
and practitioners to elicit feedback on its clarity, presentation, form and length (Litwin, 1995).
The survey contained questions about general Web use, Web use for learning, knowledge of the
Web, use of Web-based tools for learning, and advanced Web use for learning. Within these
categories, the participant’s ability to use keywords, comprehend multiple Web texts, make
inferences about those texts, and make unique insights about those texts was measured. The
survey also asked each respondent to rate the comfort level with thinking out loud in a face-toface research session. As used it was 45 minutes long. More than 500 doctoral, graduate
business and law students received an invitation to the final survey via email. Eighty-one
completed the survey. From these eighty-one surveys, I identified the top eight responses based
on the criteria above. Three experts in the field of Web-based reading and learning verified these
eight respondents as advanced learners who demonstrated the potential to use the Web in
sophisticated ways for learning. All eight respondents agreed to participate in the study. Each
collected a stipend for their time.
The eight participants who participated in this study were Beth, Janet, Jim, Jeff, Karen,
Molly, Olivia, and Troy (all names are pseudonyms). Five females and three males participated.
Their backgrounds were as follows: one law student and seven doctoral students from the fields
of education, philosophy, criminal justice, computer science, food science, economics, and
zoology. The average age of the participants was 32 years old. They reported averaging more
that 4 hours of Web use on a typical day, including an average of between 3 and 4 hours of Web
use for learning. The average self-assessment score on a scale from 1-5 (beginner-expert) for a
series of five questions related to finding, organizing, and evaluating information on Web sites

44

was 3.9. The average self-assessment score for a series of three questions related to synthesizing
information and using the Web for creative solutions and conceptual breakthroughs was 3.4.
Tasks
The specific task prompts chosen for this study reflected the following characteristics: (1)
highly ill-structured; (2) of general interest to the participants, but not topics so common that
participants would likely know a lot about them; (3) not too broad, not too narrow; (4) two clear
sides to explore and extensive resources available for both; and, (5) “Wikipedia unfriendly” results based on the initial expected keywords did not include Wikipedia on the first page. After
two separate pilot tests with five doctoral students in education, feedback from them about the
clarity or questions, and analysis of the results provided, the following tasks emerged:
Reading-to-Learn Task: Please use the next hour to learn about why some people
make decisions to change their behavior in ways that may lessen the effects of
climate change and why some people do not choose to change their behavior in
ways that may lessen the effects of climate change. Please use the learning time
you have to prepare as if you are going to be part of a public panel on this topic
that will represent a diversity of perspectives.
Reading-to-Do Task: After such a good showing in the public panel about
individual choices related to climate change, you have been hired by an
environmental non-profit.
Your first day on the job, the Director of the non-profit asks you to come up with
some new, creative ideas about how persuade three specific groups of people to
voluntarily change their behavior in ways that will help lessen the effects of
climate change: (1) college students; (2) high SES professionals; and (3)

45

individuals with strong religious affiliations. The Director would like to hear any
thoughts you have about targeting these three groups at the afternoon staff
meeting 45 minutes from now. Though the Director was clear that your ideas
should be new and creative, he also indicated that they should be back up with
some evidence of their potential to be successful strategies.
The first task replicated the sort of open-ended reading and learning on the Web that is common
in schools, at work, and at home. The second task was intended to elicit a creative synthesis of
the type outlined above; that is, it requested a perceptible idea that was both novel and useful
within a situated social context.
Two arguments guided the choice to focus this study on ill-structured topics. First, there
is an increasing need for learners of all ages, both in schools and out, to better understand
complex, ill-structured problems. This includes myriad fields of knowledge such as accounting
(Friedman, 2004); multi-party regional water issues (Adams, 2001); genetic engineering
(Hollander, 2004); cyberterrorism and changing weather patterns (Mitroff, Alpasian & Green,
2004); and, teaching with technology (Mishra & Koehler, 2007). Second, the learning
affordances on the Web appear to match very well with the unique needs of learning about illstructured topics (DeSchryver and Spiro, 2008; in preparation). That is, among other things,
learning ill-structured topics requires multiple open representations; real-world cases with
naturally occurring and ecologically based integration and interconnectedness; an open-mindset
(which learning on the Web may actually facilitate); and the ability to criss-cross the knowledge
landscape from multiple directions (DeSchryver & Spiro, 2008; Spiro & DeSchryver, 2008). As
a precursor to Web-based learning, closed hypertext learning systems (e.g., Cognitive Flexibility
Systems) demonstrated these characteristics (Mishra, Spiro & Feltovich, 1996) and provided

46

learning benefits in ill-structured domains (Jacobson & Spiro, 1995). The Web provides similar
learning affordances.
Data Collection
Each participant met with me for approximately 2.5 hours. During the first five minutes,
I recorded each participant explaining his or her background and experiences with the topic of
climate change. Then, they practiced thinking aloud for 10 minutes while using the Web to learn
about Thomas Malthus’ main ideas about population growth and using any techniques or
strategies for saving and organizing information that they might need to prepare for a test,
presentation, paper, or important discussion about this subject. Each participant had a pad of
legal paper, pens, and access to any digital resources (software or Web tools) they requested to
do the tasks. Their computers had the Firefox browser with an empty Google search page open,
one page on each of two monitors. Thereafter, I gave each participant a printout of the first
reading-to-learn task spelled out above and the following instructional text:
Explore what you can discover that you don’t already know. Be open to new
ideas. Try not to get stuck in ruts; that is, don’t allow your learning and searching
to close down on any one Web site or particular idea. This is a complex topic that
necessitates sophisticated and open learning strategies to even begin to fully
appreciate it. Also, continue to use any techniques or strategies for saving and
organizing information that you think you might need to prepare for this panel.
I reminded them to think aloud constantly and told them that they would be reminded to do so if
they remained silent for any period of time.
Once the participants began this task, verbal protocols, screen video, and digital trails
gathered data. The software system Morae collected video of their screen and the audio of their

47

think aloud comments, as well as metrics on the Web browsing conducted. I watched the video
of each session from a remote computer in real time, and coded the video stream based on a
preliminary coding scheme derived from the theoretical framework above.
After each reading-learn task, I interviewed and audio taped each participant for
approximately 15 minutes about the task. The protocol for this interview reflected informal
conversational interview and interview guide approaches to interview research (Patton, 2002).
That is, I designed the protocol based on the theoretical framework above to guide the process
(i.e. as an interview guide), including questions about locating information, evaluating
information, synthesizing information, and note taking. However, I conducted the interviews as
informal conversations, since such an environment offered the “maximum flexibility to pursue
information in whatever direction appears to be appropriate, depending on what emerges from
observing a particular setting…” (p. 342).
Upon completion of the first interview, I gave each participant a printout of the readingto-do task. I provided similar directions about goals, mindset, techniques and strategies, as noted
above. The data collection procedures reflected those carried out in the first task, and one final
retrospective interview finished the session in a similar fashion to the first.
The Research Sessions
I carried out the research sessions in campus meeting rooms one participant at a time.
Participants varied in their ways of reacting to the environment provided and completing the
tasks outlined. All of them demonstrated an acceptable comfort level with thinking out loud in a
short amount of time (i.e., I asked all of them to stop “practicing” earlier than the 10 minutes
allocated for this in the design plan). Everyone revisited the task instruction right away, rereading it to him or herself, annotating it (e.g., underlining key words or writing short notes on

48

the task paper) or highlighting important ideas in it (e.g., emphasizing out loud that the task was
specifically interested in the “why” people do or don’t change behavior). The choices about
what to do first varied across participants and tasks. In the reading-to-learn tasks, many started
immediately with open ended Google searches - [green living tips], [opinions on climate
change], [climate change] (twice), and [why aren’t people more concerned about climate
change]. Both participants who used the generic (and somewhat unexpected) [climate change]
search phrase navigated to Wikipedia right away to refresh their memories about climate change.
Others took some time to note their own thoughts about the question. One took a short time to
make a bullet list of reasons in her notes – fear, guilt, compassion, change is difficult, don’t care
about environment or other people, not hearing messages, don’t know what to do, overwhelmed,
burying head in sand – followed soon thereafter by a search of [behavior change related to
climate change]. Two others took extend time to think out loud about what they already knew
and take some notes. For example, Janet started right away by saying the following:
“I would just write down a couple assumptions on what I think they might
be doing. You know…so climate change is real…so then some people I
would think one of their questions is going to be yes, but what can I do at
an individual level that really makes an impact? And, is there any proof
that if I do alter behavior it does make an impact? Because I think that’
ultimately, from what I already know about this, it seems a lot of people
just figure that what they do doesn’t really matter…whether they use
paper bags or plastic bags, they don’t seen the impact on climate...and
getting back to the original idea that they don’t believe its really occurring,
or is it really occurring at a rate that is relevant to them. And that’s

49

probably the biggest problem. Reminds me of when my parents, for years
I was talking to them about the importance of well heads, and they totally
ignored it until the people across the street starting having well problems
because a new development, and suddenly, it became pertinent to them,
and they suddenly became all conservative about – gee we have to protect
out well heads – and I’m thinking yeah we saw that before, but you
weren’t interested until it actually directly impacted you. And,
unfortunately, it seems like that’s human behavior in a nutshell. So, what
we’re going to do now it to find out what specifically has been looked at
in terms of why people change behavior or don’t change behavior so that I
can be educated here for this public panel.”
After planning to find out what the average person finds on the Web (misinformation and
otherwise), and then looking for the actual facts about climate change, she then searched for
[attitudes toward climate change] five minutes after the task started.
During the reading-to-do tasks several participants again started right away with quick
searches - [how to lessen effect of climate change] [how to change behavior climate change],
[college students and climate change], and [how to convince Christian right wing people to be
concerned about climate change]. The others spent anywhere from two minutes to seven minutes
listing out how they thought the three groups identified might be convinced to change their
behavior related to climate change, then commenced with searching. Several made reference to
articles or ideas from the first session, but only two specifically visited any sites initially from
that first session.

50

From there, the skills and strategies for completing the tasks varied. Most used the
primary search option (Web) on Google. A few used Google scholar. One searched extensively
on Amazon for books related to this topic. Another used the online library catalog at her
university quite a bit. The keyword search phrases used were almost entirely different from
participant to participant. Several participants used PDF files available from the Web. A few
used bookmarks, but otherwise, no adjunct aids were used to organize information. Most used
multiple tabs in the browser, and three did this extensively, as shown in Figure 1.

Figure 1. Use of multiple tabs in the browser. For interpretation of the references to color in this
and all other figures, the reader is referred to the electronic version of this dissertation. The text
in this figure is not meant to be readable. The figure is meant to demonstrate the use of multiple
tabs, regardless of their content.
51

Everyone took some notes; three used only the legal pad provided, one started on the pad
and extended her notes in Microsoft Word, one used Google Docs exclusively, two used
Microsoft Word with a few additional offline notes, and one took a very few notes on the legal
pad. Among those taking notes on the computer, bullet points, lists, and outlines were common;
offline notes also included lists, but they were more freeform in their structures (e.g., see Figure
2). Among the seven taking extensive notes, between two and five pages of notes were
completed. Each participant had access to two monitors during the tasks (a laptop and an
external monitor). Only two participants made use of the external monitor in any significant way,
mostly to take notes while Web browsing on the main laptop monitor.

Figure 2. Computer based versus written notes. The text in this figure is not meant to be
readable. The figure is simply intended to demonstrate linear vs. non-linear note-taking.

52

Only three times did I ask a participant to continue speaking out loud, and once I asked
someone to speak louder. The participants asked very few questions of me, primarily about
things like where to take notes, if they should bookmark or just say they would bookmark, and
other logistical issues. There were no technical issues with the equipment being used in the
research sessions, either by the participants for their tasks or with data collection, save for one
instance of a slow Internet connection which lasted about one minute.
Data Analysis
The think aloud data and screen video captures served as primary data sources for this
study. My analysis of this data used abductive (Morgan, 2007) coding that started deductively
and then inductively added new codes that emerged from the data. I first listened to the think
aloud audio and watched the concurrent screen video several times in order to gain a general
sense of the data (Tesch, 1990). Then, the coding was conducted.
The first coding pass on the data focused entirely on synthesis moments, using codes for
(1) synthesis for meaning, (2) VCS; and (3) creative synthesis. In approaching the data, the
notion of propositional clusters (van dijk & Kintsch, 1983), whereby the unit of analysis
comprises a “cluster of propositions focused on the same idea” (Coiro & Dobler, 2007, p. 227),
provided the initial parsing lens. I soon adapted it to identify the data as “synthesis clusters”
comprised of propositions that defined an instance of at least one element of synthesis, given the
complexity and multiplicity of the clustering found to frame the synthesis taking place.
During this first coding pass, several instances of synthesis type experiences emerged that
were not explained by the preliminary coding system. As such, I followed an iterative process of
modifying the existing codes and generating and applying new codes, as outlined by the constant
comparative process (Bogdan & Biklen, 2003; Dye, Schatz, Rosenberg, & Coleman, 2000;

53

Merriam, 1988). Once I identified a new set of codes, I utilized experts in the field of online
reading to evaluate them. I first asked these experts to provide a critique of the final coding
schemes used and developed over the course of this analysis. Then, I showed them multiple
samples of each synthesis element. As was the case with Hartman (1995), the primarily purpose
of this exercise was to confirm or disconfirm the “perceptions, hypotheses, and interpretations”
made during the construction of these elements. There was 100% agreement (with minor tweaks
in descriptive language) regarding the elements discussed with these experts. Based on their
feedback, I finalized codes for six elements of Web mediated synthesis - divergent keyword
searches, synthesis for meaning, in-the-moment insights, repurposing, reinforcement, and
creative synthesis. I then conducted one more pass through the data to re-code synthesis
experiences (i.e., based on the final code definitions, some forms of synthesis were changed to
others).
Thereafter, I performed one additional coding on two of the participant’s data (Beth and
Olivia) to inform the in-depth case studies provided below. This pass provided information
about the conditions for synthesis that appeared in the data. That is, consistent with the
allowance for multiple codes in this study, each synthesis proposition was revisited and codes
related to note-taking and background knowledge were applied to the data stream just prior to,
during, and just after the synthesis proposition. Two expert raters determined interrater reliability
of the application of these codes. These raters were shown 10% of the synthesis moments coded
in each category applied in Olivia and Beth’s data (the case study text in chapters four and five
includes most of the examples). The initial interrater agreement for these examples was 50%.
After discussion, we agreed on 90% of the examples. We also agreed that the final 10% on

54

which we disagreed were complex interactions that could accurately be coded several ways or
with multiple codes.
As it relates to reliability, it important to note that I concentrated the bulk of my efforts to
establishing reliability in two specific areas. First and most importantly, I wanted to have full
agreement based on the synthesis elements that were initially proposed (and subsequently
revised) and those that emerged during the study. Once reliable corroboration of these categories
emerged, they did not change and there is high confidence and agreement associated with them.
Secondly, given the thickness of the descriptions provided below in the case studies, I wanted a
high level of reliability of the application of theses coded based on the examples provided
therein. This was achieved as well. As this paper evolves, however, it will become clear that
even with high levels of reliability, there are many vagaries when applying codes with a concept
as indeterminate and irregular as synthesis. As such, the total numbers reported below for the
seven elements of synthesis discussed represent accurate and reliable interpretations based on the
multiple coding passes noted above. As such they can be used confidently to make broad claims
about the data. However, the impact of the important results reported below is not dependent on
a pinpoint accuracy of these numbers. The findings of most interest are based on the final
operationalization of these synthesis elements and the thick descriptions of them that emerged
from the case studies. That is, none of the important findings below depend on whether there
were 40 insights across the participants or 45.
Based on the results of the above analyses, I performed two additional analyses of the
think aloud and screen capture video, specific to the use of keyword search phrases. First, I
coded every keyword search phrase as either “consistent” or “divergent,” based on whether any
words in the search phrases used were included in the task prompts provided to the participants,

55

or not. Second, I used data from audio data of participants talking about their background
knowledge about climate change and their interviews to corroborate (triangulate) and extend
ideas that emerged from the above analysis.
Subjectivity in Qualitative Research
Peshkin (2005) wrote of researchers “It is no more useful to acknowledge simply that
subjectivity is an invariable component of their research than it is for them to assert that their
ideal is to achieve objectivity” (p. 17). Though Peshkin indicated that this comment applied not
only to observational investigations, but all research, it is clear that a qualitative study such as
that conducted here is more subject to investigator persuasions that may affect the results of the
observation and analysis.
As such, the following techniques were employed throughout this study to reduce
investigator subjectivity to the extent that it was possible. First and foremost, at several times
during the study, the use of experts was used to examine my perceptions, inferences, and
interpretations. This is a common practice when employing think-aloud data (e.g., Hartman,
1995). When designing a survey to help determine the most advanced users of the Web for this
study, experts in the field of online reading established content validity based on the intended
constructs. Later in the participant selection process, the same experts verified that the
individuals selected to be invited to participate in the study demonstrated the potential the use the
Web in sophisticated ways for reading and learning.
Similarly, during the constant comparative analysis of think-aloud data, experts were
used to critique both the initial coding categories and the new coding categories that emerged
from the data. Experts were shown both the specific operationalization of these codes and
several examples of the concepts assigned these codes during their evaluations. Given that the

56

final codes themselves would end up as the foundation for a Theory of Web-mediated
Knowledge Synthesis (detailed in Chapter 6), the negotiation about these ideas was lengthy.
Though the experts confirmed all of these codes with 100% agreement, they contributed more
than three hours of feedback and dialogue, providing insight into the evolution of these codes
and the language used to describe them. This proved invaluable. During this process, the
subjectivity of my assumptions and interpretations was directly challenged, and the resulting
constructs were inherently more objective that before the expert insights.
Experts were employed at one additional step in the investigation to further ameliorate
subjective analysis on my part. After the coding scheme had been modified and finalized based
on expert feedback, I reviewed and re-coded the case study data for Beth and Olivia, the two
participants from which a majority of the discussions below are derived. After this final coding
analysis, experts once again evaluated the assignment of codes. This time, they reviewed 10% of
the examples to which each code applied. Though initial inter-rater reliability on these examples
was only 50%, further dialogue and debate resulted in 90% agreement. For the final 10%
disagreement, the experts and I concluded that the disagreement was important evidence as to the
complexity of the very nature of Web-mediated synthesis. That is, we came to agreement that the
examples should be coded in multiple ways. This conclusion corroborated one of the
assumptions identified in Chapter 1 to frame the study from the outset: synthesis is often too
complex to reduce to singular interpretations.
In all, the use of experts to evaluate the coding schemes and their application was more
beneficial to the evaluation of ideas in this study than providing percentage agreement statistics
to report here. This is consistent with Barbour’s (2001) assertion that in qualitative research

57

the degree of concordance between researchers is not really important; what is ultimately
of value is the content of disagreements and the insights that discussion can provide for
refining coding frames. The greatest potential of multiple coding lies in its capacity to
furnish alternative interpretations and thereby to act as the “devil's advocate” implied in
many of the checklists in alerting researchers to all potentially competing explanations.
Such exercises encourage thoroughness, both in interrogating the data at hand and in
providing an account of how an analysis was developed. Whether this is carried out by a
conscientious lone researcher, by a team, or by involving independent experts is
immaterial: what matters is that a systematic process is followed and that this is rendered
transparent in the written research project (p. 1116)
In addition to the use of experts, a literature review conducted after the data analysis
provided a form of scholarly triangulation to the theoretical findings reported. That is, when
detailing the Theory of Web-Mediated Knowledge Synthesis below, references to both the
original literature used to frame this study and to a literature that was explored after the
emergence of new phenomenon are discussed. For instance, the emergence of repurposing as a
component of Web-mediated synthesis was not predicted or framed by the initial literature
review. However, after its emergence, evaluation by experts, and further analysis, I explored
literature in related areas. Typically, one of the primary contributions of this sort of literary
analysis would be that it provides a frame for readers to better understand the phenomenon and
how it emerged in the context of Web-mediated synthesis. However, this review of related
literature also provided additional insight into the foundations of the new phenomenon and
served to question my interpretations by providing alternative interpretations and competing
explanations, albeit from ideas that are largely ancestral to repurposing in the context of Web-

58

mediated synthesis. That is, examining the roots of repurposing through lenses of learning,
creativity, and design, provided one final way to increase conceptual objectivity.
Finally, it is also important to address the role of objectivity and subjectivity in theory
development. Given the lack of scholarly work that directly addresses synthesis in a Webmediated environment, the subjective insights that I had prior to embarking on this study
regarding learning on the Web and creativity (e.g., DeSchryver & Mishra, 2008; DeSchryver &
Spiro, 2008; Mishra & DeSchryver, 2009; Spiro & DeSchryver, 2008; DeSchryver & Spiro, in
preparation;) were valuable in order to determine what to look for and what I was seeing when
evaluating related data. With that in mind, the study was designed with the most appropriate
design to maximize the utility of that specific researcher experience and expertise, balanced with
objective analysis of the conceptual insights that evolved from that experience and expertise.
The result is an exploratory case study. As specific evidence of the contributions my prior
expertise balanced with objective analysis, pay particular attention below to the treatment of
VCS, virtual conceptual simultaneity. As should be clear from above, at the outset of the study I
operationalized VCS as one of the three primary forms of synthesis to look for in advanced
learners using the Web. However, VCS did not emerge as an independent construct in the data.
The idea evolved, through a transparent process of data collection, initial analysis, expert insight,
secondary analysis, and final expert insight to a more broadly defined category of generative
synthesis with several possible manifestations, one of which is VCS. Like any scholarly study,
the literature framed the idea of VCS, but it was my work with Dr. Rand Spiro that provided that
literature. As such, the subjective insights and relative expertise that I had about VCS could not
be discounted before the study, or during the observation and analysis, nor should they have
been.

59

Chapter Summary
This chapter described the design, participant selection procedures, task development,
data collection, research session details, and data analysis procedures used in this study. In
particular, it highlighted the use of abductive coding, constant comparative analysis, and experts
in the data analysis. These experts helped in several areas: (1) establishing content validity in
the survey used to assist in participant selection; (2) selecting final participants who
demonstrated the ability to use the Web in sophisticated ways for leaning; (3) creating clearly
worded tasks that fit the characteristics desired for studying synthesis; (4) authenticating the
modification to existing synthesis elements and the conceptualization of emergent synthesis
elements; and, (5) providing reliable interrater evaluations of these synthesis elements for the
two in-depth case studies.

60

CHAPTER 3
FINDINGS ACROSS ALL EIGHT PARTICIPANTS
Chapter Introduction
This chapter presents the results of data analysis across all eight participants. These data
provided insights into the process of Web-mediated synthesis from two levels. First, the data
indicated that the process is more complicated than that outlined by the theoretical framework
above, resulting in the expansion and elaboration of my theoretical operationalization of the
elements of Web-mediated knowledge synthesis. In doing so, the data corroborate the existence
of synthesis for meaning and creative syntheses, contribute to an expanded and recategorized
notion of VCS as in-the-moment insight, and substantiate three new synthesis elements that
emerged from the qualitative analysis. Second, the data provide detailed statistics about how
these elements manifested across all eight participants.
In doing so, this chapter primarily addresses RQs 1, 2, 4, and 5. That is, with regard to
RQ 1, this chapter demonstrates how synthesis for meaning and creative synthesis show up in the
data mostly as expected, and how VCS did not appear as expected. It also describes the “other
forms of generative synthesis” that emerged from the data to address RQ 2. The role of
background knowledge is discussed in broad strokes and differentiated from the idea-play
activities of the participants, targeting RQ 4. And, the use of notes across the eight participants
is detailed to address RQ 5.

61

Overview
The theoretical framework constructed above operationalizes Web-mediated synthesis as
three separate processes that work independently or in concert to produce synthesis experiences.
However, this lens was too narrow to explain the phenomenon fully during data analysis. A
more complex explanation of both synthesis for meaning and generative synthesis developed
whereby the initial elements evolved and new elements emerged. Table 1 highlights the
descriptive statistics gathered for the final six elements that more fully explained Web-mediated
synthesis as it appeared in this study, including (1) divergent keyword search phrases; (2)
synthesis for meaning; (3) in-the-moment insights; (4) repurposing; (5) reinforcement; and, (6)
creative synthesis. Thereafter, I provide more details for each.
Table 1
Descriptive Data of Synthesis Elements,
All Participants
Reading-to-Learn

Reading-to-Do

Totals

52

70

122

71

41

113

40

28

68

Repurposing

17

24

41

Reinforcement

19

10

29

3

13

16

Divergent Keyword
Search Phrases
Synthesis for
Meaning
In the Moment
Insights

Creative Synthesis

Divergent keyword search phrases. As noted above, the use of keyword search phrases
is one of the primary reading strategies that differentiates Web-based reading from its offline and

62

hypertext ancestors. This data in this study confirmed this literature. The eight participants
conducted 158 keyword search phrases and utilized 242 unique keywords.
Table 2 indicates that the eight participants used 158 unique keyword search phrases
during both tasks. They used 67 during the eight hours of reading-to-learn tasks and 91 during
the six hours of reading-to-do tasks. Of the phrases used, 36 were consistent (i.e., using words
that were provided in the task prompts), and 122 divergent (i.e., using words not provided in the
tasks prompts). Of the consistent phrases, 15 were used during the reading-to-learn tasks and 21
during the reading-to-do tasks. Of the divergent phrases, 52 were used during reading-to-learn
tasks and 70 during reading-to-do tasks.
For example, based on the tasks provided above, the following keyword search phrases
were considered consistent: [how to lessen the effects of climate change], [why people don’t
change behavior climate change], [college students and climate change], and [climate change and
people’s behavior]. However, the following keyword search phrases were considered divergent:
[does God care about climate change], [economic reasons for behavior change], [how people
assess risk and their behavior regarding climate change], and [theory of planned development].
Table 2
Descriptive Data from Keyword Search
Phrases Used, All Participants
Consistent

Divergent

Totals

Reading-to-Learn

15

52

67

Reading-to-Do

21

70

91

Totals

36

122

158

63

Across all participants, 242 unique keywords comprised the phrases used. During the
reading-to-learn task, 133 unique keywords were used, and 153 during the reading-to-do task.
When keyword search phrases were consistent, 53 unique keywords were used, and when they
were divergent, 222 unique keywords were used. As shown in Table 3, the average keywords
used per search phrase were 5.87 across all participants and tasks, 5.52 in the reading-to-learn
task, 6.11 in the reading-to-to task, 5.77 in consistent phrases, and 5.9 in divergent phrases.
Across all participants and tasks, only once did the exact search phrase get used - [college
students and climate change] - two different participants used it.

Table 3
Average Length of Keyword Search
Phrases, All Participants
Consistent

Divergent

Totals

Reading-to-Learn

6.13

5.38

5.52

Reading-to-Do

5.52

6.28

6.11

Totals

5.77

5.90

5.87

Synthesis for meaning. Above, both reader intent and outcomes described synthesis
across the literature for reading comprehension, multiple text comprehension, and Web-based
reading. That is, most prior conceptions of synthesis in these areas focused on how readers
organize, combine, compose, rearrange, rewrite, compile, and structure text(s) when synthesizing
information with the explicit goal to make meaning and understand of the text(s) being read
(Duke & Pearson, 2002; Dole et al, 1991; Glister, 2000; Hartman, 1995; Leu, Zawilinski, et al.,
2007; Mayer, 2000; Rouet, 2006; Stromso & Braten, 2002). This active construction involved
64

sorting relevant from irrelevant information and organizing it effectively (TICA Basic Skills,
n.d.) and common reading comprehension strategies such as summarizing (Dole et al, 1991; Hidi
& Anderson, 1986), inferencing implied author meaning or intent (Dole, et al; Kintsch, 1994;
Scharere, Pinnell, Lyons, and Fountas, 2005), and relating to background knowledge (Dole, et
al).

The data in this study confirmed extensive use of this process to make meaning of the Web

text(s) encountered with 113 instances of synthesis for meaning across all eight participants and
both tasks.
In-the-moment insights. In the theoretical framework for this study, Spiro’s (2006e)
enabling cognitive affordance of virtual conceptual simultaneity (VCS) underscored the first
form of generative synthesis – which I called VCS synthesis. The data from this study
confirmed several instances of VCS based synthesis, but many more were demonstrated based on
a broader category of in-the-moment insights. That is, the participants in this study frequently
demonstrated insights based either on: (1) a single Web resource; (2) multiple Web resources
(e.g., as explained by VCS); (3) activity unconnected to the Web (e.g., based on background
knowledge or idea play); or, (4) a combination of the prior three possibilities. The process of
selective combination (Sternberg, 1985), whereby the reader combines “what might originally
seem to be isolated pieces of information into a unified whole that may or may not resemble its
parts,” appeared in play within a single resource, from background knowledge, or through ideaplay. VCS synthesis as I operationalized it extended that notion based on the speed of the Web
to allow for conceptual simultaneity across multiple resources. And, the enabling affordances of
virtual conceptual simultaneity underscored the instances demonstrated. In-the-moment insights
were primarily task agnostic, where the conceptual/unified whole that emerged was more often
related to the resource(s) recently visited than the primary task or purpose for reading on the

65

Web. The data demonstrated multiple instances of this process with 68 instances across the
participants and tasks.
Repurposing synthesis. Several instances of synthesis that were not explained by the
preliminary framework appeared in the data. Among them, a group based on the idea of
repurposing ideas emerged. That is, many times, the readers recast, reused, or otherwise
tweaked ideas they encountered in the text(s) to fit a new context. These instances demonstrated
generative experiences given that the value-added to the idea originated with the reader. They
actively transformed a singular idea much in the way that Kress (2000) described how
transformation, remaking, editing, and juxtaposition of texts create new and alternative
meanings.

In these cases, the relationship between the original and repurposed idea was one-to-

one (as opposed to many-to-one) with a parallelism of meaning remaining intact from the first
occurrence to the repurposing. The data demonstrated multiple instances of this process with 41
instances across the participants and tasks.
For example, after reading about the idea of “going green for lent” one participant wrote
about how “finding a day in religion where they can give up something…maybe a common
thread between religious…religions can join together in interfaith.” The commonality between
Lent and his idea is clear, but he transformed this idea to include the ecumenical component and
the importance of interfaith. Similarly, after reading through a PowerPoint on the Web about
persuasion and specifically how expert opinions are used in advertising, another participant
noted how business leaders in high SES situations might be less inclined to listen to experts, and
asked himself whether they might be more willing to “listen to one of their own.” He continued
writing, “expert opinion might fit here, with an industry leader or expert giving testimony.”
Again, this notion of “one of their own” as an expert is a transformed idea based on the use of

66

experts in advertising, and reshaped what an expert is for the context of high SES business
leaders.
Reinforcement synthesis. Another form of synthesis that emerged from the data
involves the reinforcement of other synthesis experiences. That is, when the participants
encountered ideas that had previously been synthesized in different resources, a reinforcement of
that idea occurred, often from a different perspective or in a different context. Given the scope
of resources available on the Web and speed of access to them, the potential for such interactions
is much greater than in offline texts or even hypertext systems. In the data from this study, these
experiences provided confirmation to the readers regarding their own ideas and interpretations,
and often presented with increased affect. It is important to note that these interactions were not
revisitations of the same text-based information, as outlined in CFT (e.g., Spiro, Coulson,
Feltovich & Anderson, 2004). Nor were they just occurrences the same idea, but just in a
different resource as accommodated by CFT-W (DeSchryver & Spiro, in preparation; Spiro &
DeSchryver, 2010). They were new instances of synthesized knowledge such as that provided by
in the moment insights or repurposing. In some cases, the reinforcement of those ideas led to
even further elaboration with even more generative ideas. The data demonstrated multiple
instances of this process with 29 instances across the participants and tasks.
For example, the same participant noted above suggesting the importance of interfaith
cooperation on the topic of climate change, later read “Earth Keepers, a coalition of 10 faiths
and 150 congregations across Michigan’s Upper Peninsula [who have] joined together to protect
the regional environment.” He then noted, “If they are already doing this…contact the ones who
are doing this well and work with them. Don’t have to reinvent the wheel.” This text confirmed
his earlier thoughts about the value of interfaith collaborations, and provided him with an

67

elaborative element in his emerging solution to contact interfaith agencies that are doing this well
with college students through their faith.
Creative synthesis. The data confirmed the existence of a task/problem based creative
synthesis as outlined in the preliminary framework. It presented as a variety of combinations of
the other forms of synthesis, especially in the reading-to-do tasks. It appeared iterative,
including as part of this process text(s) from the Web, background knowledge, and idea-play.
Note taking was also a crucial part of creative syntheses. In several cases, the participants started
the process of creative synthesis by brainstorming without the Web, and then used the Web to
explore, confirm, and elaborate on those ideas. Many participants also discovered new ideas
thereafter directly on the Web and added them to the evolving creative synthesis. Across all
eight participants, the reading-to-do tasks included at least one extended interaction that
demonstrated creative synthesis. Given the detail and space required to properly address how
this happens, I present examples of creative syntheses in the detailed illustrative case studies for
Olivia and Beth, below (Chapters 4 and 5).
Data from the case studies provided three additional considerations to the environment of
Web-mediated synthesis, the role of background knowledge, the use of “soft” resources, and the
process of note taking.
Background knowledge. Background knowledge played a key role in many of the
instances of synthesis demonstrated in this study. This is not surprising, since background
knowledge is foundational to successful reading comprehension (Afflerbach, 1990; Anderson &
Pearson, 1984; Clymer, 1968; Kintsch, 1998; Kintsch & Rawson, 2005). However, though at
various times it played a role in each of the component processes of Web-mediated knowledge
synthesis, the role was rarely explicit. In an attempt to deconstruct its contributions to the

68

processes to the extent that it was possible, the in-depth case analyses below operationalize
background knowledge considered in the context of how it informed each of the synthesis
elements. For instance, if Olivia integrated background knowledge to understand the text(s) she
read, I report it as part of the synthesis for meaning process. However, background knowledge
also contributed to the various generative processes Olivia experienced. That is, background
knowledge at different times informed the choice of divergent keywords search phrases she
selected and at others contributed to her in-the-moment insights. Still in others, insights were
based entirely on background knowledge. She also repurposed and reinforced ideas that were
themselves informed by background knowledge. Across all of these cases, the extent to which
background knowledge informed her syntheses represents a continuum. However, in most cases,
it is difficult to determine the level of contribution background knowledge provided. As such,
where possible, I identify and detail these instances of background knowledge used, but do not
speculate as to the strength of the contribution that it makes.
There are also times that I differentiate background knowledge from “idea-play.” This
notion of idea-play is based on the conceptions of play underlying creativity identified above in
the creativity literature. For instance, Einstein discussed how combinatorial play helps to
connect ideas and concepts that have rarely, if ever, been combined. Amabile, Hadley and
Kramer (2002) described this as “throwing a bunch of balls into cognitive space, [and] juggling
them around until they collide in interesting ways” (p. 52). And, McWilliam (2009) described
how creativity of this type often required that this idea play occur among disparate ideas for a
long enough period of time to generate new idea spaces or, as Jackson (2006) phrased it, move
ideas from one state to another. As such, this juggling of ideas over a period of time as identified
below represents idea-play. It manifests differently from a recall of background knowledge or an

69

insight based on background knowledge in that recall or insight is typically apparent in a
relatively quick time frame and is often singular in nature. Conversely, idea-play as a juggling of
multiple ideas from background knowledge, multiple ideas from the Web texts(s), multiple ideas
from the participants’ notes, or various combinations thereof, all occurring over a longer period
of time than is apparent for recall of background knowledge or an insight based on background
knowledge.
Soft resources. The following descriptive data regarding Web sites visited and keyword
search phrases used was collected from all eight participants, differentiated by the reading-tolearn and reading-to-do tasks, and then aggregated across the study. Table 4 indicates that the
eight participants visited 233 unique Web sites during both tasks. They visited 137 during
reading-to-learn tasks and 96 during reading-to-do tasks. Of the sites visited, 103 were hard
resources (e.g., books, NASA, universities, national and international newspapers/magazines,
governmental agencies), and 130 soft resources (e.g., blogs, aggregators, individuals Web sites,
institutes, sites based on a cause). Of the hard resources, 67 were used during the reading-tolearn tasks and 36 during the reading-to-do tasks. Of the soft resources, 70 were used during
reading-to-learn tasks and 60 during reading-to-do tasks.
Table 4
Descriptive Data from Web Sites Visited,
All Participants
Hard Resources

Soft Resources

Totals

Reading-to-Learn

67

70

137

Reading-to-Do

36

60

96

103

130

233

Totals

70

Participants visited the same Web sites only seven times. Two participants visited three
of these sites (a blog entry about persuasion techniques and propaganda, the global climate
change and health page on the World Health Organization site, and Greenbiz.org). At least two
different participants visited the other four of these sites, but always to different articles or
resources on the site (The Guardian newspaper, by five participants; the Washington Post,
Huffington Post, and Nature.org by three participants, but always to different articles; the
Climate Progress and Green Facts Web sites two different participants; and, Wikipedia which
was visited 12 times, by four different participants, but always to different entries).
Note-taking. Seven of the eight participants took extensive notes, at least two pages and
as many as five. One participant took only about one-quarter of a page of notes. Three
participants used all offline notes, two used Microsoft Word and some offline notes, one used
only Google Docs, and one started offline and finished in Microsoft Word. Six of these seven
started separate notes for the reading-to-do task.
The notes varied from verbatim to summary. All notes taken on the computer were
linear in nature, and often took bullet, list, or outline form. The offline notes were largely linear,
but non-linear affordances were utilized (e.g., arrows, circles, brackets, sectioning). Additional
mark-ups identified important elements (e.g., asterisks, stars, underlining). Two participants
used bookmarks. Two others pasted URLs into their computer-based notes. Four participants
revisited their notes during the reading-to-do tasks, but only minimally. More importantly,
nearly all of the synthesis elements reported above resulted in or from note taking.

71

Chapter Summary
This chapter addressed issues relates to RQs 1 and 5, by corroborating several ideas about
synthesis based on the theoretical framework for this study - synthesis for meaning, creative
synthesis, and the importance of note-taking. With additional attention to RQ 1, it recategorized
the generative concept of VCS to be described as in-the-moment insights, which better explains
what is seen the data (that is, insights are less constrained by the expectation of multiple sources,
and the updated definition allows for insight to occur in a under a variety of simpler conditions).
With regard to RQ 2, the chapter also discussed how the definitions of synthesis provided by the
theoretical framework were not sufficient to account for the phenomenon that emerged in this
study and described three new ways to conceptualize Web-mediated synthesis - divergent
keyword search phrases, repurposing, and reinforcement. Data regarding keyword search
construction and soft resources used was presented to address how the synthesis elements from
RQ1 and RQ2 manifest on the Web. To address the role of background knowledge in Webmediated synthesis examples from the data are utilized to differentiate how background
knowledge and idea-play activities differed across participants, targeting RQ 4. The chapter also
highlighted the extensive use of soft resources in both the reading-to-learn and reading-to-do
tasks. And, finally, the chapter reported statistical data on how often each of these elements
occurred.

72

CHAPTER 4
IN-DEPTH CASE STUDY #1 - OLIVIA
Chapter Introduction
This chapter provides detailed illustrations and discussion of the broad themes about
Web-mediated synthesis elements outlined in Chapter 3, based on an in-depth case study of one
individual participant, Olivia. The chapter explores each synthesis element individually and then
demonstrates how they worked in concert to comprise creative syntheses. The chapter finishes
with conclusions drawn from the case study about specific patterns of use, interactions, and
relationships among the synthesis elements.
In doing so, the chapter provides specific examples of the broad themes that emerged
from analysis of all eight participants. However, the focus of this chapter is on how synthesis for
meaning, insight, repurposing, reinforcement, and creative synthesis manifested for Olivia, by
providing detailed examples and illustrations that further explore RQs 1 and 2. Variation and
nuance in the application of these themes is demonstrated across specific examples in this one
case study (e.g., by highlighting the uniqueness of the various ways Olivia synthesized for
meaning). This chapter also emphasizes the multiplicity of these ideas and how they interact and
work in concert, the primary concern of RQ 3. The specific roles of background knowledge and
idea-play are highlighted in several examples with respect to RQ 4. And finally, to explore RQ
5, most examples emphasize the substantive contributions note-taking provides to Olivia’s
synthesis processes.

73

Olivia
Olivia was a 25-year-old food science and integrative toxicology doctoral student. I
selected Olivia’s data for closer examination based on data indicating that she demonstrated the
most extensive use of synthesis in her reading processes of any participant (120 instances). She
also demonstrated a diversity of synthesis experiences and the ability to think-aloud in great
detail. A detailed profile of Olivia is provided in Appendix A, including a reporting of her
background knowledge and experiences on the topic of climate change collected at the beginning
of her research session.
Olivia demonstrated 120 instances of synthesis during her sessions. Table 5 outlines the
descriptive statistics gathered.
Table 5
Descriptive Data for Olivia
By Task
Reading-to-Learn

Reading-to-Do

Totals

2

3

5

10

8

18

Hard Web Sites

7

4

11

Soft Web Sites
Synthesis for
Meaning
In the Moment
Insights

16

3

19

25

15

40

6

8

14

8

15

23

12

9

21

0

4

4

Consistent
Keyword Search
Phrases
Divergent Keyword
Search Phrases

Repurposing
Reinforcement
Creative Synthesis

74

In addition to these statistics, several other characteristics described her session. First,
she took a lot of notes; all by hand on the legal pad provided. Second, she revisited and
referenced the task instructions multiple times during each session. Third, she was largely nontechnical in her approach to reading online. That is, she rarely used multiple tabs, did not
bookmark or clip resources, did not save any URL’s during the sessions, and rarely used any of
the additional services provided by Google beyond the basic Google Web search. Olivia
integrated well with the think-aloud protocol and never had to be prompted.
The data collected during Olivia’s sessions provided multiple illustrations of each
element of synthesis noted above. Just below, several illustrations are provided, and then
conclusions about the specific patterns of use, interactions and relationships among these
elements are explored.
Divergent keyword search phrases. Olivia balanced her use of divergent keyword
search phrases between the reading-to-learn and reading-to-do tasks. Table 6 lists all of the
phrases used in the first session.
Table 6
Keyword Search Phrases
Olivia, Reading-to-Learn
Time

Keyword Search Phrase

Type

Origin

:38

opinions on climate change

D

BG knowledge

3:23

people’s opinions on climate change
why people change behavior on
climate change
Why not to change my behavior on
climate change

D

iteration

C

task

C

task

7:03
19:35

75

Table 6 (cont’d)
28:44

D

Web text
review of notes & Web
text

D

insight

D

Web text

42:39

mitigation
why climate change mitigation
beneficial to government

D

iteration & Web text

43:53

benefits of climate change

D

49:53

religious views on climate change
different countries perspectives on
climate change

D

iteration
Web text, BG
knowledge & task
Web text, BG
knowledge & task

34:30
38:04
38:57

57:02

Barriers in market for climate change
why people don’t care about climate
change
why climate change beneficial to
government

D

D

Several key findings emerged from this data. First, consistent with the participants as a
whole, Olivia used many more divergent search phrases than consistent. Second, her
keyword searches originated from a variety of sources. This included the Web text,
background knowledge, the task (for consistent searches) and iteration. She also often relied on
more than one source for keyword development. For instance, just before switching her
direction to search about [religious views on climate change], Olivia said “So, and when I just
read about the science, part of this topic says to open up to others, I’m wonder if any…if religion
affects your view on climate change.” When interviewed, Olivia further described her decision
to use the [religious views on climate change] search phrase as follows:
I was reading about the government citing all these scientific articles, and in my
mind when I think scientific articles, we try to search for the truth, I’m not a
particularly religious person, but in religion they try to do the same thing through

76

different means. So I was like that can give me perspective, people tend to follow
their religion, people tend to do their behaviors, based on their religion. Think
about how the government can cite all of these articles, but if people aren’t
listening to the science, and listening to something, and something else that can be
viewed kind of opposite, or kind of go against science would be religion, so I
wanted to get the opinions of that.
In this case, a combination of the Web text, the task instructions, and her background
knowledge together guided her choice of the next direction to go and the keywords to
select. Specifically, as she re-read the task, the instructions to “be open to new ideas”
impacted her at this point.
In addition to the notes Olivia took also informed and guided her decisions about
keyword search phrases. At one point, she stopped reading the Web, and just took time to
review her notes “And now I’m just looking over what I’ve done thus far…Why people
change behavior…why people don’t change behavior…so I’ve gotten in to…but why
would…I guess…something that would boggle my mind, is to why people knowing there is a
problem, not choose to care…” Thereafter, she searched for [why don’t people care about
climate change]. This demonstrates how the notes served as a one of the sources of her ideaplay about constructing keywords. However, it is also indicates that Web text played a role
in the process, given that “they don’t care” was listed as the first reason under her “people
don’t change behavior” section, which was based on repurposed ideas from an article read 25
minutes earlier. In this way, notes provided “downstream” access to ideas that the reader
synthesized earlier, which can then be used to inform keyword choices.

77

Finally, one of Olivia’s keyword search phrases was constructed by what was best
represented by an in-the-moment insight, with no direct connection to background
knowledge, the Web text, the notes, or the task, but. That is, at one point, after reading Five
Reasons Why We don’t Care About Climate Change and taking several notes, the last of
which is “not enough time to make a big difference,” Olivia immediately said “And I wonder
why then, uh, for some reason in my head I just got thinking…I wonder why it’s beneficial
for countries to make climate change a problem. So I’m actually going to go back and type
in…” then following the search phrase [why climate change beneficial to government]. The
Web text just prior was completely unrelated. Just before that she read about production,
tariffs, exports and international trade in renewable energy products, which may have
provided a seed for this search, but was not a clearly direct influence. In this case, the idea to
move in this particular direction is best described as an insight, in that her phrasing was
generative of the text.
Table 7 lists all of the phrases Olivia used in her reading-to-do session. The
development and use of searching in this task was quite similar to her first session. She
utilized various origins, used almost the same number of phrases over the course of the task,
and relied primarily on divergent phrasing. A few differences emerged. First, the phrases
used in the learning-to-do tasks were equally divergent at the keyword level, but less
divergent at the topical level. That is, Olivia used about the same number of divergent
keyword search phrases in each task, but those used in the second only diverged within the
task parameters. As such, the nature of the keyword phrases constructed was very highly
connected to the type of task provided (i.e., during the learning task she expanded her

78

searches beyond the specific topic, where as during the reading-to-do she was more
“narrowly divergent”).
Second, through the task provided specifically listed three groups to focus her ideas
on, Olivia only searched for phrases related to high SES, college student, or religious groups
three times. The rest of her phrases were comparatively group agnostic. In doing so, Olivia
also almost doubled (from 8 to 15) the times she repurposed information and reduced her
number of syntheses for meaning (from 25 to 10). In this way, she evidenced a possible
relationship between task, keyword search phrases, and synthesis type.
Table 7
Keyword Search Phrases
Olivia, Reading-to-Do
Time
5:06
14:31

Keyword Search Phrase
Rich people change behavior global
warming

Type
D

task + BG knowledge

C

Web text & task

C

task

25:05

How to change a behavior voluntarily
How to change climate change
behavior voluntarily
Improving efficacy of voluntary
behavior change

D

Web text

26:31

Improving voluntary behavior change

D

Web text

31:55

Fun projects for climate change
Fun things for people to do to prevent
climate change
How to make people want to change
behavior about climate change
Effective way to reduce climate change
college
Recycling program Michigan State
University
High socioeconomic status
professionals and recycling

D

BG knowledge

D

iteration

C

task

D

task +notes

D

BG knowledge

D

task & BG knowledge

18:46

33:41
39:21
41:24
42:07
43:39

79

Synthesis for meaning. Olivia compiled and recombined information across texts and
from her background knowledge extensively during both tasks. She did this 25 times during the
reading-to-learn task and 15 times during the reading-to-do task, synthesizing information for
meaning at various times based on her background knowledge, a single Web resource, and/or
multiple Web resources. She also incorporated skills consistent with high achieving offline
readers, including summarizing and inferencing.
One of the simplest examples of Olivia’s activities that represented synthesis for meaning
is when she would read something out loud from a text and then take verbatim (or mildly
condensed) notes on that text. For instance, as she read a PDF of a research article from Purdue
about Public Attitude and Behavior About Climate Change, she encountered a list of factors
affecting pro-environmental behavior. She read “worldviews, perception of personal and social
risks, political philosophy, personal discomfort, personal responsibility, and social norms” and
took verbatim notes of these ideas. Multiple incidents similar to this comprise her construction
of an intertext (Afflerbach & VanSledright, 2001; Hartman, 1995; Stromso, Braten, &
Samuelstuen, 2003; Wolfe & Goldman, 2005) that aids in developing a global coherence and
meaning of the multiple texts she reads with respect to the topic she is learning about.
Similarly, after reading a Web page about the public opinion on climate change, and then
an article about the impact of climate change in Bangladesh, Olivia paused to reflect on what she
had learned thus far and synthesized for meaning based on inference and summary (see
Appendix B for details). During her comments and notes Olivia made an inference (Dole, Duffy,
Roehler & Pearson, 1991) about an idea implied by the author, but not explicitly stated in the
text (i.e. he did not directly say the type of media received impacts public opinion on climate
change, but the text insinuated this), and then summarized the authors perceived intentions in her

80

own words (i.e., through comprehension, condensation, and transformation of the ideas (Hidi &
Anderson (1986)). In this sense, her comments and notes made up a single entry into an
evolving “synthesis for meaning” of the texts she was reading on the Web, based primarily on
the text from a single page (though their is likely some influence of background knowledge
given its general contributions to inferencing).
In another case, Olivia synthesized meaning through a single soft resource that
provided multiple perspectives to her understanding. This occurred when reading a comment
page on Reddit (a “social news” site that is user generated and allows comments on the stories)
about the environment. The details of her interactions on this page are provided in Appendix C.
Based on her reading of comments from several different users, she synthesized meaning about
the role of inconvenience, beliefs, industrial/corporate influence and externalities, the sum of
individual behavior, being affected personally, and morals through both verbatim and summary
notes. The extensive multiplicity of ideas she synthesized from just one resource was a direct
result of the structure of the page - that of a discussion based on comments about a single article.
As such, it demonstrated the potential importance of soft resources to the Web-mediated
synthesis process.
Finally, Olivia also synthesized information for meaning from multiple sources at the
same time for meaning. While skimming a long PDF, she read, “A recent survey found that,
among Americans who believe there is solid evidence of global warming, only a minority…think
that we will have to make major sacrifices…” and then commented “Oh, so people know that its
happening, but they don’t know…so they know it is happening, but they don’t
know…sacrifices…but they don’t care to make a change because they don’t actually believe it is
a big problem.” She wrote in her notes “know happening ≠ make Δ à actual problem.” Her

81

comments and notes synthesized ideas from the current page, the idea that people just are not
concerned or don’t care from a different article she read 10 minutes earlier (and took notes
about), and the idea that about of individual perceptions of the seriousness of the problem from a
resource five minutes earlier.
In addition, a deeper examination of the conditions surrounding the various processes
demonstrated to synthesize meaning indicates how Olivia’s path to synthesis for meaning was
a complex interaction among various elements. For example, the very first synthesis for
meaning example above demonstrates how various decisions and precursor events contributed to
its manifestation (see Appendix D for a detailed accounting). Though this appears to be a simple
interaction, the resulting information copied is central to her developing understanding of the
topic, and her path to finding it is important to consider. This path shows how Olivia’s
background knowledge about the text structure (i.e., the value of Google scholar vs. Google
search), her ability to evaluate the relevancy of an article (both the fact that she recognized the
first article was not about global warming, and that the Purdue article looked good), her
skimming proficiency, and her ability to separate relevant from irrelevant information within a
relevant text all contributed to the synthesis for meaning she experienced.
In-the-moment insights. Insight comprised or contributed to many of the ideas that
Olivia synthesized during both of her tasks, and the conditions for insight varied across her
activity. Some insights were isolated incidents unrelated to the text(s), others based on providing
generative value-added ideas based on the text, and still others in the midst of synthesizing
information for meaning. For instance, after reading about technology, personal sacrifices, and
making related comments and notes, she read the text “most people are willing to make some
financial sacrifices.” She then commented “Yeah, that can be it, too, so not change

82

behavior…money…” making notes about money. Without hesitating, she added, “also probably
public perception, I’d say…of why people would change their
behavior…because…it’s…probably because it’s like cool…it’s cool to recycle…” During these
comments she wrote “public perception (cool)” in her why people Δ behavior section.
It is impossible to know exactly where the idea of public perception came from based on
the data collected. However, the context of her reading provides several clues. First, this idea
was not explicitly stated nor implied in the then current or any previous text(s) she had read thus
far in the session. In that way, it did not represent a synthesis for meaning as defined in this
study. It is also not background knowledge used to help her understand the meaning of the
current or previous text, since public perception or a related idea is not in the text. Second, in the
18 minutes of online reading prior to this insight, Olivia read about various topics, including:
climate change around the world, media, personal interest, world views, social risk, political
philosophy, personal discomfort, money, demographics, incentives, social norms, values,
responsibility, fear, risk, health, technology, and sacrifice. The density of ideas available to her
and the rapid successful with which she moved from one to another created an infinite number of
possibilities for combination and recombination of ideas, a creative stew of potential for insight.
As such, it is viewed as a generative form of synthesis - an insight. That is, it may represent a
selective combination (Sternberg, 1985), enabled by the cognitive affordance of virtual
conceptual simultaneity (Spiro, 2006e) of the above ideas, and likely some background
knowledge, that once amalgamated represented a “conceptual whole great than the sum of the
parts” (Spiro) “that may or may not resemble its parts” (Sternberg).
Olivia made several other insights explicitly related to the text(s) she read, when she
provided value-added ideas to those text(s). Four of these examples are detailed in Appendix E.

83

Of these four examples, the conditions for a generative insight about the text(s) vary. In the first
example, Olivia skimmed text from a hard resource very quickly, reading only text segments and
titles as she moved through several pages of text in less than a minute, which presented many of
the ideas from the extensive subset of topics similar to those listed for the public perception
example above. That is, the rapid successful of ideas in the text may have contributed most to her
insight. In the second case, Olivia read a soft resource with a much greater attention to the text,
reading at least part of, if not the entirety of, five different paragraphs before reading a sixth and
providing her own value-added to the text. That is, her attention to detail and a deeper reading
of the text may have contributed most to her insight. In the third example, there was very little
text on the page and the previous five minutes had been spent exploring only articles about
religion and climate change. Olivia read the title and just one paragraph of two sentences from
this page. In this case, given the isolated context of her activity, background knowledge may
have contributed most to the insight. In the final example, Olivia read only a few ideas in a hard
resource, but conducted a deep read of one paragraph where she repeated several sentence
fragments in her think-aloud. The pausing and repetition of ideas demonstrated by Olivia is
consistent with the notion of idea-play, which may have provided the foundation for this
particular insight.
Another example from Olivia’s sessions demonstrated how an insight occurred in the
middle of a synthesis for meaning. As noted above, while she read the Purdue research, she
synthesized a list of factors affecting individual behavior, worldviews, perception of personal,
and so one. While writing down these ideas, she read “political philosophy, knowledge, personal
discomfort” from the text, and then said “yeah, that’s a big one…personal discomfort, I never
thought of that…um…probably money, as well. I’m just going to write that in there, but I’m

84

going to make a note that he didn’t actually say that, yet.” She then continued synthesizing the
rest of the list into her notes, mostly verbatim. The notion of money as a contributor to
behavioral decision-making had not, to this point, presented itself in the text(s). It is worth
noting that her insight also occurred after making a few comments about the list of ideas she was
synthesizing. That is, stepping away from the text for just a spilt second to reflect in a metacognitive manner may have provided the opportunity for her to have this insight, in the
same way that incubation supports creativity.
Repurposing synthesis. Olivia repurposed information from multiple texts and her
background knowledge during both tasks. She did this eight times during the reading-to-learn
task and 15 times during the reading-to-do tasks. The examples provided in Appendix F
illustrate how repurposing occurred in several ways, from one context to another, by using an
idea in the opposite way it was intended, by reorienting the ideas to a different scale, by
repurposing an idea in two different ways, and by combining a repurposed idea with an insight.
Taken as a whole, these examples demonstrate several key ideas about repurposing.
First, it can happen with both task specific resources (e.g., the Purdue PDF on behavior change)
and resources that are somewhat far afield from the task (e.g., social justice, & trade barriers
for renewable energy). Second, repurposing happens in both soft and hard resources. Third,
an open mindset, such as that proposed by CFT-W (Spiro and DeSchryver, 2010) may be
important to finding information that is amenable to repurposing. Finally, the keyword
search phrases that directly preceded each repurposing may in fact provide the lens that
facilitates it. For instance, the fact that Olivia was specifically searching for [rich people change
behavior global warming] may have prefigured her ability to adapt the carbon tax to a tax
benefit, where as, if she had been searching for information about college students and climate

85

change, the potential relevance of carbon taxes may not have occurred to her. Similarly, in this
way, the keyword search process serves to reconnect the reader to their task in important ways
that may in fact facilitate different types of synthesis.
Reinforcement synthesis. Olivia reinforced her knowledge synthesis experiences
several times during each task. These included reinforcement of ideas synthesized for meaning,
as well as her own insights, including insights that were reinforced multiple times about the same
idea. Examples are provided in Appendix G.
Across these instances, several important benefits of reinforcement to generative
synthesis emerged. First, each reinforcing activity offers a different context to related ideas.
This provides the sort of loose-associations among ideas that are key to developing cognitive
flexibility and may make these ideas more flexibly adaptable for future application (Spiro, et al
2004). Many of the different resources provide different perspectives, as well, which helps to
ameliorate the danger of relying on the biases of just one or two resources. The reliability of the
different resources also compliment each other. That is, soft resources and hard resources
may serve to fill in the gaps that the other inherently demonstrate. For instance, opinions are
often more difficult to find in hard resources, and data may be more difficult to find in soft
resources. When one reinforces the other, the effect of the triangulation of the idea is that much
more powerful. Finally, reinforcement also occurs through resources from different time and
dates, lending a sense of stability to the idea.
Synthesis in multiples. At various times, Olivia demonstrated the ability to utilize
several elements of synthesis while reading the same text, as evidenced by how several of the
examples above overlap in some way. One example was particularly revealing of this point.
When reading the HuffingtonPost article entitled Five Reasons Why We Don’t Care About

86

Climate Change, Olivia utilized three different synthesis techniques in processing the text in
Figure 3.
First, from an evolutionary perspective, we are not programmed to take future threats as
seriously as immediate one; and though ice caps are melting and storms worldwide are
intensifying, for most of us, this is still going on somewhere out there. It’s not yet
happening where we can feel it and see it…
Second, it costs money to do some of the right green things. Much to my children’s
embarrassment, for example, I still drive and SUV. I bought it 130,000 miles ago when we
lived where snow could fall four feet in a weekend…Now that I live in a place where it
never snows (to my children’s chagrin) I have no reason to drive it, except the fact that I
can’t afford a hybrid.
Third, we have become so distance from nature, after centuries of trying to wield our power
over it and generations of relegating our children to spending more time indoors…
Fourth, many of us no longer believe we can make a difference in the world. During the
past 50 years, America’s population has doubled and industries from banking to cat food
have been supersized….
Finally, we think we never have enough time—and, in some ways, we’re right. In recent
decades, the rise of families composed of two working parents (or a single working parent)
has created a lifestyle that ill-equips us to address issues outside our own personal,
immediate concerns…
Figure 3. Text from Five Reasons Why We Don’t Care About Climate Change
She synthesized information for meaning by summarizing the first point in the article
when she read
we are not programmed to take future threats seriously,” then said, “we are not
programmed to take future threats seriously” says “evolutionarily…to takes
threats seriously…as immediate ones…though ice caps are melting…still going
on somewhere out there…it’s not yet happening..we can’t feel it or see it...so we
worry about more immediate things
and then noted “evolu: take threats seriously...worry about immediate.”

87

She used the text in the second paragraph to reinforce and clarify the idea of how money
might be a reason for people not to change behavior. The third paragraph resulted in an insight
on her part about valuing nature. The fourth paragraph reinforced and simplified the her idea
about collective vs. civilian considerations and how many people feel one person cannot make a
difference. And, she synthesized not having enough time to make a difference in the fifth
paragraph for meaning.
All of this occurred in a short editorial about climate change behavior. Her interactions
with the text in that article provided only a glimpse, however, of the level of sophistication and
complexity the data later demonstrated. Now, I explore how these elements worked in rapid
sequence, such as that shown with this example, and also worked in relative simultaneity,
providing the foundation for creative synthesis.
Creative synthesis. The above examples and those provided in related appendices of
Olivia’s synthesis for meaning, insights, repurposing, and reinforcements derive solely from
Olivia’s reading-to-learn sessions. There are two reasons for this. First, in her reading-to-do
(second) task, Olivia made reference to knowledge she had synthesized during the reading-tolearn (first) task. This provided more complexity to the notion of synthesis that is best
understood after discussing the “simpler,” mostly isolated and independent, examples above
from each element.

In her second session, “downstream” synthesis (or synthesis of syntheses)

occurred. For instance, when Olivia copied or summarized a segment of her own notes from the
reading-to-learn session (either from memory or by looking) into the notes from the reading-todo session, it represented a synthesis for meaning, with no value added from the reading-to-learn
notes text. However, when she used text from her reading-to-learn notes, regardless of where it
came from (synthesis for meaning or generative synthesis), in a way that provided a value-added

88

to it, it was a form of generative synthesis. For example, if she had an insight based on
something from her reading-to-learn notes that originally had been synthesized for meaning, then
integrated that insight into the reading-to-do notes, it represented a generative use of her own
notes from the reading-to-learn sessions. Conversely, if an insight from the reading-to-learn
session was copied or summarized in the reading-to-do notes, that was considered a synthesis for
meaning in the reading-to-do session.
The second reason that most of the examples above are from the reading-to-learn session
is that a majority of the synthesis experiences in the reading-to-learn task were relatively isolated
in their manifestation, simpler to identify, and interacted only minimally. However, in the
reading-to-do task, many of the synthesis activities were either multiple in nature (e.g., synthesis
for meaning and insight occurring at the same time), or in the context of broader creative
synthesis chunks. These are best considered in groups based on how they interacted. Though
more difficult to parse, it is this multiplicity and complexity that defines creative synthesis.
Three examples of Olivia’s creative syntheses are detailed in Appendices H, I, and J
(creative synthesis #1, #2, and #3, respectively). Several important ideas about creative
synthesis are demonstrated in these examples.
In the first example (see Appendix H), Olivia largely synthesized knowledge that had
itself been synthesized during the first session. She commented and took notes about increasing
the value of Mother Nature as related to the bible, which was a synthesis of as many as three
ideas, from three different resources, in the reading-to-learn task. That is, in Olivia’s reading-tolearn task the concept of valuing nature emerged from an insight based on Web text about how
people were too distanced from nature; the concept of religion possibly leading to passion about
climate change was an insight based on Web text about “environmentally minded clerics,”

89

“greens with a spiritual bent,” and “acquaintances that transcend ethnic, ideological and
theological obstacles;” and, the concept of religion possibly helping people be more invested in
mother nature emerged was repurposed from text indicating how radical environmentalists have
many parallels with Judeo-Christian traditions. In the first note during her reading-to-do task, she
synthesized all three of these ideas for meaning as she summarized them in her notes. Her second
comment and note about hymns demonstrates how Olivia synthesized this same idea for
meaning a second time (i.e., the idea was synthesized for meaning in her reading-to-learn task
directly from the Web text, and it was synthesized for meaning from her reading-to-learn notes
into her reading-to-do notes). The third comment and note about the community aspect of
religion demonstrates how Olivia again synthesized several ideas from across different
resources in the reading-to-learn task, in particular repurposing the notions of collective action
and groups. That is, in session one, Olivia used the ideas of collective versus individual action to
explain why people might not change their behavior. Similarly, in the context of religion, she
proposed that a way to bring people together would be to do environmental things. In the
reading-to-do task, Olivia repurposed those ideas by reversing the relationship, so instead of the
environmental activity bringing people together, it was the established community of religious
people that might encourage people to clean up the environment or recycle. The seeds for this
repurposing were from a variety of synthesis interactions in her reading-to-learn session. She
synthesized the idea of collective vs. individual behavior for meaning from Reddit comments,
and later reinforced it with a HuffingtonPost article; recycling was an insight based on
background knowledge along with the importance of public perception; and, she synthesized the
idea of linking faith and greenery together for meaning from an Economist article.

90

Olivia’s second example (see Appendix I) comprises a creative synthesis with several
features worth highlighting. First, almost all of the syntheses that occurred in this example
were generative. Olivia oriented herself for searching based on repurposing the idea of caring
from the first task (which was in the context of climate change in her reading-to-learn task) to
whether rich people care about spending their money. She then repurposed carbon taxes on an
international scale to tax credits for rich individuals. An insight that rich people might care
about their families followed, just before she repurposed the idea of the wealthy transferring
resources to the poor from the context of distributive justice to the individual behavior of caring
and giving rich individuals. She paused and thought aloud about what she had been reading,
resulting in an insight about the importance of considering investment. Thereafter, she had an
insight about fame, recognition, and incentives. She then repurposed the idea of incentives to
college students and again to the religious group. This lead to the insight that college students
want to learn, followed by related insights about classes on global warming, integration of these
ideas across the curriculum, and a specific project related to exercise equipment producing
energy.
Second, these interactions demonstrate the complexity and layering of synthesis
elements between the reading-to-learn and reading-to-do tasks and among various
synthesis elements within the reading-to-do task. For instance, when Olivia repurposed the
idea of incentives from rich people to college students and then to religious people, she
incorporated and repurposed ideas from several different sources and times. Her proposal that
“we could reduce the cost of meals, for example, if they learn to recycle...I would definitely say
incentives” repurposed incentives from the discussion of rich people/fame/recognition, and was
connected to insights about recycling in the reading-to-learn session. The implied casual

91

relationship between lower cost meals (saving money) and pro-environmental behavior may also
have been repurposed from her ideas from about 10 minutes earlier in the reading-to-do session
when she repurposed carbon taxes and said about rich people “if they do this voluntarily...they
get a tax benefit.” When talking about the religious implications of incentives, she noted that the
incentive (itself repurposed from both rich people and college students) would be to do it for
God, the latter of which may be a repurposing of the notion of “creation care” in the reading-tolearn task, or an insight based on her background knowledge about religion. “Doing it for your
community” reinforced her repurposed ideas about community and religion from 10 minutes
earlier in the reading-to-do task. She also said “more of a moral obligation for those reasons”
which represented a repurposing of the ideal of morals being a reason for people to both change
and not change in the reading-to-learn task (i.e., it is repurposed for the religious context here,
given the absence of religion in the moral argument the first time around).
Third, though it would be easy to look at this series of events that make up the creative
synthesis and come to the conclusion that many of her ideas were a result of background
knowledge alone, this would be overly reductive. The data provide evidence of the impact of
the current and former Web text(s) on her interactions. First, the idea of “caring” emerged in
Olivia’s initial comments about rich people. When she said “if they care about spending their
money or if they don’t care...” Olivia repurposed the idea of caring about climate change from
the reading-to-learn to the high SES task. The importance of caring emerged even more during
the ensuing creative synthesis in her reading-to-do task. First, Olivia had an insight that rich
might care about their families. Then, she talked about how caring rich people might give to
future generations. Furthermore, after reading a paragraph about
preferences/beneficiaries/benefits and reviewing her notes about what rich people might care

92

about, she had an insight about determining what people are invested in. This concept of
investment undergirded her subsequent insights about fame, recognition, and the importance of
incentives. It is this switch in terminology from caring to investment that kept the momentum of
this particular creative synthesis moving forward, and the switch came just after reading a long
paragraph in the Web text. Though it is unclear what, if any, specific text in that paragraph made
Olivia switch from caring to incentives, the paragraph contained several words and phrases that
are financial in nature (poor, resources, benefits, cash, money, beneficiaries, wealthy,
productivity), and the word incentive connotes a more financial relationship that caring. Given
this relationship, and the timing that occurred (i.e., the switch to incentives came right after
reading this paragraph), it is arguable that this text informed her switch from caring to
investment. In this way, Web text potentially provided seeds to the two ideas that permeated
nearly all of the generative elements in this creative synthesis.
Fourth, this example also demonstrates how notes can be important to a creative
synthesis. This was especially evidenced in the final two synthesis elements. Just before having
an insight about how college students want to learn, Olivia reviewed her notes. Her comments
“they want to be cool, save money, have fun” were a summary of the first four lines of her notes
about college students. This idea-play directly preceded her insight about students wanting to
learn, which then evolved to her ideas about classes on global warming and cross-curricular
integration. She then went back to her notes, reading “So they can take a class…reduce cost of
meals…they can go someplace…” then said, “actions that they can do that would be fun…” and
continued reading her notes, “fun other countries…travel in other countries…groups, fun, meet
people…feel good about yourself…reduce cost…” before having the idea about the exercise
project.

93

Fifth, the assertion that these activities comprise a single creative synthesis represented
by one long series of related interactions is largely based on what Csikszentmihalyi (1990)
called flow. Among other things, flow states are based on complete immersion, focus, and
engagement. Very little, if any, time elapsed between even the most divergent of ideas in this
sequence. For example, when transitioning from incentives that might influence religious people
and her insight about college students wanting to learn, there was no break between the last word
related to religion and the first words about college students. In fact, Olivia took a breath just
after saying “moral” at the end of the religious arguments, and did not take another until after
saying “college students” to begin the next thought. All of the transitions in this sequence were
similarly paced. Where as, at the end of this series of syntheses, Olivia finally took a break from
this flow experience (actually pausing for three seconds) before saying “Let’s go back here,
though, and look at um, look at how, so, I’m actually going go back here and I’m going to look
at, let’s see here, what was I just thinking...I want to go back, I had read something earlier that
talked about how to get people to change behavior...“ and then finally searched for [how to
change a behavior voluntarily].
Finally, is it important to highlight the integrated nature of all of the different
elements discussed just above for her second creative synthesis. In this example, the Web text,
her background knowledge, incidents of idea-play, and her notes all lead to several different
synthesis elements that comprised the creative synthesis she experienced. Given the organic way
in which these various components interacted, it is impossible to parse which was most
important. They all contributed to the ideas she constructed and were part of the flow of creative
synthesis she experienced.

94

In the third example of creative synthesis (see Appendix J), there are several more things
worth noting. First and foremost, it demonstrates the near-simultaneous occurrence of
syntheses elements. When reading the short one paragraph abstract, Olivia synthesized text for
meaning, had an insight, and reinforced two ideas. In addition, the Web text about “reducing gas
and reducing energy” was itself used three different times leading to various generative synthesis
experiences. The first time she read this information from the abstract of a Journal of
Preventative Medicine article, she had an insight about how these represent the easy things to
change in behavior. Then, she viewed this same information through the religious lens saying
“reduction in gas, reduction in...” before reinforcing the idea of creation care. Finally, she said,
in the context of rich people “how to get them to save...reduce energy...reduce gas...” which lead
to her repurposing the idea of hybrid cars and the eventual insight about the value of
unattainability and uniqueness with this group. Given that this text was encountered in the
context of an abstract (i.e. a Web page with one paragraph of text) this example provides an
argument for the potential value of short snippets of text when the task requires creative
synthesis. Similarly, early text about government regulations and subsides facilitated both
repurposing and reinforcement and later it supported insight, repurposing and reinforcement at
nearly the same time.
Creative synthesis #3 also demonstrates how both of the important articles selected,
Public Perception of Climate Change: Voluntary Mitigation and Voluntary versus Mandatory
Approaches to Climate Change Mitigation were based on ideas from the first, reading-tolearn task. That is, during the first session, Olivia encountered the word mitigation in her
Google results, but did not know what it meant. She opened up a new tab, searched for
mitigation, and explored several different definitions before finding one that fit the context of

95

climate change (i.e., to lessen). At that point, she also rephrased her search to include mitigation
as a term. The notion of mitigation, which she did not previously understand, became an
important part of her ongoing synthesis of ideas in her reading-to-learn task. Similarly, the
phrase “public perception” was an idea about which she had an insight in the reading-to-learn
task regarding its importance to individual behavior decision-making. As such, she may have
been primed to select these ideas in the different titles above, especially as they appeared
together in her reading-to-do task. This phenomenon did not apply to just Google results, either.
While browsing through a 17 page PDF, Olivia skipped many sections, but stopped at Clinton,
the Carbon Tax, and the Climate Change Action Plan, possibly because carbon taxes were
discussed in Web text earlier in her reading-to-do session (at the time leading to her repurposing
an idea about tax breaks for the rich).
Two of the instances of keyword use across the series of events in this creative synthesis
demonstrate how search phrase development itself can be more than just a gateway to
synthesis, but synthesis-like itself. The first time, as Olivia switched gears from the creative
synthesis #2 sequence, she said the following
...so let’s see...I’m going to go back here though and look at...um...look
at...how...um...so I’m actually going to go back here [to Google search box]...and
look at how...let’s see what was I just thinking...I want to go back, I had read
something earlier that talked about how to get people to change behavior...and
this thing says that my ideas need to be backed up...so I guess I’m going to
consult the literature...
and then typed [how to change a behavior voluntarily] for her search phrase. This process took
her about 45 seconds to complete, and though it is not a generative synthesis, it is very much a

96

synthesis for meaning of her own disparate ideas, bringing together ideas from the task and from
articles she had read earlier, in an idea-play like environment that resulted in her preferred
phrasing.
Similarly, after determining that she was “going to be done with” a PDF article about
climate change policy, Olivia returned to the Google search window, and said,
so, let’s see here...um...so let’s see here...I’m going to look up efficacy (as she typed the
word in the search box)...or improving efficacy (as she deleted efficacy and then typed
improving efficacy)...um...of...of...voluntarily behavior change
resulting in the search phrase [improving efficacy of voluntary behavior change]. This search
phrase was based on an insight she had made about enhancing the efficacy of the weak voluntary
programs she encountered in the previous article. In this sense, it is a form of generative
synthesis designed to open up her knowledge to a new area. After a quick review (less that one
minute) of one article about the “RP model” of behavior change, she returned to her results and
immediately revisited the search phrase, and said, “improving...maybe take out efficacy and put
in climate change” which resulted in the search phrase that netted several syntheses based on the
PowerPoint outlined above. That is, her visit to one site and then right back to the search box
indicated that looking at one result and determining it was not what she wanted was itself part of
the search phrase development. In this way, her total search phrase development time was over
two minutes, but the resultant phrase provided several generative syntheses based on the water
stewardship PowerPoint she found. Finally, to the extent that her successful search phrase
development was guided by the one result she browsed after the preliminary search, this form of
development is something that cannot be explained by keyword search literature grounded in
online catalogs of offline resources. This is because the time it would take to access each offline

97

resource would be excessive and essentially undermine the idea-play activity afforded by the
immediacy of the Web.
Conclusions - Case Study #1
Several key findings emerge from an analysis of Olivia’s overall synthesis experiences.
First, the complexity of synthesizing information in Web-mediated environment is
demonstrated across the examples and elements. Her interactions showed how keyword
search phrases can be constructed in a variety of ways: (1) through a combination of Web text,
task instructions and background knowledge; (2) from interactions between notes and Web text;
and, (3) based on in-the-moment insights. In a synthesis for meaning, arguably the simplest
element of synthesis proposed, the data demonstrated how Olivia’s background knowledge about
the text structure, ability to evaluate the relevancy of an article, text skimming proficiency, and
ability to separate relevant from irrelevant information within a relevant text interacted as
precursors to synthesis for meaning events. The data also showed how a variety of conditions
may lead to in-the-moment insights, including the rapid succession through a multiplicity and
density of ideas in multiple texts, attention to detail and deeper reading of specific text,
background knowledge, and idea-play, across a myriad of text structures and types, including
both hard and soft resources. Repurposing manifested several ways, from one context to
another, using ideas in opposite ways, reorienting ideas based on scale, repurposing the same
idea in two different ways, and combining a repurposed ideas with an insight; from task specific,
divergent, hard, and soft resources. Examples of reinforcement indicated how information
resulting from synthesis for meaning and insights were reinforced, often multiple times across
multiple texts. The data also indicated how in a single text, several of these elements could be
utilized in rapid sequence.

98

The concept of creative synthesis was explored, providing even greater evidence of the
complexity and layering of synthesis elements in a Web-mediated environment. The
implications for “downstream” syntheses of ones own previous syntheses were explored. The
various elements of synthesis that can appear elsewhere in insolation can also interact, work in
concert, and comprise a larger unit of knowledge construction that makes up a creative synthesis.
For example, one sequence from the reading-to-do session demonstrated how a series of
generative syntheses (repurposing, repurposing, insight, repurposing, insight, insight,
repurposing, insight, insight), scaffolded by the Web, but also informed by Olivia’s syntheses
from the reading-to-learn session, comprised Olivia’s foundational creative synthesis during her
reading-to-do task. This example also indicated that flow (Csikszentmihalyi, 1990) is a
consideration when examining the process of creative synthesis. In another creative synthesis
example, the simultaneous occurrence of synthesis elements working in concert was explored;
for instance, how the same text and ideas lead to insight, repurposing, and reinforcement.
Together, these illustrations indicate one thing about synthesis in a Web-mediated environment
quite clearly - it is anything but predictable or regular in its manifestation. In this way, the
notion of Web-mediated synthesis is itself an ill-structured concept.
Second, Table 7 indicates the average times each element occurred per minute during
each task. Olivia synthesized for meaning .42 times per minute during her reading-to-learn
session and .33 times per minutes during her reading-to-do. She had insights .10 and .18 times
per minute, respectively; repurposed .13 and .33 times per minute, respectively; and, reinforced
at the same rate during both. These data are not surprising in that the reading-to-do task was
essentially a creative task, where as the reading-to-learn task was exploratory. That is, synthesis
for meaning helped Olivia make sense of the text(s) she encountered in an exploratory task,

99

but when more generative reading was required, she relied more heavily on insights and
repurposing. She also relied more heavily on repurposing during the reading-to-do task than
insights, which may be an indication of the relative difficulty of having an insight when
compared to repurposing other information, and the significant value of repurposing to a creative
synthesis.
Table 8
Descriptive Data for Olivia
Synthesis elements per Minute
Reading-to-Learn
Synthesis for
Meaning
In the Moment
Insights
Repurposing
Reinforcement

Reading-to-Do

0.42

0.33

0.1

0.18

0.13

0.33

0.2

0.2

Third, soft resources were a valuable part of both Olivia’s reading-to-learn and
reading-to-do syntheses. As shown in Table 5, Olivia used more than twice as many soft
resources (16) as hard (7) during her reading-to-learn task, and about the same number of each
during the reading-to-do task (three to four, respectively). These soft resources were used in a
variety of ways. A blog named Old man in a cave helped her synthesize the importance of
media for meaning. A Reddit comments page provided her with ideas about inconvenience, sum
of behavior, money, collective vs. individual behavior, how people are affected personally, and
moral behavior. These ideas facilitated several synthesis activities, including synthesis for
meaning, repurposing, and reinforcement. An editorial on the HuffingtonPost helped Olivia
synthesize ideas about how humans deal with threat from an evolutionary perspective, a new
perspective on money, human relationships to nature, our ability to make change alone, and the
100

implications of time; again with a variety of synthesis techniques. The idea that religion could
make individuals more invested in nature came from repurposing a commentary on
WorldNetDaily, a site often considered a right-wing media outlet. Similarly, an insight about the
relationship between opposition to environmental protection and some individuals desire for less
government was spurred by a commentary on an equally biased, but left-learning site, Grist.org.
Finally, Olivia repurposed multiple ideas from a PowerPoint hosted at the Empowerment
Institute, a consulting and training organization.
When asked about these soft resources, Olivia said they were “extremely” helpful, further
commenting,
I really feel that even though they are softer, in the sense that I wouldn’t really
want to necessarily publish a paper and cite a blog, it gives me background
knowledge that I didn’t even know, and so I can begin to read credible sources on
this opposite to what I believe, I need to figure out what is opposite, I guess, basic
bullet points under this category, before I can begin to go look at more of the
credible resources...
With respect to the Reddit comments, she added
So the bulletin board, um, I felt like I could relate more to the people who are
writing. You know, it’s actually them, and they’re stating their opinion. It’s that
one opinion, and I know it’s that one opinion, but it’s interesting to then see two
people go back and forth, because that show me, yeah that’s right, yeah OK, I can
see that too, even though I don’t agree with it, I can totally see that...and, if it’s a
scientific article or the news, it's some sort of, they want you to think this way and
they write the article for that. But, the blog [discussion board], you can see the

101

back and forth, back and forth, and I really...that really helped me get some basic
ideas down
Her comments emphasize a fourth key idea about Web-mediated synthesis that is
highlighted by many of the examples outlined above: the topics and the task contribute
greatly to the types of Web resources that will be useful addressing them. In this case, given
the ill-structured nature of the topic and the exploratory and creative focus in the tasks, Olivia
was interested in ideas. She found ideas that she could borrow, repurpose, and that gave her the
foundation to make her own insights. These kinds of ideas can come from anywhere, and
sometimes the best came from sources that might not be considered reliable based on library
research rubrics (e.g., “Evaluating”) or Web reading guidelines (e.g., “TICA”).
Fifth, the relationship between keyword search phrases and the synthesis elements
that followed soon thereafter was inconsistent. Table 9 demonstrates the variability in the
types of syntheses the resulted from consistent and divergent keyword search phrases that
themselves preceded multiple instance of synthesis. No clear relationship emerged between the
type of keyword search phrase and resulting types of synthesis. However, given that the phrases
that lead to the most generative search elements were both consistent and took place during the
reading-to-do task, in may be that the task has more to do with the emergence of generative
syntheses than the type of keyword search phrase.
Table 9
Selected Keyword Search Phrases
Olivia

Keyword Search Phrase
why people change behavior on
climate change

Type Session Generative Syntheses
Syntheses
for
Meaning

102

C

RTL

8

6

Table 9 (cont’d)
why not change my behavior on
climate change
religious views on climate
change
benefits of climate change
rich people change behavior
global warming
how to change a behavior
voluntarily & how to change climate
change behavior voluntarily
improving of voluntary behavior
climate change

C

RTL

4

8

D
D

RTL
RTL

5
2

3
2

C

RTD

9

2

C

RTD

12

3

D

RTD

5

0

Sixth, synthesis for meaning provided additional value to Olivia beyond just helping her
understand text(s). As ideas were synthesized for meaning, they also contributed to the ideaplay that undergirded insights and creative syntheses. For example, as discussed above,
Olivia synthesized the idea of reducing energy and gas use in the context of ways to change
behavior for meaning from an abstract in the Journal of Preventative Medicine. However, by
repeating the phrases she had synthesized for meaning when considering several different lenses
of behavior, these simple ideas contributed to several instances of insight, reinforcement, and
repurposing thereafter.
Seventh, Olivia visited the task and supporting instructions extensively while both
reading-to-learn and reading-to-do, often directly supporting her synthesis experiences.
During the reading-to-learn task, and while reading about How the science of behavior change
can help with sustainability, she looked back to her task instructions several times which helped
her realize “I’m not interested in the theory of behavioral change, necessarily.” This sort of
filtering relevant from irrelevant is critical to synthesis for meaning. At about that same time, rereading the task out loud directly preceded her repurposing the idea of caring from sustainability
to climate change. About five minutes later, she re-read the instructions again in the midst of
103

starting on a 48 page PDF, after which she was again able to separate relevant from irrelevant,
this time within a specific document. About eleven minutes later, re-reading the task and
emphasizing to herself the instructions to “discover what you don’t already know” she said,
“well I know a lot of reasons why people make decisions to change behavior in many ways, but
why people may not...” which contributed to the development of her next search phrase. Another
15 minutes later, she re-read the entire task, soon thereafter she read about the science of climate
change, and said to herself “when I just read about the science...part of this topic (task) says to
open up to others...I wonder if religion affects your view on climate change.” This quick change
in direction was directly facilitated by the task instruction, lead to her divergent searching and
several related syntheses. Similarly, just a few minutes later, she read the task one last time,
specifically the instructions to “not close down and be open to new ideas.” After reviewing the
current results from religions, she switched abruptly again in her direction and searched for
[different country perspectives on climate change].
In her reading-to-do task, about 15 minutes after starting, Olivia read from the task that
she should be focusing on “creative ideas how to persuade. ” She then commented on the
“actions we can take,” reviewed her notes, and had an insight about projects for college students
to do. Just thereafter, she reviewed her notes and said “this thing [the task instructions] says I
need to have my ideas backed up, so I’m going to consult the literature on how to change a
behavior voluntarily, which guided her next search phrase. Later, she read again from the task
“persuade three different groups...lessen the effects of climate change...college
students...socioeconomic” and immediately said “so I’m going to type in ‘fun project for climate
change.’” Only a few minutes later, she read the task again, “...creative ideas to persuade three

104

different groups, so show them what it would be like, go back and go through my list...” which
led to the following insight
Well…we can I guess we can make for college students we could give them an
example of what the world would be like in so many years…make them…have
them enter this new world…a day…and hour in the life of someone so many
years from now…from global warming…this would also be fun, too, for
people…like a church group could do this…everybody could do that together…
Olivia summed up her use of the task instructions in her interview by saying “I just
needed to continuously go back and clarify, because for me it’s easy sometimes to get
distracted.” However, the examples above indicate that the task served several more purposes in
the Web-mediated synthesis environment. That is, re-visiting the task helped her change course
and construct new search phrases five times, three of which were divergent. Re-reading the task
instructions also directly preceded four of her synthesis moments.
Eighth, Olivia used many more Web resources during the reading-to-learn task than
the reading-to-do task (23 to seven, respectively), which she described as task related. After
her reading-to-learn task, during an interview, Olivia mentioned several times that her strategy
was to search for opinions and information, “I was searching...I had kind of like a preconceived,
I guess, I tried to get rid of that, but I kind of knew the type of information that I wanted.” That
is, her mindset was to search, and Google was used as a search tool. Where as, in the second
session
I just sat and I thought about my ideas first..before I went to Google. Like what
do I know, what are the ways that I know that could achieve this goal…and then I
used Google to basically verify those ideas and to gain more ideas for each

105

separate group. But then, as I was going through the groups, I realized the
fundamental thing between all of them is this fundamental behavior change. And,
then I wanted to get some background information on how to effectively change
behavior. So I went back and I did that…in Google, and then I came up with
some more ideas...
That is, when the task was more creative, her mindset was more about thinking, and Google
became part of her creative process as more of a thinking tool.
Ninth, Olivia’s examples demonstrate how both background knowledge and idea-play
contributed to her synthesis activity. Background knowledge both contributed to and fully
comprised synthesis elements related to keywords, synthesis for meaning, insights, repurposing,
and creative syntheses. At the same time, several examples demonstrated how assigning
background knowledge as the source for a several ideas that emerged would be overly reductive.
The multiplicity Web text(s) and use of idea-play explained or partially explained several
instance of synthesis that may have otherwise been contributed to background knowledge alone.
Similarly, idea-play interacted with keyword construction, insights, repurposing, and creative
syntheses. There are also times that both background knowledge and idea play worked in
concert with synthesis elements. However, while these examples provide ample demonstration
of the importance of both background knowledge an idea-play, the extent to which either
contributed to any given synthesis element is unclear from Olivia’s examples.
Finally, Olivia’s use of notes was extensive and supported all of her synthesis
processes. Synthesis for meaning was supported by both verbatim and summary notes in almost
every case. Insights were prompted by reviews of her notes in several instances. Review of her
notes even informed the development of her keyword search phrases at times. Adding ideas to

106

her notes (either for meaning for in a generative way) also facilitated the repurposing of that
same idea to another section of the notes. For instance, as she synthesized morals as a reason
people might change their behavior for meaning, she also realized that the same idea might apply
to why people don’t change their behavior, and added a note repurposing the idea of morals to
that section. During the reading-to-do task, her reading-to-learn notes served as a new
standalone text from which Olivia was able to synthesize several ideas for meaning into her
reading-to-do notes. The discussions of the creative syntheses in her reading-to-do sessions
above outline several of such instances. In addition, just stopping to synthesize ideas for meaning
into her notes sometimes led to other insights. For instance, as presented above, she synthesized
several ideas that might affect individual behavior from the Purdue study. In the middle of
writing them down, she had an insight about money. Similarly, after making note of a different
insight about how money might be a reason people do not change their behavior; she
immediately had another insight about public perception. It is possible, therefore, that the act of
stopping to take notes provides either idea-play among the ideas in the notes, or a small
opportunity for internal incubation (i.e., a letting up of her focus on the Web text(s) allowing for
subconscious idea-play), both of which may facilitate generative syntheses.
Chapter Summary
This chapter provided detailed illustrations and discussion of the broad themes about
synthesis elements outlined in Chapter 3, based on an in-depth case study of one individual
participant, Olivia. The chapter explored each element individually and then demonstrated how
they worked in concert to comprise creative syntheses. The chapter finished with conclusions
drawn from this case study about broader patterns of use, interactions, and relationships among
the synthesis elements that directly address the research questions for this study.

107

Conclusions that contributed to a better understanding of RQs 1 and 2 included (1) Webmediated knowledge synthesis is complex, multiplicitous, unpredictable, and ill-structured; (2)
synthesis for meaning was used more during the exploratory task, while insights and repurposing
were used more during the creative task; (3) soft resources were a valuable part of both Olivia’s
reading-to-learn and reading-to-do syntheses; (4) the topics and the task contributed greatly to
the types of Web resources that Olivia used to address them; and, (5) Olivia utilized the task and
supporting instructions extensively while both reading-to-learn and reading-to-do, often directly
supporting her synthesis experiences.
Conclusions that contributed to a better understanding of RQ 3 included (6) the
relationship between keyword search phrases and the synthesis elements that followed soon
thereafter was inconsistent; (7) in additional to aiding understanding of text, synthesis for
meaning also contributed to the idea-play that undergirded insights and creative syntheses; and,
(8) Olivia used many more Web resources during the reading-to-learn task than the reading-to-do
task, based on the type of task.
Conclusions that contributed to a better understanding of RQ 4 included (9) background
knowledge and idea-play contributed to all of the other synthesis elements, though the extent to
which either contributes is unclear.
Conclusions that contributed to a better understanding of RQ 5 included (10) Olivia’s use
of notes was extensive and directly supported her synthesis processes a majority of the time.

108

CHAPTER 5
IN-DEPTH CASE STUDY #2 - BETH
Chapter Introduction
This chapter provides detailed illustrations and discussion of the broad themes about
synthesis elements outlined in Chapter 3, based on an in-depth case study of one individual
participant, Beth. The chapter explores each synthesis element individually and then
demonstrates how they worked in concert to comprise creative syntheses, primarily by
comparing and contrasting these findings with the data provided in Chapter 4 regarding case
study #1. In so doing, this chapter highlights further nuances in the application of the broad
themes of Chapter 3. It finishes with conclusions drawn from the case study about specific
patterns of use, interactions, and relationships among the synthesis elements that are unique to
Beth.
The chapter focuses on how synthesis for meaning, insight, repurposing, reinforcement,
and creative synthesis manifested for Beth by providing detailed examples and illustrations.
Specific to RQs 1 and 2, variation in the application of these themes is demonstrated both across
specific examples in this case study (e.g., by highlighting the uniqueness of the various ways
Beth synthesized for meaning) and by highlighting differences between the two cases studies
(i.e., between Beth and Olivia). The chapter also emphasizes the multiplicity of these ideas and
how they interact and work in concert, the primary concern of RQ 3. The specific roles of
background knowledge and idea-play are highlighted in several examples with respect to RQ 4.
And finally, to explore RQ 5, most examples emphasize the substantive contributions note-taking
provides to the participants’ synthesis processes.

109

Beth
The second participant selected for an in-depth case study was Beth, a 23-year-old
economics doctoral student. Her selection was also primarily based on her extensive
demonstration of synthesis activity in both reading-to-learn and reading to do tasks (she had the
second highest total after Olivia, 72 instances). Like Olivia, she also demonstrated a diversity of
synthesis experiences and the ability to think-aloud in great detail. However, three additional
factors contributed selecting Beth for the second in-depth case study, all related to the initial goal
of diversity in participant selection. First, her background in economics was far enough afield
from Olivia’s in food toxicology to provide a significantly different perspective and learning
background (e.g., hard science vs. social science). Second, her reading style for the Web-based
texts generally involved less deep reading than Olivia. Finally, she took all her notes online
(unlike Olivia who took notes offline). A detailed profile of Beth is provided in Appendix K
including a reporting of her background knowledge and experiences on the topic of climate
change collected at the beginning of her research session.
Beth demonstrated 72 instances of synthesis during her sessions. Table 10 outlines
descriptive statistics gathered.

Table 10
Descriptive Data for Beth
By Task
Reading-to-Learn

Reading-to-Do

Totals

Consistent
Keyword Search
Phrases
Divergent Keyword
Search Phrases

2

1

3

7

6

13

Hard Web Site

6

3

9

110

Table 10 (cont’d)
Soft Web Sites
Synthesis for
Meaning
In the Moment
Insights
Repurposing
Reinforcement
Creative Synthesis

20

9

29

7

10

17

12

12

24

4

7

14

12

2

14

0

3

3

In addition to these statistics, several other characteristics describe her session. First, she
took a lot of notes, all on the computer in Google Docs. Second, she revisited and referenced the
task instructions multiple times during each session. Third, a vast majority of the resources she
read were short articles on soft Web sites. Within these texts, she did not deep read much of the
text, mostly skimming and skipping around. Finally, in both sessions, Beth “slowed down” a bit
toward the end of the allotted time. In both cases she asked when the session would be over, and
the number of synthesis experiences declined toward the end of the task.
The data collected during Beth’s sessions provide further illustrations of the various
synthesis elements. Similar to Olivia, several of these are explored in detail below, and
conclusions about the patterns of use, interactions and relationships are discussed.
Divergent keyword search phrases. Beth’s use of divergent keyword search phrases
was relatively balanced between the reading-to-learn and reading-to-do tasks. Table 11 lists all
of the phrases used in the first session.

111

Table 11
Keyword Search Phrases
Beth, Reading-to-Learn
Time

Keyword Search Phrase

Type

1:20

green living style (tips)

D

BG knowledge +
Google

2:59

why recycle

D

Web text

4:30

D

Web text

6:29

how to lower energy consumption
how to lessen the effects of climate
change

C

task

14:00

why not go green

D

BG knowledge + task

25:11

reasons to go green

D

iteration

35:18

why I don’t recycle

D

iteration

42:50

can I lessen climate change

C

iteration

54:06

why I won’t go green

D

iteration

Several key ideas emerged from this data. First, Beth’s keyword development was grounded by
a single keyword - “green.” It was the first phrase she used [green living tips], and resurfaced
three more times during the session. Before using it the first time, she was reading the task
instructions, and finished with “explore what you can discover that you don’t already know.”
After which she commented
all right, I’m aware of a lot of things, but I don’t know a lot of things, if that
makes sense. It’s like, I kind of know that people drive hybrid cars and use cloth
diapers and stuff, but that’s...I don’t know. All right, so let’s get going...
She navigated to Google and typed in [green living style], as she said “green living style...tips,
tips, I like tips better, thank you Google.” She switched to “tips” based on the recommended
112

search phrases that Google presented below the search box as she typed the words “green
living.”
In this case, both Beth’s background knowledge and Google informed the final search
phrase she selected. Her interview comments provided an indication of the origins of her
“green” orientation to the task. When asked what guided her searching throughout the readingto-learn session, she replied
it was pretty much what I underlined on the paper [task instructions], which is
what I thought were the main topics...and basically, why do people...the words I
underlined were why behavior lessen the effects of climate change and not change
behavior...are the exact words I underlined
She continued “yeah, so that was what kind of guided my search was the word like behavior,
green, why to go green, why not to go green.” When asked where green came from, she said
I think it’s because I associate being green with like environmentally friendly
decision making, people call it like the green movement, or people buy green
products...I also had a friend who worked for, I don’t remember exactly who, but
like the green movement in Washington, D.C. and she helped organize the green
festival out in California...so I think that’s why I associate green with
environmentally friendly. She is a Prius driving person...I didn’t even notice it
wasn’t on there [on the task instructions]. I swear it was on there.
As such, the influence of her background knowledge and experience was significant, enough so
that she assumed her “green” lens on climate change was in the instructions.
•Equally important to consider was the impact of Google’s suggestion to her. That she was
open to the recommendation from Google and switched to it without hesitation is indicative of

113

the confidence users may place in their technology. Though it cannot be determined whether
[green living style], Beth’s first choice at this moment in the session, would have resulted in
better resources (i.e., she only spent less than two minutes on the [green living tips] results), it is
important to consider the implication of using Google recommended keyword phrases. If, as
discussed above, keyword selection is an important part of the overall synthesis of ideas for a
creative task, it is arguable that using too many Google recommended search phrases would
effectively “close down” the breadth of the learner's exploration since the Google recommended
phrases are likely based on popular combinations of words based on the first few words entered
by the learner.
The second characteristic of Beth’s keyword searches was their commonality across her
reading-to-learn task. Though she used nine different search phrases, they represented only
four different general areas of search: (1) being green (or not); (2) recycling; (3) lessening
climate change; and (4) lowering energy consumption. Compare this to Olivia’s use of keyword
search phrases (Table 6). Though Olivia only used three more phrases (12) than Beth, topically
she explored many more areas: (1) opinions on climate change; (2) climate change behavior; (3)
barriers; (4) caring about climate change; (5) benefits of climate change; (6) mitigation; (7)
religious views; and (8) different countries. This may be related to Beth’s waning motivation
toward the end of her reading-to-learn session. That is, during the 60 minutes task she said “all
of these Web sites say the same thing” at 47:55, said she felt “stuck in a rut” at 53:30, asked how
much longer the session had at 53:42, and seemed to be out of ideas for keyword search phrases
when she used [why I won’t go green] at 54:06 and said “Have I already done that? I might have
already done that...” Beth also demonstrated only five synthesis moment in her final 20 minutes
of the reading-to-learn task, using two search phrases that were only mild changes to phrases

114

used earlier in the session ([how to lessen the effect of climate change] was used at 6:29 and [can
I lessen climate change] at 42:50; [why not go green] was used at 14:00 and [why I won’t go
green] at 54:06). Whereas, Olivia demonstrated 13 synthesis moment in the final 20 minutes of
her reading-to-learn task, using four largely different phrases that did not have similar
antecedents ([why climate change mitigation beneficial to government], [benefits of climate
change], [religious views on climate change], and [different countries perspectives on climate
change]).
Together, these ideas illustrate the potential pitfalls of ignoring Web text(s) when
developing keywords. Beth lifted only the ideas of recycle and energy consumption from the
text to expand her exploration, even though in the interview, she commented that she commonly
does this. That is, when exploring ill-structured topics, the combination of task, background
knowledge, and Web text demonstrated by Olivia may lead to more effective development of
divergent keyword search phrases.
Finally, though both Olivia and Beth used several soft resources in their reading-to-learn
sessions, more than hard resources in both cases, the soft resources visited by Olivia were more
serious. While individual choice of the resources to visit is the primary factor in determining
what resources were visited, it is possible that the different keyword choices made contributed
to the quality of soft resources by impacting what was available. That is, while Olivia
generally used the term “climate change” to frame her searches, Beth used the variants on “going
green,” a more colloquial term. Of the soft resources visited by Olivia, most had more serious
titles like “global governmental organization,” “the ‘public opinion’ on climate change,”
“environmental economics,” “social behavior and public policy,” “behavioral economics,” “the
climate change religion,” “Christian right-wing views are swaying politicians and threatening the

115

environment,” and “how governments can creative incentives for mitigation.” Where as, the soft
resources visited by Beth included titles like “8 ways not to get tricked going green,” “how to
stop unwanted mail,” “Ultimate way to go green? Don’t have kids,” “composting worms,”
“Convince me to go green! Eco Friendly Motivation,” “reusable paper towels,” “the four-day
week,” and “create jobs, don’t go green.”
Table 12 lists all of the phrases Beth used in her reading-to-do task. When compared to
Olivia’s reading-to-do search phrases, the data provide further argument for the importance of
consistent and continued use of divergent search phrase. That is, Olivia used four topically
dissimilar keyword search phrases in the final 10 minutes of her reading-to-do session and
experienced five synthesis elements during that time, while the five search phrases Beth used
during the final 10 minutes covered only two general topics and she only had one synthesis
moment.
Table 12
Keyword Search Phrases
Beth, Reading-to-Do
Time
1:36

Keyword Search Phrase

Type

39:50

how to lessen effect of climate change
marketing green products to upper
class
how to convince other to make
environmentally friendly decisions
convince others to make green
decisions

40:44

convince friends green choice

D

iteration

41:44

inspiring lifestyle change in others

D

BG knowledge

44:00

posh green products

D

idea-play

10:23
35:09

116

C

task

D

task + BG knowledge

D

task + BG knowledge

D

task + BG knowledge

Synthesis for meaning. While Beth compiled and recombined information across texts
and from background knowledge across tasks, she did so less often than Olivia. Olivia
synthesized information for meaning 25 times during her reading-to-learn task; Beth did so only
seven times. Several issues may have contributed to this difference. First, as noted above, Beth
visited less varied resources. As she began to see the same information over and over, though it
was reinforced for her (see below), it did not required a new synthesis for meaning to be useful.
Second, as noted above, the resources Beth visited were also sometimes less serious in the
topics they discussed, and may not have provided information in substantive enough ways to
be worth synthesizing for meaning. Third, Beth made several more insights than Olivia during
the reading-to-learn task, indicating that the less serious resources visited may have been more
amenable to insight.
Of the times Beth synthesized information for meaning, she did so in its simplest form.
That is, while Olivia demonstrated several different ways to synthesize meaning (i.e., based on
her background knowledge, from a single Web resource, and/or from multiple Web resources),
in every case, Beth primarily copied short phrases from the text(s) when synthesizing for
meaning. Appendix L provides several examples of this.
In-the-moment insights. One of the things that differentiated Beth from Olivia was her
extensive use of insights during both the reading-to-learn and reading-to-do sessions. While
Olivia had six and eight insights during the first and second tasks, respectively, Beth relied more
heavily on this process, having 12 insights during her reading-to-learn task and 12 during her
reading-to-do sessions. Of these 24 experiences, she demonstrated insight under a variety of
contexts and conditions.

117

At one point, Beth had several insights quickly in a row while reading through a list of
ways for individuals to reduce climate change (see Appendix M, examples 1 and 2). Beth’s rapid
movement though this list and corresponding generative insights was similar insights Olivia
experienced as a results of the rapid succession of ideas, but differed in two important ways.
First, in this case the rapid sequence of ideas is from one Web site. That is, just this list of
ideas seemed to create an environment similar to that which promoted insight over the course of
several articles for Olivia. Second, while many of the ideas in Olivia’s case were substantive in
nature, the page Beth explored was very soft, even including what she considered some silly or
unrealistic examples. However, given the lack of traditional reliability or accuracy, the text on
this page still established something similar to the creative stew of ideas Olivia experienced. As
such, Beth had insights based directly on the text (text about paying bills online resulted in an
insight about people doing things that are easier; the vinegar text lead to an insight about
chemicals and safety) and based on background knowledge (change stuff too much and you can
feel good about it were both identified as specifically not from the Web, but a “personal opinion”
and “from my brain”) while reading the list. In the case of the third list item, Web text about
purchasing local food, two different insights were facilitated, the importance of higher quality
and the notion of feeling good. Finally, it is important to recognize that the articles Beth visited
in the first two examples in Appendix M were not about why people make the decision they do,
but actions people can take to reduce the effects of climate change. Part of the reason Beth was
able to (or had to) make insights about these resources was just that slight difference from the
task she was assigned. So, it is possible that a keyword search phrase that included keywords
like “choices” or “behavior” would not have required or facilitated as much insight on her part
when compared to resources that resulted from the more general [how to lessen the effects of

118

climate change] that led her to this site. Both examples of Beth using lists on the Web and the
resulting insights indicate that reading through bulleted material may provide smaller
segments of text amenable to idea-play. That is, the reading process of lists itself may again
allow for small breaks in thought, fostering internal incubation (subconscious idea-play), which
may facilitate generative syntheses.
In another case (third example in Appendix M), Beth made an insight that appeared to be
based on a VCS synthesis of ideas across several articles. The build up of ideas in this example
is different than the creative stew identified just above and in Olivia’s data. All of the ideas in the
texts Beth encountered had the notion of “tradeoffs” in common (though not explicitly stated),
where as the creative stew is a number of relatively unrelated ideas.
In the final example, Beth’s use of the text comprehension strategy of prediction
(Brown, Pressley, Van Meter & Schuder, 2004) provided a generative insight. And, like Olivia’s
experience with the Purdue research, her insight occurred after stepping away from the text to
take notes on a synthesis for meaning, which may have provided just enough letting-up of ideas
or incubation to promote insight.
Repurposing synthesis. Beth repurposed only four times during the reading-to-learn
tasks, but these instances reflected similar ways of doing so as Olivia. For instance, after
searching for [how to lower energy consumption] she repurposed ideas from one context to
another when reading about “13 Simple Ways to Lower Your Electric Bill.” As she quickly
scrolled through this article, she read
keep it clean...clean air filters...program the thermostat...seal up the house (saying
I should do this in my apartment, my heating bills are obscene)...um..let’s

119

see...avoid chores...limit the use of heat generating appliances such as the
oven...change light bulb to the compact fluorescent ones...
and then said, “another reason people might go green is to save money...like on heat bills or on
grocery bills...go back to gardening thing,” while she noted “save money (i.e. on heat bills or
grocery bills).” In this case, the idea of saving money was repurposed from the context of
lowering electric bills to climate change behavior. After returning to the same article, she read
further “consider time-of-use plans...fix the bill...” then returned to her notes and said, “so,
saving money is a reason...but then I guess it could kind of be a reason not to go green...because
things that are like organic and stuff are more expensive,” noting, “save money (organic more
expensive).”. In this latter case, Beth repurposed her own repurposing of saving money in a
way opposite to how she herself intended it.
Beth repurposed another idea in the opposite way it was used in text and reoriented it to
a new scale while reading the Washington Post article referenced above. After synthesizing
morals for meaning from the article, she returned to the text and quickly read, “he’s trying to say
that we should look at the history of civil rights...uuum...should they [carbon emissions] be
mandated by law...” and returned to her notes saying “another reason people choose not to go
green, they don’t have to, it’s not legislated,” as she noted the latter text. In this case, the idea of
mandating behavior by law was reversed from a way to promote behavior to a reason people do
not behave a certain way. It is also an example of “double” repurposing, given that Beth
reoriented the same idea to a different scale, that is, from carbon emissions at the
corporate/industrial level to individual behavioral mandates.
Reinforcement synthesis. Similar to Olivia, Beth encountered several opportunities for
reinforcement of both Web text(s) and her own ideas during the reading-to-learn task. However,

120

many of the reinforcing instances for Beth did very little to enhance her knowledge in generative
ways. Where as Olivia’s examples often provided new context, perspective, reliability, and
temporal stability, many of Beth’s opportunities for reinforcement were reduced to mere
repetition. Several of these examples were based on bullet points with limited text, such as that
provided in Figures 4 and 5 (See Appendix N for detailed examples). In the first example, Beth
read through this list, but did not read the supporting text. In the second example, she read
through a list of hyperlinked ideas in a similar fashion. Though in both cases supporting text did
provide new perspective on an issue she had read prior, her thin reading strategies did not allow
for this. Whereas, in the second example, her deeper reading of a text structure similar to that
provided in Figure 4, did lead to reinforcement. As such, both the depth of reading on the
part of the reader and the structure of the text may enhance or diminish the generative
potential of reinforcement opportunities.

121

Figure 4. Bulleted Paragraphs in Web Text. The text in this figure is not meant to be readable.
The figure is simply meant to demonstrate the structure of the text.

122

Figure 5. Bulleted Lists in Web Text. The text in this figure is not meant to be readable. The
figure is simply meant to demonstrate the bulleted structure of the text.
Beth also demonstrated how reinforcement is not always directly from the Web text but
can be more circuitous. For instance, she skimmed some comments from an article about “5
reasons why people don’t recycle,”
Ah, here we go...reasons recycling is bad for the environment...we are not running
out of landfill space...recycling actually harms the amount of trees we have..it’s
not good for the environment because there is a lot of pollution to recycle.
She then said/wrote in her notes “reasons people aren’t going green...technology is not perfect
yet, i.e. it costs more to recycle than to not.” In this case, the text itself did not reinforce
anything, but it helped her think about technology again (i.e., reminded her of something she
read earlier), which reinforced the idea that she noted earlier about the technology to go green
not being ready yet. Reinforcing that idea here lead to an elaboration about recycling costs, and
this elaboration was based on the text she had just read.

123

Finally, as mentioned above, Beth’s use of keyword search phrases did not provide as
many paths to new information as did Olivia’s. One direct result of this was that near the end of
her reading-to-learn task, when faced with information that might serve to reinforce earlier
text(s) or insights, Beth’s reaction was exasperation. As she pulled up a Web site entitled
“Reduce Global Warming - Slow Down Climate Change with These Easy Green Tips!” she read
“eat less meat” and then lamented “all of these Web sites say the same things” while laughing.
In this way, the relationships between keyword development and the potential to experience
generative reinforcement was highlighted. That is, the tendency for many of Beth’s resources to
have the same information may indicate the potential for decreasing returns on the utility of
reinforcement after a certain point.
Creative synthesis. As with Olivia, the illustrations used above for Beth were culled
from just her reading-to-learn task. As such, multiples of synthesis elements are now examined
in the context of the creative syntheses they comprised during her reading-to-do task. Two
examples of creative synthesis are detailed in Appendices O and P (creative synthesis #1 and #2,
respectively). Several important ideas about creative synthesis are demonstrated in these
examples.
The activity in Beth’s first creative synthesis (Appendix O) comprised the first creative
synthesis during Beth’s reading-to-do task, and was quite different from Olivia’s. While Olivia
synthesized ideas from her first session for meaning, repurposed ideas from the first session, and
had insights during her initial creative synthesis, she did so by working only from her notes. She
did not use the computer at all in doing so. Where as, the activity from Beth’s initial creative
synthesis integrated multiple sources: (1) her reading-to-do notes; (2) her reading-to-learn
notes; (3) the task; (4) the browser history text; (5) four articles from session one; and, (6)

124

background knowledge. Together, these resulted in ideas synthesized for meaning, insights, and
repurposing.
For instance, the very first idea Beth proposed “it doesn’t have to be expensive” was
based on three separate ideas from the first session, “it’s good for you” (a rephrasing of “you can
feel good about it”), “saving money,” and “accidentally doing it.” In this way, she synthesized
several ideas together for meaning from memory. That is, even though both “feeling good
about it” and “accidentally going green” were insights in her first session, they were synthesized
for meaning at this point, given that taken together with the idea of saving money, their
combined meaning comprised “it doesn’t have to be expensive” without further value-added.
The next idea that she discussed was “God’s kingdom” representing her background knowledge
about religious groups and soon thereafter, the idea of “future generations” was synthesized for
meaning from notes. (As a point of clarification, her reading-to-learn notes now represent a
new text resource available in the universe of texts on the Web.)
Just considering future generations prompted Beth’s memory of altruism in one of the
reading-to-learn articles, which, even though she did not take notes on altruism, was available for
her to access. The availability of the idea of altruism as a consideration and her ability to access
the details of it quickly during the second task is indicative the complex environment for
synthesis that ubiquitous Web access facilitates. You only need to remember core ideas and key
markers on the Web. That is, Beth needed only a basic familiarity with the content she read and
then the Web was available as a near-instant adjunct to her long-term memory. In Beth’s case,
she used the browser history to help her find this article, but, if she had Googled the phrase [US
news world report altruism green] (i.e., all she really needed to remember was the publication,
and two keywords), it would have likely been immediately available, as well (in May 2012, more

125

than one year after her research session, that search still listed the target article second in Google
results). As this sequence of events related to her synthesis of information, though Beth could
have just added altruism to her notes without revisiting the original article, actually finding it
again accomplished two things. First, it allowed her to check the article for any explicit
connections between religion and altruism (as she searched the page text for “relig” and
“Christian”). Second, it facilitated elaborative and evaluative comments on her part before
synthesizing for meaning the concept of altruism, as she wondered if it was too much of a
generalization to assume religious people are altruistic before eventually synthesizing the idea
into her notes.
Thereafter, returning to her notes prompted several more synthesis moments. While
reviewing what she had already written, she had an insight about Christianity vs. religion and
how the Bible fits (or doesn’t) with this task. Given the lack of any supporting Web text(s) on
this topic, her ideas here were largely based on background knowledge. She synthesized future
generations again for meaning in the college student section. She had an insight about making a
difference, which, though could be entirely based on background knowledge, may well have
been informed by several of the ideas synthesized in the reading-to-learn task. That is, given that
she just used the idea of future generations again, it, along with the idea of a moral argument that
“we have the power to prevent death/poverty,” both from the reading-to-learn task, may have
contributed to the development of “making a difference.” Beth was clear where “could be this
generation’s social movement” came from, which she synthesized for meaning from a
Washington Post discussion about the parallels between civil rights and climate change action.
At this point, she still was starting to lose some of the momentum from this particular
creative synthesis. She had been trying to find the Washington Post article, but when she did, it

126

replaced her notes in the browser tab. Once she got her notes back, the article was lost, and she
kept trying to find it again in another tab. However, the sequence of reading some text from
another article about peer pressure and greenwashing, re-reading part of the task, re-reading her
reading-to-do notes, and then returning to her reading-to-learn notes prompted two final insights
about trendsetting and responsibility. It would be easy to just assume both of these ideas were a
result of background knowledge alone. However, given the multiplicity of resources and quick
switching that occur adjust prior to having these insights, it is arguable that attesting these
insights to background knowledge alone would be overly reductive. Beth read information from
four different sources in the 40 seconds prior to the insights. As such, in this case, Web text, task
instructions, current notes, session one notes, and background knowledge appeared to work in
concert to facilitate the two final insights.
Beth’s second creative synthesis (see Appendix P for details) is unique from those
already discussed in that it was completed entirely without Beth ever doing a Google search or
visiting a Web text directly. And, though it might be easy to attribute all of her insights during
this time to background knowledge, close analysis of the interactions provides a more complex
scenario. Indeed, Beth’s vision of a product for religious people is largely based on ideas that did
not appear in Web text(s), and were either new insights (the aesthetic of a beautiful landscape),
or synthesizing for meaning her own previous insights that were originally based on background
knowledge (e.g., God’s planet, the bible verses from an insight earlier in this session and “feeling
good about yourself” from an insight in the reading-to-learn task). But, a close examination of
the other activity indicates how much she synthesized information from her reading-to-learn
notes, which at this point represented its own Web text of her construction, based in large part on
original Web text(s) and insights/repurposing she contributed based on these Web text(s). Her

127

vision for college students repurposed “everyone is doing it” (originally synthesized for meaning
from Web text in the reading-to-learn task) in the opposite way - “personal decision” for college
students. The economic aspects mentioned here were synthesized for meaning and reinforced
from Web text(s) in the reading-to-learn task.
Finally, Beth’s vision for high SES individuals in this example was equal parts
background knowledge and ideas based on Web text(s). The poverty issue was synthesized
for meaning from the reading-to-learn session. However, her insights about the importance of
charity and visibility derive, by her own admission, from her biases (background knowledge).
Similarly, the idea of emphasizing small changes can be traced to an insight in the reading-tolearn task that was based on background knowledge. But, not moving into a smaller house was
synthesized for meaning from a Web text in the first session. And, though the specific idea of
health benefits had not appeared to this point anywhere in Web text(s) to this point, she did read
about walking and biking instead of driving, had an insight about the higher quality of local
foods, and discussed the higher prices or organic foods in session one.

A such, the idea of

health benefits could be based on her own background knowledge about health and climate
change, but it may well be an insight based on her previous interactions in the reading-to-learn
task, or a combination of the two.
Next, Beth read her session one notes with the explicit reason to “counteract” the reasons
people don’t go green. That is, when she read from her notes (now representing its own Web
text), she repurposed the idea of “changing stuff too much” in the opposite way to emphasize
small changes to the high SES groups. She then had an insight about the lack of certainty about
the tradeoffs that using data would be helpful for the high SES group. Soon thereafter, when
reviewing her session two notes, she synthesized turning off lights, insulation, and eating less

128

meat for meaning, which were originally synthesized for meaning in the first session. And,
finally, biking and walking was repurposed from the context of transportation in her reading-tolearn task to health benefits in the reading-to-do task.

Conclusions - Case Study #2.
Beth’s synthesis experience was for the most part “thinner” than Olivia’s. That is, she
visited resources with less text, didn’t read them as closely, used more colloquial keyword search
phrases with less topical breadth, and the notes she created were less detailed. At the same time,
while Olivia experienced several synthesis moments that were “multiple” in the sense that they
reflected two or more elements of synthesis (especially during her reading-to-do creative
syntheses), Beth’s were largely independent of each other. However, Beth still produced almost
50 interactions of generative synthesis. In fact, Beth experienced several more in-the-moment
insights than Olivia (24-14). And, though Beth’s insights were often based on bulleted or listed
Web text and Olivia's were often based a paragraph or more of text from longer and richer
sources, Beth’s insights were no less meaningful. For instance, while Olivia made an insight
about how religion can be a way for people to become more passionate about climate change
based on the following, rather complex, text from a long research based PDF …and greens with a spiritual bent, confer in increasing numbers, in particular
over climate change, acquaintances are being struck that transcend many ethnic,
ideological, a theological obstacles….And such encounters are not just a
curiosity. The might make all the difference to the planet’s future…
- Olivia made an insight about how technology to go green might not be ready based on a blog
rant about how the author blew out a “10 year light bulb” in five minutes. As such, the data from

129

these case studies provide evidence that synthesis can be facilitated in both complex and
simple online reading environments, from both hard and soft resources.
Similar to Olivia, Beth also made extensive use of soft resources. She used 20 soft
resources (as compared to six hard resources) in the reading-to-learn task, and nine (as compared
to three) in the reading-to-do task. Like Olivia, Beth used these resources to synthesize for
meaning, spur insight, and repurpose information. When asked about her use of these soft
resources, Beth commented,
I think it depends on what I’m doing, if I’m writing a paper, I don’t do that...but,
if I’m just trying to get a feel for the various opinions out there...and just trying to
like, even if I’m writing a research paper, a lot of times depending on the topic,
I’ll start with the Washington Post and BBC and things like that or even blogs,
cause its just, to me, it helps give me context...like what are the opinions on this
issue, what is the space I’m trying to navigate...
Beth’s experiences also indicate that different soft resources may be more amenable to
certain forms of synthesis. For example, when reading soft resources with less serious content,
Beth had more insights than she synthesized for meaning. As such, Beth’s experience
corroborated a possible interaction between task, resources, and synthesis type for Webmediated knowledge synthesis.
Beth’s case study provided several examples where insight occurred in the context of
idea-play resulting from bulleted lists. There were also times in both Beth and Olivia’s sessions
where they “let-up” from the text to revisit or take notes and insights occurred while doing so.
Both of these situations indicate how important the role of the creative process might be
when considering Web-mediated synthesis. That is, both idea-play and letting-up (or

130

incubation) are central to theories of creativity (see Bruner, 1962; Csikszentmihalyi, 1997;
Osborn, 1953; Sternberg & Lubart, 1999; Wallas, 1926); Ward, Smith, & Finke, 1999) and
should be considered when further exploring insight as part of the process of Web-mediated
synthesis.
Beth and Olivia also differed greatly in their tendency to repurpose (14 to 23,
respectively), especially in the reading-to-do task. This highlights the role of Web text and
searching in the reading-to-do task. Neither participant used the Web extensively in this task,
however, all of the sites Olivia visited were new, while Beth went back to several sites she had
visited in the reading-to-learn task. Olivia used specific keywords ([rich people change behavior
global warming]) that guided her to texts with enough similarity to her task that repurposing was
afforded (e.g., carbon taxes for industrial customers repurposed to tax breaks for private
individuals). However, when using the Web text(s) from her reading-to-learn task during the
reading-to-do task, Beth either visited them to look for a specific idea (e.g., remembering the
specific idea of altruism led Beth to revisit the article that originally presented that idea and
synthesize it for meaning), or revisited the article because the title appeared in her history list
while looking for something else (e.g., when she was looking for information on social
movements and pulled up one on peer pressure from her history, which itself led to an insight
about first adopters and trendsetters). In this way, keyword development may prefigure the
cognitive lens through which readers view their text(s) and the types of synthesis that
result. That is, Olivia repurposed her text to fit the constraints of her keyword search phrase.
Where as, Beth’s specificity of intent at times lead to a synthesis for meaning, and her lack of
focus at others (e.g. when selecting the peer pressure article essentially because it was there),
facilitated insight.

131

Though some of the differences above were rather stark, Olivia and Beth were quite
similar in other areas. First and foremost, when faced with resources of extended text, both were
very capable at combining the reading comprehension skills of skimming and closer
reading, and switching between them while reading Web text. Both demonstrated several
times where she would read several section headers or the first sentence or a short phrase from
several paragraphs before finding something particularly interesting and focusing in on that.
After reading more closely (and/or re-reading), and often having a synthesis experience, they
would continue skimming until another closer read or the end of their article (or the end of their
interest). (It should be noted that, in general, Olivia read more deeply that Beth, but both
demonstrated this strategy of skimming to closer reading). This ability to vacillate between
skimming and closer reading is different than that documented for skilled K-12 online readers.
That is, the Taxonomy (n.d.) indicates how skilled K-12 online readers skim prior to a deep
reading of an article (i.e., a text walk prior to a close reading (Taxonomy, n.d.)), but does not
make reference to changing between these skills as quickly and efficiently at Olivia and Beth
did. In this study, Olivia and Beth likely demonstrated this skill given that the extended length
of some of their resources (not many in Beth’s case). These resources were also likely longer
than those used by K-12 readers in similar research settings. With shorter resources, a reader
might not need to switch back and forth, a quick skim is enough to determine whether and entire
short article is worth reading. However, for advanced online readers of ill-structured topics, the
need to sort relevant from irrelevant ideas (Taxonomy, n.d.) within long texts may be especially
useful as it relates to synthesizing for meaning online.
Both Olivia and Beth revisited the task instructions several times during both tasks,
as well. This may be task specific (i.e., the ill-structured and complex nature of these tasks may

132

have required that they remind themselves of their goals extensively). However, given the
emphasis on meta-cognitive skills for advanced readers offline (e.g., Baker & Brown, 1984), it is
not surprising that doctoral students did this. However, another explanation is that the many and
varied texts of the Web provide even more topical uncertainty on the part of the reader than
offline texts that are pre-selected for a topic. For instance, Beth revisited her task 10 times
during the two tasks. This included doing so after reading about ways to lower you electric bills,
climate change and poverty, reusable shopping bags/laundry/cloth napkins, greenwashing, and
earth friendly clothes washing. Olivia similarly checked her task instructions after reading
about such varied topics as behavior change theory, a carbon footprint calculator, some
outlandish advantages of global warming (e.g., mountains get higher as ice melts and boundary
disputes between nations over low-lying island disappear), the social justice of global warming,
and water stewardship, and while reviewing the results of a search for [fun things for people to
do to prevent climate change].
Beth’s use of background knowledge was similar to Olivia’s in that she used background
knowledge to facilitate various elements of the synthesis process. That is, background
knowledge contributed to her keyword search phrase development, synthesis for meaning,
insights. For instance, the use of “green” to guide most of her keyword search phrases in the
reading-to-learn session was, by her own comments, entirely based on her background
experiences. Beth’s data also demonstrated how background knowledge contributes at times
when multiple resources inform the knowledge synthesis. For example, at the beginning of her
first creative synthesis Beth relied on (1) her reading-to-do notes; (2) her reading-to-learn notes;
(3) the task; (4) the browser history text; (5) four articles from session one; and, (6) background
knowledge when synthesizing. Such level of complexity further complicate any attempts to

133

determine the exact level of contribution from background knowledge and corroborate the
findings from Beth’s data that the extent to which background knowledge contributes to any
individual element of Web-mediated knowledge synthesis is unclear.
One of the other most significant differences between Beth and Olivia was Beth’s
extensive use of notes during the second session. Like Olivia, Beth made extensive use of notetaking during both tasks. The included the same elements of note-taking, verbatim notes and
summary notes, which supported Beth’s syntheses for meaning. Insight and repurposing were
also both promoted by revisiting her reading-to-learn and reading-to-do tasks. However, Beth
differed in how she used her reading-to-learn notes during her reading-to-do task. Though there
were times that Olivia used ideas that were in her notes, she never actually looked at them during
her reading-to-do task. Where as, Beth’s reading-to-learn notes were open in a tab through the
entirety of her reading-to-do task, and she used them as if they were one additional stand-alone
text in the universe of available texts on the Web. Many of the ideas that she synthesized into
her reading-to-do notes were either directly from her reading-to-learn notes, insights based on
those notes, or repurposed from them. This contrast highlights how one of the primary
advantages of online notes is that they integrate well into the Web-mediated synthesis
environment.
Chapter Summary
This chapter provided illustrations and discussion of the synthesis elements detailed in
Chapter Three, based on an in-depth case study of one individual participant, Beth. The chapter
explored each element individually and then demonstrated how they worked in concert to
comprise creative syntheses. It also compared and contrasted these findings with the findings
from Chapter 4 about Case Study #1. The chapter finished with conclusions drawn from this case

134

study about broader patterns of use, interactions, and relationships among the synthesis elements.
Conclusions that contributed to a better understanding of RQs 1 and 2 include: (1)
synthesis can be facilitated in both complex and simple online reading environments, from both
hard and soft resources; (2) like Olivia, Beth made extensive use of soft resources in various
forms of synthesis and indicated her use of soft resources was largely based on the task; (3)
elements of the creative process (e.g., idea-play and incubation) informed the environment for
Web-mediated insight; and, (4) like Olivia, Beth revisited the task instructions several times
during both tasks.
Conclusions that contribute to a better understanding of RQ 3 include: (5) keyword
development may prefigure the cognitive lens through which readers view their text(s) and the
types of synthesis that result; and, (6) both Olivia and Beth were very capable at combining the
reading comprehension skills of skimming and closer reading, and switching between them
while reading Web text.
Conclusions that contribute to a better understanding go RQ4 include: (7) Beth’s
experiences corroborated Olivia’s with regard to background knowledge in that it contributed to
several elements of the synthesis process. However, findings from Beth’s data further complicate
any attempts to ascertain the extent to which background knowledge contributes given its role in
concert with as many as five other resources that contributed to her creative synthesis.
Conclusions that contribute to a better understanding of RQ 5 include: (8) one of the
primary advantages of online notes is that they integrate well into the Web-mediated synthesis
environment.

135

CHAPTER 6
TOWARD A THEORY OF WEB-MEDIATED KNOWLEDGE SYNTHESIS
Chapter Introduction
This chapter utilizes the data and findings provided above, integrating them with related
literature to propose a Theory of Web-Mediated Knowledge Synthesis.
Background
To date, scholarship regarding both online and offline reading comprehension has largely
approached the concept of synthesis from the perspective of understanding the text(s) being read
(Duke & Pearson, 2002; Dole et al, 1991; Glister, 2000; Hartman, 1995; Leu, Zawilinski, et al.,
2007; Mayer, 2000; Rouet, 2006; Stromso & Braten, 2002). This study endeavors to expand this
perspective in two ways. First, above, it proposes a reconceptualization of synthesis to expand
beyond synthesis for meaning to include generative synthesis. Synthesis for meaning facilitates
understanding of what is explicit or implied in text(s). However, generative synthesis is based on
a post-comprehension stance that emphasizes the value-added readers provide or experience
when constructing knowledge beyond what is explicitly or implied in the text. Second, also
above, this study provides data from advanced learners that demonstrate how synthesis for
meaning and generative forms of synthesis manifest in reading-to-learn and reading-to-do tasks
about an ill-structured topic in a Web-mediated environment. Together, these contributions
provide a new lens with which to view reading and learning online. That is, they provide the
foundation for a Theory of Web-Mediated Knowledge Synthesis.

136

Before presenting the details of this theory, it is important to first revisit the value of
theory in general. The value of theory stems from its role as a “conceptual system, that is, a
system of concepts with four properties” (Olson, 2005, p. 6). These properties of these concepts
are as follows. First, the concepts identify entities that can be “pointed out and referred to”
(Olson), further highlighting relevant entities and ignoring irrelevant ones (Mishra & Koehler,
2005) In so doing, theory provides a common language with which to examine them. Second,
the concepts are linked logically to one another in a way that allows for some entities to be
defined in terms of others (Olson). Third, theories provide the basis for “causal laws” that link
concepts to one another, allowing for inference, prediction and explanation of the phenomenon.
Finally, a theory based on such conceptual systems is open to elaboration and refinement over
time. As such, a good theory balances the disaggregation of complex phenomenon into its
essential components, without being overly reductive, and makes the phenomenon more
amenable to rigorous investigation. At the same time, theory identifies and respects the
interdependence among these components.
Another way of looking at the value of theories is that they “abstract from the
complexity of everyday events and practices” (Olson, 2003, p. 6). In this way, theories
succinctly describe the various essences that make up the whole of any complex phenomenon.
These abstractions are important to moving our understanding forward in that they guide
observation. That is, when observing complex phenomenon, “simply looking, even patiently, is
not sufficient. Part of seeing...is know what to look at or for,” and abstractions provide a
framework for what to look for (Root-Bernstein & Root-Bernstein, 1999, p. 36). In this way,
theory provides the “prerequisite” framework for “precise observation statements” (Chalmers,
1976).

137

Finally, Mishra & Koehler (2005) identified how educational theories need to provide
information about how the concepts described can be applied in the real world. That is, theories
of education need to help guide the design for better ways of learning, by providing “the right
level of analysis in order to bridge the gap between description and design” (p. 29). In this way,
theories of complex educational phenomenon that respect its complexity can facilitate the
critique of more simplistic approaches to the same phenomenon. Good educational theories can
also provide a mirror that educators “can hold up to their own practices to see the ways that their
problems are both similar and different from those facing teachers in other settings” (BultermanBos, 2008, p. 413) and provide a “normalizing lens” to educators, broadening their scope of
understanding about issues for which they have significant local (or narrowly focused)
experience and understanding.
In these ways, the Theory of Web-Mediated Knowledge Synthesis offered below makes
contributions to the descriptive, inferential, and applied understanding of synthesizing
knowledge online. Below, I propose this theory in detail by outlining seven interacting elements
that emerged from this study: (1) divergent keyword search phrases; (2) synthesis for meaning;
(3) in-the-moment insights; (4) repurposing; (5) reinforcement; (6) note-taking; and (7) creative
synthesis.
Divergent Keyword Search Phrases
Henry (2006) noted how “all other decisions and reading functions on the Internet
emanate from the decisions that are made during the search process” (p. 616). In particular, this
study demonstrates the extensive use of divergent keyword search phrases in a Web-mediated
synthesis environment - keyword search phrases that integrate words and phrases not provided in
task instructions - to guide the reader’s searching and ongoing construction of a personal text.

138

In should be noted that divergent keywords search phrases as operationalized here differ from
previous conceptions of sophisticated keyword search development. For instance, LICRA
searching (learner-initiated, complex, reciprocally adaptive searching (DeSchryver & Spiro,
2008; Spiro & DeSchryver, 2010), is a cognitive perspective on keyword search development. It
describes where the keywords come from (i.e., mostly externally oriented), and emphasizes the
critical role of iteration and adaptation. As defined herein, the use of divergent keyword search
phrases is much simpler. The only requirement for a keyword search phrase to be divergent is
that one or more words are not consistent with the task at hand. Two arguments for the
importance of divergent keyword search phrase use in the process of Web-mediated knowledge
synthesis are proposed here.
First, across all eight participants in this study, the use of divergent keyword search
phrases outnumbered consistent search phrases more than three to one. That is, the advanced
learners in this study demonstrated activity consistent with what White and Drucker (2007)
called “explorers,” where their search paths, guided by divergent keyword phrases, branched
frequently, included many different keyword queries, and visited new domains of information.
Additionally, in the case studies, a comparison of Olivia and Beth’s experiences demonstrated
that divergent does not just apply to the total number of different phrases. For instance, though
Beth used nine different search phrases in her reading-to-learn task, they only represented four
different domains of search, while Olivia used twelve phrases representing eight domains. These
data suggested that Beth’s waning motivation and decrease in synthesis experiences toward the
end of her reading-to-learn task might have been related to her lack of divergence across
domains. In this way, the sequential iteration of search phrases based only on synonyms for the
same word(s) is divergent within a domain, but divergence across domains may also be

139

beneficial to successful Web-mediated knowledge synthesis. Finally, in this study, divergence
was partially achieved through the use of multiple keywords per search phrase. For example,
unlike the participants in Nachmias and Gilad’s (2002) study of search behavior, for whom the
use of single keywords was the most common search strategy, Olivia and Beth averaged 5.9 and
4.8 keywords per search phrase, respectively.
Second, the actual selection of divergent words for inclusion in a search phrase may itself
be a form of generative synthesis. That is, the choices of divergent words to include in any given
phrase may represent an active integration of entirely new categories, subjects, or fields of
information by the reader. For example, when Olivia decided to use the divergent search phrase
[religious view on climate change] in her reading-to-learn task, she said “So, when I just read
about the science, part of this topic [in the task instructions] says to open up to others, I wonder
if any…if religion affects your view on climate change.” When interviewed, she elaborated on
how this phrase developed in her mind:
I was reading about the government citing all these scientific articles, and in my
mind when I think scientific articles, we try to search for the truth...I’m not a
particularly religious person, but in religion they try to do the same thing through
different means. So I was like, that can give me perspective, people tend to follow
their religion, people tend to do their behaviors, based on their religion. Think
about how the government can cite all of these articles, but if people aren’t
listening to the science, and listening to something, and something else that can be
viewed kind of opposite, or kind of go against science would be religion, so I
wanted to get the opinions of that.

140

Her decision to chose that particular keyword search phrase was based on her synthesis of
meaning across the Web text, task instructions, and generative insights based on her background
knowledge. This is also consistent with White and Drucker’s (2007) conception of the Web
explorer that is cognitively complex, applies a global approach to learning, and focuses on the
relationships between multiple ideas during the learning process. The data from this study
suggest that not only does the variable behavior of an explorer directly support the sense-making
White and Drucker proposed (in this study considered synthesis for meaning), but also
generative syntheses beyond the text(s).
Synthesis for Meaning
Though the primary emphasis of this project was to investigate post-comprehension
synthesis (i.e., generative synthesis), the data demonstrated that synthesis for meaning, both
within a single Web resource and across the multiple sites visited, was a critical element in the
overall consideration of Web-mediated synthesis. It was the most common form of synthesis
across all eight candidates, and knowledge that was synthesized for meaning was at various
times: (1) a direct antecedent to generative synthesis; (2) later repurposed in a generative way;
(3) the content and context for combinatorial idea-play leading to insight; or (4) integrated
directly with generative syntheses to comprise a creative synthesis. Synthesis for meaning was
often from the current resource, but at other times combined as many as three ideas from recent
articles when combined and rearranged in the participants’ notes.
When first operationalized above, synthesis for meaning included a variety of activities.
Each of these was apparent in the data. Organizing, combining, rearranging, rewriting, and
compiling were integral to the process (Duke & Pearson, 2002; Dole et al, 1991; Glister, 2000;
Hartman, 1995; Leu, Zawilinski, et al., 2007; Mayer, 2000; Rouet, 2006; Stromso & Braten,

141

2002). Several instances of synthesis for meaning resulted from common reading
comprehension strategies such as summary, inference, and relating to prior knowledge (Dole et
al, 1991; Hidi & Anderson, 1986; Kintsch, 1994; Scharere, Pinnell, Lyons, and Fountas, 2005).
When synthesizing for meaning, the participants were by the very act sorting the relevant Web
information from the irrelevant (Taxonomy, n.d.). And, the participants’ notes during their tasks
represented well the active construction of their own personal text (Leu, Zawilinski, et al., 2007).
However, though it was expected that several different “technical” activities would
support synthesis for meaning (e.g., note-taking, bookmarking, clipping, highlighting), the vast
majority of synthesis for meaning was demonstrated by rather traditional note-taking, both
offline and online. In this way, note-taking was particularly important when synthesizing for
meaning online. The act of combining ideas across the text(s) visited into one text is crucial and
may be qualitatively different from merely highlighting in context or making a bookmark. That
is, seeing the relevant text clips together may foster the global understanding that emerges from
synthesis for meaning.
It is also not unexpected that synthesizing for meaning appears to be so important in the
overall environment supporting the generative synthesis processes. Generative syntheses, such
as insight and repurposing discussed below, are inherently creative activities, and many theories
of creativity propose that domain knowledge is fundamental to the creative process (e.g., see
Csikszentmihalyi, 1997; Lubart & Sternberg, 1995). In the absence of significant prior
knowledge in a domain (such as that of climate change behavior for the participants in this
study), synthesizing meaning in the moment may serve to fill in gaps in this domain knowledge
and allow for more generative syntheses to emerge. For instance, though Olivia admitted she did
not know there were any perceived benefits to climate change, after synthesizing a list of those

142

possible benefits directly from an article for their meaning, she soon thereafter repurposed the
idea of longer growing seasons to explain why certain farmers might not want to change their
own behavior in ways that would lessen the effects of climate change, if the longer growing
seasons were more advantageous to their crops. She also had an insight based on these ideas
that, in general, “even though you know there are many bad things going on with global
warming, it just may be good for you.” Sternberg (1988) summed up how important
synthesizing meaning may be in the absence of domain knowledge when he said “It is
impossible to have novel ideas about something if one knows nothing about it” (p. 137).

As

such, synthesis for meaning may be the foundation that facilitates generative syntheses.
In-the-Moment Insights
If divergent keywords are the gateway to Web-mediated insight, and synthesis for
meaning is the foundation, then insight and repurposing provide the bricks with which to build
up. A combination of Spiro’s (2006e) cognitive affordance of VCS and Sternberg’s ideas on
insight (1985) comprise in-the-moment insights based on: (1) a single Web resource; (2) multiple
Web resources; (3) activity unconnected to the Web (e.g., based on background knowledge or
idea play); or, (4) a combination of the prior three possibilities. The generative connotation of
insights requires that they provide value-added information neither explicit nor implied in the
text(s). And, they are primarily task agnostic. That is, insights are not necessarily connected
directly to the general task or purpose for reading the Web, but related specifically to the current
context of reading.
The data in this study demonstrated the complexity of insight behavior by the fact that it
occurred under the most varied and unpredictable conditions and contexts of any synthesis
element. Insights occurred based on bulleted lists of text in a single article, after the succession

143

of as many as 18 different topics across several articles, when skimming text, when reading text
at a deeper level, with significant contributions from background knowledge, and based on ideaplay. They were at different times based on both soft and hard resources, and also occurred as a
result of one-to-one (i.e., one idea in the text facilitated an insight) and many-to-one (i.e., several
ideas in the text considered in VCS lead to a single insight) relationships. They occurred during
both reading-to-learn and reading-to-do tasks. Insights also occurred in various multiples of
synthesis. That is, they occurred in the midst of several syntheses for meaning, and were at other
times reinforced themselves by background knowledge or Web text. Finally, insights comprised
a large portion of the creative syntheses by both Beth and Olivia.
As such, Web-mediated insight is a relatively unpredictable phenomenon, with the
exception of two common themes. First, in most cases, insight occurred in the context of the
reader being concerned about ideas, as opposed to remembering. Given the creative aspects of
insight, this is not surprising. Most theories of creativity emphasize the importance of “piling up
alternatives” (Osborn, 1953) and “associating or combining those ideas” (Ward, Smith, & Fink,
1999). In this way, the mindset of the reader may be most important to promoting insights.
Second, that focus on ideas often lead Beth and Olivia to skim and skip around the text
(including their own notes). Given the emphasis on incubation in creativity literature (e.g.,
Wallas, 1926), the cognitive pauses when skimming and skipping around may provide just
enough time for subconscious combination and recombination of ideas to promote insight.
Finally, it is important to differentiate how insight is conceived in this theory from its
common use in psychological and creative literature. As used here, insight is not a “solution” as
it is often described when insight is studied in the context of “insight problems” (e.g., Weisberg,
1985; Weisberg & Alba, 1981). As used in this theory, it is described as any idea or knowledge

144

that emerges as a value-added to the text that was neither explicit nor implied in the text. It does
not have to be a “light bulb” moment that happens suddenly and unexpectedly, as is commonly
associated with insight. In this way, the insight described in this study should be considered
“little-i” or “everyday” insight, just as creativity itself has been differentiated as little-c and bigC creativity (e.g., Beghetto & Kaufman, 2007; Sawyer, 2012).
Repurposing Synthesis
While insights essentially provide “new” information to the reader, the act of repurposing
allows readers to modify existing ideas in substantive and generative ways by recasting, reusing,
or otherwise tweaking them. A repurposed idea retains one or more of the important qualities of
the original idea (i.e., there is a clear parallelism between the original and the repurposed idea),
while changing or adding other qualities. Olivia’s case study demonstrated the generative
implications for repurposing particularly well. During the course of both her reading-to-learn
and reading-to-do tasks, she repurposed ideas from one context to another, by using ideas in the
opposite way they were presented in the text, by reorienting ideas to a different scale than that
explicit or implied in the text, by repurposing ideas in two different ways, and by combining
repurposed ideas with an insights. She did so from both soft and hard resources, and focused
especially on repurposing during the reading-to-do task. Repurposing occurred in both task
relevant resources and those that were more far afield. However, viewed together with Beth’s
experiences, it is likely that resources that are very close in content area to the task may be less
amenable to repurposing, given the topical consistency of Beth’s resources and her limited use of
repurposing. Finally, the keywords that preceded repurposing in Olivia’s case may have helped
to prefigure her orientation and lens toward repurposing and provide resources most amenable to
it.

145

The roots of repurposing can be found in learning, creativity and design scholarship. The
notion of transfer in learning is a broad concept that explores how learners apply or generalize
what they have learned in similar or dissimilar contexts (e.g., Barnett & Cici, 2002). It is best
considered across nine different dimensions - learned skill, performance change, memory
demands, knowledge domain, physical context, temporal context, functional context, social
context, and modality - and applies equally to skills and knowledge. Given the emphasis of
repurposing on ideas, the most relevant dimension of transfer is knowledge domain. However,
the difference between transfer among knowledge domains and repurposing in the context of this
theory is that transfer is typically the application of already learned knowledge, where as
repurposing is a component of the learning process. That is, a repurposed idea is integrated into
an ongoing knowledge construction, while transfer takes the results of a knowledge construction
and applies it.
Though these differences exist, both the conditions and the mechanisms of transfer can
inform how repurposing manifests. In particular, the idea of abstraction is helpful to
understanding how readers may repurpose ideas. That is, transfer may depend on (1) the level of
abstraction of the phenomenon to be transferred (i.e., highly abstract ideas may be more
transferrable), and (2) whether learners have actually abstracted the critical attributes of that
phenomenon (Perkins & Salomon, 1992). As such, given that any situation, process, skills, or
idea can have multiple abstractions (or essences) that together explain it fully (Root-Bernstein &
Root-Bernstein, 1999), the more abstractions that exist for a given phenomenon, and the extent
to which the learner understands all of those essences will impact how well transfer is
accomplished. As such, transferred knowledge is typically not considered “generative” in the
sense that all of it essences remain intact and are just applied elsewhere. In contrast, a

146

repurposed idea often keeps intact one or more of the abstractions while changing one or more
others to comprise the generative and value-added component.
Creativity literature often explains the process of creativity through similar lenses of
adaptation, reuse, and repurposing. For instance, Hofstadter (1985) describes creativity as
“variations on a theme.” The common tests for creative ability utilize “divergent thinking” tasks,
which are often a measure the ability to “list the different ways to use” an object (e.g., Guilford,
1967), or how it can be repurposed. Ward, Fink and Smith (1999), in describing their Geneplore
model of creativity (i.e., generate ideas and then explore them), propose that three of the
processes for idea generation are (1) transformation of ideas into new forms; (2) analogical
transfer of ideas from one domain to another; and, (3) categorical reduction of ideas. Each of
these processes describes how one of the abstracted essences of an idea (form, content, scope)
can be modified during the creative cognition process. As such, the application of these
conceptions of creativity to the reading process demonstrated online in this study provide a
foundation for the generative component of repurposing in Web-mediated knowledge synthesis.
Design literature also contributes to the concept of repurposing. Kress (2000) described
how “(re-)shaping of the potentials of existing resources” leads to transformation of ideas.
Similarly, Bereiter and Scardamalia (2003) proposed the concept of a “design mode” for
learning, whereby the goals of learning, are, among others, the improvability and developmental
potential of ideas. These ideas of transformation, improvement, and development provide
further evidence of the generative capability of repurposing existing ideas. In this way, the
design perspective emphasizes that repurposing starts with an existing idea and facilitates it
evolution.

147

Reinforcement Synthesis
The data in this study also made apparent the value of reinforcement as an element of
Web-mediated synthesis. Reinforcement is not just seeing ideas again in Web text(s). It is when
the affordance of multiplicity of the Web facilitates a strengthening of the reader’s emerging
knowledge construction. This may happen once, or many times to the same idea. It applies to
both information that has been synthesized for meaning and that which was a result of generative
synthesis (i.e., insight or repurposing). In this way, reinforcement can at different times facilitate
meaning and understanding of the Web text(s) or be generative. This depends on whether the
reinforcement applies to knowledge that was originally synthesized for meaning or a result of
generative synthesis. It also depends on the level of value-added during the reinforcement.
Reinforcement of ideas that were originally synthesized for meaning can strengthen them
ideas in several ways. First, it can cement ideas in the reader’s knowledge construction that
might otherwise be questionable or unclear. For instance, hard resources can reinforce ideas
synthesized from soft resources. Originally ambiguous ideas can also be reinforced by a simpler
and more easily understood representation. Conversely, additional details can reinforce ideas
that were too simplistic in their original state. All of these ways help to facilitate how well the
reader understands what they have synthesized for meaning. However, if, while reinforcing
ideas that were originally synthesized for meaning, the reinforcement prompts elaboration on the
original idea, the reinforcement becomes generative. These elaborations can often be based on
differences in context, perspective, and time that appear in the reinforcement Web text
environment.
Reinforcement of ideas that resulted from generative synthesis can be similarly
strengthened, in additional ways to ideas originally synthesized for meaning. When the reader

148

questions or disagrees with the text (representing a generative insight), their disagreement can be
later reinforced and strengthened in the knowledge construction. Insights that were originally
based on background knowledge or idea-play can also be reinforced by later Web text. Finally,
reinforcement of ideas that were originally generative that includes elaboration on those ideas is
also generative.
Previous literature supports the value of reinforcement. For instance, re-reading is
commonly associated with increased recall and comprehension (e.g., Bromage & Mayer, 1986;
Rawson, Dunlosky, & Theided, 2000; Howe, 1970). This would explain the value of
reinforcement to ideas that had previously been synthesized for meaning. Similarly, revisitation
of previously read information has been shown to increase facts learned and strengthen weak
connections among the material read (Lawless & Brown, 1997). On the Web, users may return
to text read previously once it is deemed relevant at a later time (Junivo, 2006; Wen, 2003). They
may also return to previously read resources to verify or retrieve information or revisit
challenging ideas after learning more about them elsewhere (Desjarlais, 2010) and revise
constructed meaning (Afflerbach & Cho, 2008). However, all of these perspectives
demonstrated the effect of readers choosing to revisit the exact text they had already read at least
once.
Reinforcement, as proposed in this theory, is when the same (or very similar) ideas
emerge multiples times, but in different resources. This is what allows for elaborative and
generative potential, often based on the new context, perspective, or timeframe of the new
resource. Reinforcement is akin to the notion of revisitation and conceptual variability in CFT
(Spiro, Collins, & Ramchandran, 2006; Spiro et al. 1992; Spiro & Jehng, 1990), which described
how hypertext systems could facilitate flexible understanding through revisiting the same ideas

149

at different times, in various contexts, with different purposes, and from different conceptual
perspectives. However, two differences are apparent between the CFT/CFH and reinforcement
as proposed here. First, most CFT and related CFH systems were closed systems with a finite
amount of information related to a specific topic. As such, the opportunities for serendipitous or
unplanned reinforcement to occur and for these to occur in resources that are topically far afield
from the task were more limited in CFT/CFH systems than that afforded in an open online
environment. The availability of resources that are more far afield from the task on the openWeb may provide the potential for more generative synthesis, given that creative insights often
result from combining ideas from different domains of knowledge (Ward, Smith, and Finke,
1999) Second, CFT and CFH systems often pre-selected the themes to be explored in this way
for users to specifically select as part of the navigational options of the systems. In this way
conceptual variability use in CFT and CFH systems promoted a synthesis for meaning of the
multiple texts included in the systems, the ultimate goal for which was to prepare learners for
flexible application of that meaning in new contexts. Where as, reinforcement as proposed here
is in the context of both synthesis for meaning and generative synthesis. For instance, if a Webbased reader has an in-the-moment insight that provides new ways of thinking about a particular
idea, the reinforcement of that idea later on serves to crystallize and/or adapt that generative
insight.
Reinforcement can also demonstrate an affective component that is beneficial to the
reader’s motivation and even their reading strategies. For instance, if the reader questions or
disagrees with the text and this insight is reinforced later by the text, the reinforcement can be
encouraging to the reader in a way that may promote even more questioning and disagreement.
Similarly, any time the reader is reinforced for synthesizing meaning or generative synthesis, the

150

potential for increased positive affect exists. As such, there may be an impact on self-efficacy
(Bandura, 1994); motivation to learn (Brophy, 2004); intrinsic motivation (Deci & Ryan, 1985);
flow (Chen, Wigand, & Nilan, 1999; Csikszentmihalyi, 1990); or, interest (Hidi & Renninger,
2006). The motivational benefits may be especially relevant for forms of generative synthesis,
which involves more cognitive engagement and personal investment than synthesis for meaning.
Finally, it should be noted that reinforcement can manifest in minimally beneficial ways.
As was demonstrated in Beth’s case study, synthesizing an idea in a list of bullet points the first
time it occurs, and then reading it again in another list elsewhere did little for the emerging
knowledge construction. Similarly, reading strategy can impact the potential benefits of
reinforcement. For example, if the reader is only reading small segments of text all of the time
and never considers the different contexts or perspectives when the same idea occurs in different
resources, the effects of reinforcement may be minimized. And, if the resources visited over
time are too similar and repeat the same information several times in simple ways or without a
change in context, perspective or time, there may be decreasing returns to the value of
reinforcement. This may be related to keyword search phrase development, in that a narrow
scope of largely consistent phrases or even divergent phrases in the same general area may
decrease the potential benefits of reinforcement by resulting in resources that are too similar in
context or complexity.
Note-Taking
An abundance of research indicates that taking notes is beneficial to learning in a variety
of contexts (e.g., Kiewra, 1989; Kintsch & Kintsch, 1996; Makany, Kemp & Dror, 2009). The
data in this study extend those findings and demonstrated that note-taking is also a critical
element in the process of Web-mediated knowledge synthesis. Across eight participants, all of

151

them took at least some notes. As demonstrated above in case study #1, Olivia was particularly
effective with her notes. She supported synthesis for meaning with both verbatim and summary
notes. Several of her insights were prompted by reviews of her notes. Review of her notes also
informed the development of keyword search phrases. And, adding ideas to her notes (either for
meaning for in generative ways) facilitated the repurposing of those same ideas to other sections
of the notes. Several of her generative insights were supported by conscious idea-play based on
the information in her notes. Finally other more “random” insights occurred when working with
her notes. These may have been facilitated by the opportunity for internal incubation that
leaving Web-text(s) to review or add notes may have provided (i.e., letting up of her focus on the
Web text(s) may have allowed for subconscious idea-play resulting in insight).
Beth also used her notes extensively, especially during her reading-to-do task. In
addition to many of the benefits to synthesis Olivia demonstrated, Beth used her reading-to-learn
notes through the entirety of her reading-to-do task as if they were an additional stand-alone text
in the universe of available texts on the Web. Many of the ideas that she synthesized into her
reading-to-do notes were synthesized for meaning from her reading-to-learn notes and several
generative syntheses in her reading-to-do task were based on her reading-to-learn notes. This
contrast highlights how one of the potential advantages of online notes is that they integrate well
into the Web-mediated synthesis environment.
As such, this theory proposes that note-taking is a fundamental part of Web-mediated
knowledge synthesis. This relationship is unique to theories designed to explain online reading.
For instance, the Taxonomy (n.d.) outlines the importance of note-taking for successful online
reading comprehension, but in the context of communication. Based on the data here, note-

152

taking should be considered integrated, and an activity that directly supports both synthesis for
meaning and generative synthesis.
While the differences between online and online notes were not within the scope of this
study, as Figure 6 demonstrates, these affordances may be very different. Among those taking
notes on the computer, bullet points, lists, and outlines were common, where as offline notes also
included lists, but were more freeform in their structures. It is possible that the freeform use of
offline notes is beneficial to some elements of Web-mediated synthesis, and the easily integrated
nature of online notes to others. These are questions for further examination in future studies.
However, based on the data in this study, both offline and online notes supported Web-mediated
knowledge synthesis in substantive ways.

Figure 6. Computer based versus written notes. The text in this figure is not meant to be
readable. The figure is simply intended to demonstrate linear vs. non-linear note-taking.
153

As it relates, one of the specific affordances of note-taking that should be considered
when integrating notes into Web-mediated knowledge synthesis is their permanence or lack
thereof. That is, Beth easily integrated her reading-to-learn notes as a stand-alone text during her
reading-to-do task, in part because they became a permanent part of the Web in Google Docs.
Similarly, the fact that Olivia effectively ignored her offline notes from her reading-to-learn tasks
(with the exception of what she could recall from them), is potentially indicative of the
temporary nature of offline notes in an online world. As such, the value of notes to the “in-themoment” processes of Web-mediated knowledge synthesis should be considered based on their
various online and offline affordances. Based on these ideas, using both offline and online notes
may be valuable to the Web-mediated knowledge synthesis. However, the existence of evolution
of touch screen technology and cloud computing may facilitate the merging of the various
affordances of offline and online note-taking, rendering concerns over permanence and structure
unnecessary.
Finally, note-taking in this environment should also be considered as a broader spectrum
of activity than is typically associated with and reading and learning. Though participants in this
study did not use many of the new types of online note-taking tools that facilitate clipping, tags,
keyword searching, and inline annotation (e.g., Evernote and Diigo) these activities and others
should be considered in the context on their note-taking value to Web-mediated synthesis. In a
follow-up study to this project (DeSchryver, in preparation), the value of clipping and tagging in
Diigo is explored in this area through data provided when Olivia and Beth returned for additional
research sessions. Prior to these additional sessions, both participants were taught how to use
Diigo, and then encouraged to continue using it during research sessions with similar learning
environments to their first sessions. The value of these activities to the synthesis elements

154

proposed in this theory will be explored, both in case studies specific to those sessions, and in
comparison to Beth and Olivia’s use of more traditional notes in this study.
Creative Synthesis
The sequential occurrence of several of the synthesis elements above, including instances
where they are experienced at virtually the same time, comprises a creative synthesis. It
represents a substantially generative way of interacting with Web text(s), background
knowledge, idea-play, notes, and the task undertaken. An online reader could read single or
multiple online texts and practice generative synthesis without ever experiencing creative
synthesis, since the data from both Olivia and Beth demonstrated how the elements above can
and do exist in isolation. However, as proposed, creative synthesis is essential if Web-mediated
knowledge is to be constructed in ways that can address the big questions that ill-structure topics
pose.
As such, in this theory, the concept of creative synthesis is largely the application of the
creative process to reading and synthesizing Web text(s). In this way, the theory moves both
scholarly conceptions of reading and creativity in new directions. As noted above, it opens the
definition of synthesis in reading research up for more creative and generative examinations.
Similarly, creativity scholarship is often from the perspective of problem solving, science,
inventions and the arts (e.g., Weisberg, 2006), workplace dynamics (e.g., Amabile, Conti, Coon,
Lazenby &Herron, 1996), or psychological processes (e.g., Sternberg & Lubart, 1999). To be
creative (or generative) in one’s thinking while reading text (or in this case Web text) is a unique
perspective on creativity.
However, both bodies of work help to explain the phenomenon that emerged in this
study. Reading comprehension scholarship explains many of the skills both Olivia and Beth

155

integrated during their creative syntheses (e.g., synthesis for meaning, summary, inference,
prediction and accessing background knowledge). At the same time, the multiplicity of synthesis
elements in the creative syntheses Olivia experienced are consistent with Root-Bernstein and
Root-Bernstein’s (1999) conception of layering their own creative thinking skills one on top
another, often in multiples more than two. Finally, as noted above, Csikszentmihalyi’s (1990)
notion of flow informed how creative synthesis can be recognized in a Web-mediated
environment.
Web-Mediated Knowledge Synthesis
Together, these elements comprise a Theory of Web-mediated synthesis. The
presentation of these elements is not provided in a sequential fashion nor intended to imply a
linear relationship. All elements of the model can and do interact and work in concert with the
other elements. Both insights and ideas synthesized for meaning can be repurposed or
reinforced. The use of divergent keyword search phrases may precede synthesis for meaning,
insights, repurposing, and reinforcement at various times. Ideas synthesized for meaning,
repurposed, and reinforced, as well as insights and notes, can all inform subsequent divergent
keyword search phrases. Similarly, notes support synthesis for meaning, insight, repurposing
and reinforcement. And, in various combinations, these elements all support creative synthesis.
Figure 7 demonstrates these relationships.

156

Figure 7. The Theory of Web-Mediated Knowledge Synthesis

157

CHAPTER 7
SUMMARY
Introduction
This study examined how eight advanced learners synthesized knowledge when using the
Web to explore and apply ideas about an ill-structured topic, climate change behavior. It was
conducted from a post-comprehension perspective in order to better understand generative
synthesis experiences, wherein knowledge is constructed that is neither explicit nor implied in
the online texts viewed. This final chapter summarizes its findings and discusses the specific
implications of these findings through descriptive, inferential and applied perspectives. It also
presents the limitations of the study.
Synthesis for Meaning and Generative Synthesis
One of the primary contributions of this study is descriptive. That is, based on an
analysis of previous literature about synthesis, with an emphasis on the implications of creativity,
this study specifically delineates the difference between synthesis for meaning and generative
synthesis. Providing these two specific lenses from which to observe reading and learning online
clarifies an important distinction related to synthesis, while respecting its complexity. For
instance, the terms combine, rearrange, and summarize are equated with synthesis. Similarly,
reading researchers have used the term synthesis to describe summarizing (e.g., Dole, et al,
1991) or inferencing (Dole, et al; Kintsch, 1994; Scharere, Pinnell, Lyons & Fountas, 2005).
However, summary and inference are typically utilized in the context of understanding textual
meaning, not a generative synthesis of it. In this study, I define generative synthesis as activity
that provides value-added knowledge, which is neither explicit nor implied in the text(s). As

158

such, this study proposes a specific terminology, synthesis for meaning, to account for common
ways of operationalizing synthesis online - these include the ability to sort relevant information
from irrelevant information, organize and understand the texts available (“Taxonomy”, n.d.) and
assimilate multiple sources across multiple modes (Glister, 2000). It also contributes new
terminology, generative synthesis, to extend ideas about synthesis so that they include more
creative activities. As such, this theory provides substantial descriptive clarity to researchers
and practitioners interested in synthesis.
Summary of Findings from Research Questions
Five research questions about synthesizing knowledge in an open online environment
guided this study. This section provides a summary of the findings specific to each question.
RQ1 - How do synthesis for meaning, VCS synthesis, and creative synthesis
manifest for advanced learners in Web-mediated reading-to-learn and reading-to-do
environments? Previous literature and scholarly insights provided the foundational framework
for this study - synthesis for meaning, VCS synthesis, and creative synthesis. The data across
eight participants operationalized definitions and corroborated findings regarding synthesis for
meaning and creative synthesis. Synthesis for meaning was the most widely used form of
synthesis across participants, was common to both reading-to-learn tasks and reading-to-do
tasks, and was supported by both soft and hard resources. Creative synthesis emerged as a related
sequence of the other forms of synthesis, and was primarily apparent in the reading-to-learn
tasks. Both hard and soft resources also supported it.
However, as first defined, VCS synthesis did not fully explain the data. As such, this
definition was broadened to include VCS as part of a larger category of generative synthesis
activity called in-the-moment insights that represented a point of overlap between creativity

159

theory and reading theory in the construction of the Theory of Web-mediated Knowledge
Synthesis. One of the foundations of creativity is that a multiplicity of ideas is necessary; this is
also one of the affordances of the Web. As such, even when insights in this study were
determined to be specific to one resource, they were in the context of many ideas from many
resources in a relatively short period of time (60 minutes for the reading-to-learn task and 45 for
the reading-to-do task). The juggling of these ideas, whether in bulleted lists, through skimming,
from deeper reading, in soft and hard resources, and with topics both close to the task and more
far afield, demonstrated a creative stew amenable to generative insights.
When reading about how people tend to hope that future technology will save the world
from climate change, one participant had the insight that the broader issue may be that people
just like to put the responsibility on someone else. This idea was neither explicit nor implied in
the text, but a generative idea on the part of the reader. As defined in the literature, VCS would
imply that such an insight be from multiple resources considered in virtual simultaneity. The
broader category of insight defined herein accounts for such activity based on singular resources
(often as simple as a sentence, as was the case in this example).
RQ2 - What other forms of generative synthesis emerge from the data, and how do
they manifest, for advanced learners in Web-mediated reading-to-learn and reading-to-do
environments? Three synthesis activities identified in the data could not be explained by the
elements found in RQ1. This resulted in the operationalization of three additional elements divergent keyword search phrases, repurposing, and reinforcement. Divergent keyword search
use (i.e., using keywords that were not included in the instructions or task text) provided a
gateway to other forms of synthesis, and was itself somewhat generative when it actively opened
up new topics and domains of information to the reading and learning processes. It was used

160

nearly three times as often as consistent search phrases across all eight participants in this study.
Divergence was also evidenced across participants, by the fact that in 14 hours of Web-based
learning, only one keyword search phrase was duplicated across participants. A downstream
impact of this was that out of 233 Web sites visited, only seven were visited by more than one
participant.
Repurposing emerged as a form of synthesis and involved modifying existing ideas,
either from Web text, notes, or background knowledge, in substantive and generative ways, with
much of its process being similar to the creativity literature. That is, not all creative ideas are
“new.” In fact, Hofstadter (1985) indicated “making variations on a theme is really the crux of
creativity” (p. 233). His analogy was that concepts can often be defined and described by knobs,
and that creativity results from tweaking one or more of these knobs. These knobs are the
abstractions, essences, and attributes of ideas. The data in this study demonstrated in detail how
participants changed one or more of these knobs, while the remaining knobs stayed the same, as
they repurposed knowledge in generative ways. This was often from one context to another, by
using ideas in the opposite way they were intended in the text(s), or by scaling ideas up or down.
For instance, when reading about how most people indicate a willingness to make financial
sacrifices to help combat climate change, one reader incorporated this ideas in the exact opposite
way, that many people who do not change their behavior do it because it is too expensive.
Reinforcement emerged as a generative act of seeing ideas (from Web text, insights,
notes, or background knowledge) in new resources and contexts in ways that facilitated a
strengthening of the original knowledge construction. This manifested in various ways - hard
resources reinforced soft resources; ambiguous ideas were reinforced with more clarity; and
simple ideas were reinforced with more complexity. For instance, after synthesizing the notion

161

that people only act about things like climate change when it affects them personally from a
generic behavioral paper, a reader saw the same idea in the context of water rights. Like
synthesis for meaning and in-the-moment insights, both repurposing and reinforcement occurred
during reading-to-learn tasks and reading-to-do tasks, and were seeded by both soft and hard
resources.
All of the elements of synthesis that emerged from RQ1 and RQ2 manifested in a variety
of ways. In fact, together they represent a process that is itself complex, multiplicitous,
unpredictable, and ill structured. They manifested in a variety of context-dependent ways. They
all appeared during both deeper and more cursory reading strategies, but in no predictable
manner (e.g., see the differences in general reading strategies in case study #1 and case study
#2). They all integrated soft resources at some point in the knowledge synthesis, but without a
discernible pattern. Each element of synthesis was supported at one time by task-relevant
resources (e.g., the National Oceanography and Atmospheric Administration Web site), and at
others by resources more far-afield (e.g., a Web site outlining how to get people to take the stairs
instead of the escalator). Some participants utilized more of one synthesis element than another
based on the task (reading-to-learn vs. reading-to-do), while others used them more equally. One
consistent finding in the environment for Web-mediated synthesis was that the description of the
task and supporting instructions were used extensively during both reading-to-learn and readingto-do, often directly supporting synthesis activities (as was the case in both case study #1 and
case study #2).
RQ3 - How do synthesis for meaning activities and generative synthesis activities
interact for advanced learners in Web-mediated reading-to-learn and reading-to-do
environments? The primary way that the elements of synthesis interacted in this study was to

162

work in concert to support creative synthesis. Once again, this occurred in unpredictable ways.
This study examined five specific instances of creative syntheses from two participants. One
was comprised primarily from knowledge repurposed from previous synthesis for meaning
activities. Another relied primarily on generative forms (insights, repurposing, and
reinforcement) working in close sequence. A third emerged from several generative elements, as
well as synthesis for meaning processing the same paragraph of text. A fourth creative synthesis
during a reading-to-do task was comprised of both syntheses for meaning and generative
syntheses based on several different resources: (1) reading-to-do notes; (2) reading-to-learn
notes; (3) task instructions; (4) browser history text; (5) four Web texts from the reading-to-learn
task; and, (6) background knowledge. A final creative synthesis occurred without any Web
searches at all. That is, the makeup of creative synthesis was not consistent in any way.
Outside of the context of a creative synthesis, the other elements of synthesis often
appeared in close proximity to each other, or complemented each other directly. Synthesis for
meaning seeded divergent keyword search phrases, as did insights and repurposed knowledge.
On the other hand, divergent keyword search phrases directly preceded insights, syntheses for
meaning, repurposing and reinforcement. However, so did consistent phrases. Knowledge
synthesized for meaning was repurposed and reinforced, but so was knowledge based on insights
and participants' backgrounds. The model that emerged indicates that an organic, but
unpredictable pattern of interactions supports Web-mediated knowledge synthesis with all of
these elements potentially supporting each other at any given time.
RQ4 - What role does background knowledge play to support the knowledge
synthesis of advanced learners in Web-mediated reading-to-learn and reading-to-do
environments? Participants in this study accessed background knowledge during keyword

163

development, made insights based on background knowledge, repurposed background
knowledge, and integrated background knowledge during creative syntheses. This was clear.
What was not clear was the extent to which background knowledge contributed to each of these
activities - all of these processes worked so often in rapid sequence or in virtual concert, parsing
background knowledge from acquired knowledge proved most difficult. This is consistent with
previous indications that the role of background knowledge “may be somehow less important
when reading on the Internet” (Coiro, 2011, p. 360; Spiro & DeSchryver, 2010), and that the
implications of background knowledge may be increasingly less obvious.
This study contributed a descriptive framework for future studies in its attempts to
differentiate idea-play, insight and background knowledge. Specifically, the ability to access
both background knowledge and insight typically occurred in relatively quick time frames,
whereas idea-play occurred over a longer period of time. However, idea-play may have been
based on both insight and background knowledge, since recalling background knowledge may
lead to an insight, and idea-play may also result in the same. These complications were detailed
in the second creative synthesis for case study #1.
RQ5 - What role does note taking play to support the knowledge synthesis of
advanced learners in Web-mediated reading-to-learn and reading-to-do environments? The
role of note-taking in the web-mediated synthesis process was much clearer. All participants
took notes, either offline or online, and sometimes both. Nearly every documented case of
synthesis activity in this study was directly supported by note-taking, or in the context of
participants reading their own notes. Unlike previous frameworks that identified note-taking as a
separate online reading process from synthesis (e.g., see “Taxonomy”, n.d.), this study
demonstrated that note-taking integrated with every aspect of synthesis. Keyword development

164

was even, at times, seeded by participants reading over notes taken earlier in their research
sessions.
One open question regarding note-taking is the extent to which online notes may better
integrate with Web-mediated synthesis than offline notes. The in-depth case studies in this
project provided the following: the first participant took all of her notes offline and did not
access any of the notes from her reading-to-learn session during her reading-to-do session;
whereas the second participant took all her notes online and made extensive use of her readingto-learn notes during her reading-to-do session. Whether this was an individual preference and
cognitive style, or was based on the affordances of online notes could not be determined in this
study. A follow-up study is underway with data from these two participants taken during
additional research sessions that examines more closely the specific affordances of online notetaking.
The Theory of Web-Mediated Knowledge Synthesis
Together, the results from these research questions provided the theoretical structuring
and illustrated examples of a new framework for considering synthesis in open online
environments. As such, the primary contribution of this study is a Theory of Web-Mediated
Knowledge Synthesis detailed in chapter 6. It is based on the interdependent elements of
divergent keyword search phrase development, synthesis for meaning, in-the-moment insights,
repurposing, reinforcement and note-taking. Taken together, these elements lead to creative
synthesis. Extensive data in this study supported each of these elements. This theory combines
information from the reading literature, previous scholarship about creativity, and results from
this study to propose a post-comprehension perspective for reading online. In so doing, it
provides a lens through which to better explore and understand generative reading and learning

165

activities online.
The need for a theory that incorporates a creative or generative perspective for reading
and learning online is based in large part on the original motivations for this study. Primary
among these is the new ecology of information, knowledge, teaching, learning, and thinking
brought about by ubiquitous access to the Web. Anderson & Rainie (2010) surveyed multiple
experts who indicated that the Web will shift cognitive capacities in ways that require that we
think harder and spend less time on memorization - “Google allows us to be more creative;”
Google “supports SOME parts of human intelligence, such as analysis, by REPLACING other
part such as memory;” “Google will make intelligence different.” At a time when this sort of
transformation is occurring, we need more theories to guide our exploration of the related
phenomenon, and the Theory of Web-Mediated Knowledge Synthesis is one such attempt.
These changes in available technologies and the cognitive implications of them are
rapidly shifting. As such, this theory is intended to be just a beginning. As Olson (2005)
indicated, one of the properties of a valuable theory is that it is open to elaboration and
refinement over time. For this theory in particular, not only is it open to elaboration and
refinement, but this is expected. New elements of synthesis will be added, and branches of the
proposed elements will emerge from further study. Given that this theory was constructed based
on data from advanced adult learners, developmental heuristics will be needed to indicate what
expectations for Web-mediated synthesis are appropriate for various age levels. This theory will
also need to adapt as technologies themselves become smarter and more refined.
Though technological progress requires new theories like this to describe and explore the
related cognitive implications, one of the other motivations for this study was cultural. That is,
the world continues to be more and more complex. Whether technology proves to offload

166

memory and allow for more creative thinking or not, we still need more creative thinking to be
better able to address the increasingly complex issues facing society. Complexity is apparent at
the individual, local, regional, national, and international levels (e.g., Blaufus & Ortlieb, 2009;
Gunther, 2010; Hanoch, Wood, Barnes, Liu, & Rice, 2011; Hollander, 2004; Mitroff, Alpasian &
Green, 2004; Schwab, 2001; Thaler & Sunstein, 2008). The complex, ill-structured and wicked
problems that face us at home and at work need better methods of learning to address them than
are currently the norm. This theory provides a starting point from which to develop such
methods in order to better address the increasing complexity of the world.
Implications of the Theory
While Chapter 5 provided the specifics for a Theory of Web-Mediated Knowledge
Synthesis, it also outlined the importance of theory. It identified how theories are typically
conceptual systems (Olson, 2005) that “abstract from the complexity of everyday events and
practices” (p. 6). They break down complex phenomenon into their essences in order to help
others know what to look for, identify how some of the essences are defined in terms of others,
and allow for inference and prediction to be made based on these essences. Specific to
educational theories, Mishra and Koehler (2005) outlined how once these essences are
determined, their value can be viewed through descriptive, inferential, and applied lenses. As
such, the following explores the specific contributions that the Theory of Web-Mediated
Knowledge Synthesis provides in each of these areas.
Descriptive. The notion of synthesizing information when reading and learning online is
a complex and multi-faceted phenomenon, and it has proven difficult to observe (e.g., e.g., Leu,
Zawilinski, et al., 2007). This difficulty may in large part be due to the lack of specificity and

167

clarity regarding what to look for when observing it. This theory endeavors to ameliorate these
issues in four ways.
First, as highlighted above, one of the primary contributions of this study is descriptive,
in that differentiating synthesis for meaning from generative synthesis provides researchers and
practitioners alike with a more precise conception of synthesis related activity than existed in
prior literature.
Second, this theory provides several specific details about the conception of generative
syntheses that learners and readers may experience online. The concepts of in-the-moment
insights, repurposing, reinforcement, and creative synthesis further clarify what researchers
might look for when observing this phenomenon. Though the activities that these terms describe
are not “new” in the sense that each can be traced to scholarly antecedents, this theory is unique
in that it includes all of them as a way to describe one specific phenomenon, that of Webmediated knowledge synthesis. In particular, the borrowing of terminology from the field of
creativity research provides additional ways to describe the various forms of synthesis when
learning and reading online.
Third, this theory integrates the importance of note-taking (both traditional and nascent
online tools that support a myriad of similar processes) in the process of Web-mediated
knowledge synthesis. Prior conceptions of online learning and reading have identified notetaking as an important component of online reading (e.g., TICA, n.d.), but they did so as a
separate process from that of synthesis. From a descriptive perspective, this theory makes the
case that it should be difficult to conceptualize online synthesis without considering a variety of
note-taking techniques.

168

Finally, the Theory of Web-Mediated Knowledge Synthesis explicitly provides a
foundation for appreciating the relational complexity among all of its components. While
components of the theory may manifest in isolation, (e.g., synthesis for meaning) it is their
inter-relationships that best describe the complexity of Web-mediated knowledge synthesis. In
this way, this theory provides a roadmap for observing and describing the components in
disaggregated ways, but also emphasizes the importance of considering how each supports or
facilitates the others.
Inferential. While the descriptive clarity and specificity and the emphasis on interrelational activity outlined just above allows researchers to better see what Web-mediated
synthesis look like, it also provides a solid foundation from which to make inferences and
predictions about how they interact. In this way, it opens up the field to several testable
hypotheses that may not have otherwise been considered. For instance, based on the specific
description of divergent keyword search phrases, one might predict that extensive and extended
use of divergent keyword search phrases will lead to increasingly creative syntheses of
knowledge. Similarly, this theory provides a foundation for predictions about the role of soft vs.
hard resources and task relevant vs. far afield resources (e.g., do task relevant resources seem
more amenable to insights or repurposing when compared to resources that are more far afield?).
Given the complexity and rapidly changing contexts within which Web-mediated knowledge
synthesis occurs, myriad inferential possibilities exist. However, this theory provides parameters
for the predictions and inferences to be made that may make the process of hypothesis
construction more manageable.
Application. The field of educational research is particularly prone to creating gaps
between research and practice, theory and pragmatics. However, as Mishra and Koehler (2006)

169

have noted, “a good theory or framework offers us the right level of analysis in order to bridge
the gap between description and design” (p. 29). In this way, the Theory of Web-Mediated
Knowledge Synthesis offers two benefits to practitioners and those working with them. First, it
provides a new lens with which to view existing, and possibly too-simplistic, perspectives of
online synthesis. For instance the TICA (“Taxonomy”, n.d.) provides an excellent framework
for developing questions, locating information, evaluating information, and communicating
information from which numerous successful professional development sessions have been run
(e.g., http://www.lite.iwarp.com/CoiroVT2008.html). The addition of the Theory of WebMediated Knowledge Synthesis to expand upon the more limited details provided by the TICA
about synthesis for future professional development sessions may serve fruitful to practicing
teachers. Second, several ideas from it can be reviewed from the perspective of educational
change. Given the speed with which the new ecology of information, of reading, of knowledge,
of teaching, of learning, and of thinking is emerging, it is not practical for schools to wait for the
traditional sequence of theory development, experimental research, and replication to be
complete before considering better ways to teach and learn. Consequently, this theory provides
an additional perspective through which educators can think about the future of education. The
Theory of Web-Mediated Knowledge Synthesis is largely based on an assumption that
educational environments will increasingly move toward 1-1 ubiquitous access to digital devices
and that this technical evolution requires an evolution in what and how teaching and learning
take place. For those who believe that there will be less emphasis on rote learning and
memorization and more time for higher-order thinking, this theory provides them with ideas
from which to build their vision of the future of education.

170

Limitations
There are several limitations in this study to consider. The limitations of subjectivity
were discussed in the method section, Chapter 2. In addition, though the use of think-alouds is
commonly used for reading and Web-based research, it only provides “only a small window into
the inner workings of the mind...limited by the awareness of [participants] of their own thought
processes and their ability and willingness to communicate this awareness” (Hartman, 1995, p.
530). Though the participants in this study were advanced learners who may have had a higher
awareness than most in this area, the limitation still applies. Topical limitations were inherent in
this study, given that the first author chose the topics. If a topic was more relevant to one
participant than another, their willingness and ability to synthesis may have been enhanced or
diminished. However, since this study endeavored to demonstrate and explicate synthesis, the
fact that one participant may have synthesized more than another would not have impacted the
development of the theory or the findings of the individual in-depth case-studies. Finally, it
should also be noted that the interviews included in the design of this project were intended to
serve primarily as a source for triangulation of the think-aloud data. Unfortunately, interview
data netted very little useful information. This was in part due to the time constraints of the
sessions, but was also due to the emergent nature of the study. The interview protocols did not
have questions designed to target several of the elements of synthesis, given that they did not
emerge until during data analysis. However, the combination of think-aloud data with screen
video and expert review provided ample data with which to address assumptions about the thinkalouds.

171

APPENDICES

172

Appendix A
Profile for Olivia
Prior to the study, Olivia provided the following information about herself in a survey of
Web Learning. She reported using the Web for more than four hours per day on a typical day,
and for three hours per day for learning. On a likert scale of 1-5 (5 being the most), she reported
that she like to use the Web for learning tasks at a 5 level. She reported having recently used the
Web to learn about animal approval forms, differences between Affymetrix gene chips and EST
arrays, performing animal surgery, the cost of gavage needles, and information about traveling to
Memphis. She reported using browser history, bookmarking, multiple browsers and notecards to
organize and manage her Web resources and was unfamiliar with the concept of tagging. When
asked to provide possible search phrases to learn about the effects of climate change on global
poverty she listed: [climate change affect poverty], [global warming affect on poverty], [climate
change third world country], [climate change poverty], [economics of climate change], [climate
change affects economics of world], and [climate change affect world poverty]. She noted that
choosing keyword search phrases was easy for her and that she “picks out the main words of
what I want to know and leaves out the other nonsense words like ‘the, of, on.’ I also will think
of multiple ways to say the same thing – also the way in which other people are most likely to
say something.” She noted that “aha” or “light bulb” moments when learning on the Web usually
came to her when “learning about something I do not understand in grad school – finally I am
able to connect enough pieces together to get the overall picture – that is when it comes together
to make sense.” On a scale of 1 (beginner) to 5 (expert) she rated her abilities in the following
categories:

173

Finding information related to your studies
Organizing/managing the Web information you find that
is relevant to your studies
Determining the reliability of Web sites
Determining the relevance of Web sites to your specific
tasks
Determining the bias/stance of Web sites that you use
for learning
Synthesizing information from the Web into coherent
solutions
Using the Web to construct creative solutions to
problems
Using the Web to have “aha” moments of conceptual
breakthrough
Setting goals in your Web learning tasks
Staying focused when using the Web to learn

4
2
3
5
2
3
2
2
1
2

At the beginning of the first research session, I asked Olivia to tell me everything she
knew and her background and experiences on the topic of climate change. Her response provided
the following details:
So, as for climate change, I guess it can be controversial with people, as to
whether or not they think that climate change and or global warming is actually
taking place. I have read about climate change, and I believe that it is in fact
taking place….and what impressed me, when I think climate change
automatically, I think about Al Gore’s documentary on the whole situation…and
being a scientist, I believe that its actually taking place when reading from the
scientific papers…and when you see the picture, too, from, for instance, that
arctic circle many years ago versus now…there’s not doubt that its taking
place…now, I know there’s controversy around it because some people will
actually say that it is the governments way of controlling what we do and how we
act by saying that we have to regulate ourselves and regulate what we use…and

174

they don’t believe it so they aren’t going to do certain things so that would help
prevent climate change and or global warming from taking place. Usually when
people confront me like this, I usually hold my tongue because I’m informed and
the only thing I can do to try to combat the problem and to make wiser decisions
involving climate change. I’ve noticed that the change, whether its warming or
cooling is taking place because our weather has been quite dramatic here as I’ve
gotten older, it’s just variable. I remember when I was younger, it used to be cold
in the winter and hot in the summer, but now it seems to be jumping all over the
place. But, that may be just because people are telling me it is taking place, but I
don’t know if personal experience can go along with that. I also read about
animals and plants and how they are becoming more and more extinct because the
climate is changing, and how different niches are not being filled and how that’s
leading to a greater catastrophe other than just global warming…and the ozone
layer and stuff like that that are taking place…I know I hear politicians talking
about it…I know now that we’re talking about climate change I was thinking
about Al Gore’s documentary and in my mind comparing it to Michael Moore,
because I know these documentaries…even though you agree with them…they
can get all hyped up and overdramaticized…I’ve not had any classes specific to
global warming. I’ve had classes that focus on the ability of organisms to change
their genetics, or evolve to adapt to the changing climate…I’ve read about that
and I’ve seen that data…

175

Appendix B
Olivia’s Synthesis for Meaning Example
As part of this activity, Olivia said ”somebody else’s opinion was that it depends on the
type of media, or subset of media, you received…” then wrote in her notes “media you receive.”
The text to which she was referring (a personal blog) had been read a few minutes earlier in the
first article, which she read out loud.
This is about how science, politics, and the media ‘place’ discussions about
climate change, global warming, emissions policy and the like...There appears to
be an underlying assumption that the ’frame’ of debate within which public
awareness is constructed is definable as a subset of the range of views on climate
change. My intuition, based on personal experience on the Netweather.tc website,
on of my favoured for a, is that the range of public opinion is much greater than
the ‘framing’ concept allows for.

176

Appendix C
Olivia Synthesis for Meaning - Multiple Perspectives from a Soft Resource
Olivia read the first comment on a Reddit page, “I don’t act against climate change
because there’s nothing one person can do = bullshit. You don’t act because it inconveniences
you.” Then, she said, “OK, so I guess inconvenience, and because they honestly don’t believe
that they can make a change,” while also writing “inconvenience” and “don’t believe make
change” in her notes under why people do not change behavior. She continued “And then
somebody here says no single raindrop believes it is to blame for the flood” reading a comment
from a different user. She read a reply to this comment, “No single raindrop is to blame for the
flood. It’s just a single raindrop” then commented, “Yeah, OK, I get what they are getting at.”
She read a second reply from another user, “I’d be interested to know how many of those
“raindrops” are actually civilians and how many of those raindrops are government-subsidized
industrial contributors, “ then said “Oh, oh, so not single raindrop is to blame for the flood
(rereading)…I’d be interested to know how many of those are actually civilians (rereading)…oh,
so… OK that’s kind of a hitting on the government here…” She read another reply from the very
first user “’government-subsidized industrial contributors are organizations of individuals…sum
of their behavior” and said “it’s the sum of their behavior that makes it bad or good.” She then
said “So…so…I guess sum of behavior” as she wrote “sum of behavior” near why people do not
change behavior in her notes. She continued reading the same reply “’Big Polluters’ wouldn’t
exist if those who worked for them didn’t support that behavior, ” then continued on to the next
reply, “Yes, that’s true, but government-subsidized industrial contributors are more able to get
special discounts…to externalize their environmental costs…civilian usage constitutes
something less that 10% of water consumption…” She then said “Oh, OK, so because maybe

177

people actually…individual people actually…it’s collective, and the individual person, or
civilians can’t actually make a change…maybe that’s what they are feeling,” while writing
“collective persons (civilians) can’t change” in her notes. This last note represented a summary
of several different ideas on the page. She read further in the same comment “Moreover the
industrial et al users get special discount rates. Even if civilians were especially wasteful, which
they aren’t, changes in the consumption rate would never have any real effect…” The, she read a
reply from yet another user “Where do you live? Most states in the Web have water rights,” and
said “Ah, that’s true, I guess why people would change their behavior, is if they have water
rights,” writing “water rights” under why people change behavior in her notes, and saying “so
again, this goes back to affecting them.” She read on further, from another new user “Being
morally superior is more important that getting results!” Then, she said “Ah, yes, so morals,”
writing “morals” under why people change behavior.

178

Appendix D
Olivia’s Path to Synthesis for Meaning based on Verbatim Notes
After first searching for [people’s opinions on climate change] and writing down just a
few notes Olivia said to herself:
I think now…normally if I wanted to know more about climate change, I would
probably go to Google Scholar, and look at...um…but again, Google Scholar I
think is just going to give me more scientific facts, and this things asking for
people’s opinions…and so do not choose, so why people choose to change their
behavior or not change their behavior (reading from the task). Ok, so I’ve learned
about some climate change, now I’m going to write in [why people change
behavior on climate change] (typed in as new keyword search phrase).
She selected the first article based on its Google results stub “How the science of
behavior change can help with sustainability” read through it, and took some notes based on
repurposing the information provided therein. As she left this page, she said “but this is totally
not about global warming.” Upon returning to the Google results, she read the next stub “How
the science of behavior change can help environmentalists” but did not select it, then read
“Putting” the first word of the next Google stub, and then read the full title to “Public Attitudes
and Behavior about Climate Change” exclaiming “Yes (with increased volume in her voice), and
this I see is by M Patchen, looks…Purdue…OK, looks like a good…” and then opened the PDF.
She went on to read through the table of contents “Summary, Introduction, Theoretical
Frameworks, Benefits, Emotions…Appraisals of the Situation, Personal Characteristics” noting
to herself “yeah, that… definitely, I can see that” and then finished “Social Influences.” She
added “This definitely seems like a PDF I want to read, but it is 58 pages long, and I only have

179

and hour, so I’m probably just going to skim it.” She skimmed the first five paragraphs of the
introduction “…effects that can cause serious harm to humans (from second para)…blah blah
blah Why do some people (from fifth para)…that’s right why do some people concerned but
others don’t care (comment linked to “care” in notes and from previous article).” Then, she read
in more detail the sixth paragraph “Since climate change has become a topic
of…discussion…(skips rest of first sentence)…pro-environmental behavior…support for green
policies…affected by a variety of factors ((parts of second sentence skipped…increased volume
on factors)…These include…” and then read the full third sentence and notes the list above
about these factors.

180

Appendix E
Olivia’s Insights Related to the Text(s) Read
First, as Olivia bounced from one idea to another in a long PDF, reading only section
titles and segments of first sentence, she jumped to a section of text about ways that people deal
with climate change and read “Some people have hope that new or future technology will save
them,” and then immediately said “so, I guess, maybe putting the responsibility on somebody
else” as she wrote “response – someone else – technology” in her notes about why some people
don’t Δ behavior.
Second, when reading through a list of reasons why people may not care about climate
change, Olivia read “Third, we’ve become distanced from nature, ” and said “yeah, I would
agree with that,” while writing “distanced from nature” in her notes. After her verbatim notes,
she then said, “so, we don’t value nature, I guess, as much,” adding “(value)” to the notes about
distance.
Third, in the context of exploring information about religious views on climate change,
she read the text “U.S. Legislators, backed by the Christian right vote again these issues with
near-perfect consistency. That probably doesn’t surprise you, but this might: Those same
legislators are equally united and unswerving in their opposition to environment protection.” She
then noted “opposition to environmental protection” as a reason people might not change. As
she did so, Olivia stated, “Could be a reason why…because you want less government, so maybe
your politics…” and then added “less government” after her previous note.
Fourth, Olivia read through an article about Religion and Climate Change. As she
skimmed the text, she focused in on a particular section and read:

181

As environmentally minded clerics, and greens with a spiritual bent, confer in
increasing numbers, in particular over climate change, acquaintances are being
struck that transcend many ethnic, ideological, a theological obstacles….And such
encounters are not just a curiosity. The might make all the difference to the
planet’s future…So these hymns…about faith…and the environment…
She then said, “So, they sing songs about the environment and faith,” and continued reading
(some of the text for a second time):
…and greens with a spiritual bent, confer in increasing numbers, in particular
over climate change, acquaintances are being struck that transcend many ethnic,
ideological, a theological obstacles….And such encounters are not just a
curiosity. The might make all the difference to the planet’s future…
Finally, she said, “Hmmm. So I guess it’s a way to make people passionate about something,”
and noted “passionate about climate” in the section of her notes devoted to religion.

182

Appendix F
Olivia’s Repurposing Examples
For instance, after entering the search phrase [why people change behavior on climate
change], Olivia selected the first result based on its title “How the science of behavior change
can help with sustainability.” She then read the subtitle “Les Robinson shares his tips on how
the science of behavior change can help make sustainability more effective,” predicting “so
how…basically…it seems to me though about how people are already interested in changing
their behavior, and this is about how to make their behavior more effective.” After skimming
and reading several headings in the article, she read a header “What if people just aren’t
interested?” and said “So, I guess, maybe if people don’t change their behavior with global
warming, maybe they’re just not interested,” and wrote “they don’t care, not interested” under
the heading “why people Δ no behavior.” After reading several more headings, she clicked
away from the page, saying, “but this was totally not about global warming.” As such, in this
case, she repurposed the idea of caring (or not caring) from the context of sustainability to
climate change. And, while it is arguable that these topics are largely similar, Olivia did not
think they were, and yet still repurposed information from the article to her evolving knowledge
about the climate change task. It is also worth noting that the search phrase she used ([why
people change behavior on climate change]) was intended to find sites about why people do
change their behavior, but she synthesized information explaining why they might not.
Openness of this sort to finding and valuing information that you might not be seeking at the
time is a characteristic of the CFT-W model (Spiro & DeSchryver, 2010) and may be particularly
relevant to finding information that can be repurposed.

183

Several times, Olivia used information found on the Web in opposite ways that it was
presented in the text. For example, as noted above, she read “most people are willing to make
some financial sacrifices,” and then commented, “yeah, that can be it, too, so not change
behavior…money…” Notice how the text is written in a way that indicates people will make
financial sacrifices related to climate change, but she repurposed this to explain why some
people might not. Similarly, as noted above, when reading through the Reddit comments about
climate change, Olivia read “Being morally superior is more important that getting results!” said
“Ah, yes, so morals,” and wrote “morals” under why people change behavior. Just thereafter,
however, she also said “so, some people, I guess, their morals aren’t strong enough
towards…um…aren’t strong enough (as she writes notes)…um…to their…um…hmm….to
support environment…hmmm,” as she wrote “morals ≠strong enough” under why people don’t
change behavior. Notice the slight differences between these two examples. In the first
example, the idea as written does not apply to either changing or not changing behavior. The
notion that people will make financial sacrifices to lessen climate change does not explain why
they might not. Olivia had to repurpose this as she did to integrate it into her knowledge
construction. However, in the second case, the text as written could (and was) synthesized “as
is” to contribute to meaning across texts, but was also repurposed.
Olivia also repurposed by scaling ideas down in her reading-to-learn task. For instance,
when reading an article entitled Trade Flows, Barriers and Market Drivers in Renewable Energy
Supply Goods, she scanned through the table of contexts, and then began skimming pages. After
looking at a few charts, she read a shaded area “This paper seeks to establish correlation between
production and export capacity in renewable energy,” then summarized it for herself “so the
ability to produce and export energy.” She then paused and said, “hmm…I don’t…produce and

184

export energy…how would this change people’s behavior? Some countries can produce and
export these goods…and I guess that maybe some companies can’t do this…so maybe that’s why
they [people] don’t [change behavior],” writing “companies can’t produce and export” under
why people might not change. Olivia repurposed this idea from international commerce to
individual behavior, but it is also important that she deliberately stopped in the middle of an
otherwise unrelated article to ask how the text might apply to her task.
A few minutes later, she did the same thing again, reading “Applied tariffs for all except
India…in the single digit range. Developing countries apply higher tariffs…” and said, “Oh, so I
guess if you come from a poorer country you might have a higher tariff on these renewables…on
renewable energy…so that’s…no wonder. That’s not really fair…” She then wrote
“pooreràhigher tariff renewable” under why people don’t change. In both cases, ideas intended
at an international scale were reduced and repurposed to explain individual behavior in ways not
intended by the authors.
Olivia repurposed another idea in two different ways. That is, after searching for [rich
people change behavior global warming] and reading through an article entitled Global Warming
and Social Justice, she skimmed text fragments from several paragraphs - “Questions of
corrective justice…our goal here is to answer those questions…to motivate…First, the world
taken as a whole would benefit…” and said “Yeah, we know that, but let’s see here…Oh, yes,
yes, yes…” She then read the first few words of a paragraph about carbon taxes “Suppose for
example, that the world…” and then said “That’s right…so a tax benefit…could make them…if
they do this voluntarily…they get a tax benefit…you know…if they switch over,” while adding
“tax benefit” to the her notes about how to change high SES people’s behavior. Two things
happened here. First, Olivia scaled the idea down from its international context (i.e., she was

185

reading about how the world as a whole would benefits) to individual behavior. The idea could
have been repurposed on this change alone, i.e., carbon taxes could be applied to individuals to
change their behavior. However, Olivia repurposed the idea yet again, essentially from a stick to
a carrot.
Finally, after reading through a list of “advantages of global warming” on about.com,
including “longer growing seasons,” saying “yeah, that would be good,” and adding it to her
notes, Olivia had an insight about this idea and combined it with a repurposed idea from earlier
in a layered generative synthesis interaction. That is, soon after reading the list, she said, “All
right, but I guess we’re looking at behavior here…um..but I guess there are advantages as to why
people maybe would want global warming, like if you’re a farmer, for example…so you
wouldn’t want to change your behavior, even though you know there are many bad things going
on with global warming, it just may be good for you.” This comment integrated an idea from an
earlier article when she read, “so many people willing to act on the negative consequences of
environmental change…affect them personally, affect a larger social unit, or affect the biosphere.
Perceptions of greater risk or threat…have been shown to increase people’s willingness [to take
pro-environmental actions]…” In this previous segment, she emphasized the importance of
“affect them personally” though increased volume in her speech and connected it to willingness
to make positive change. In her comment about the farmer, she repurposed this in the opposite
way, in that if something affects someone personally, they might not change their behavior in
positive ways. This was integrated with an insight about longer growing seasons providing a
specific advantage to farmers, making them act regardless of their knowledge of the larger
issues.

186

Appendix G
Olivia’s Reinforcement Examples
Reinforcing synthesis for meaning. Several times, ideas that Olivia synthesized for
meaning in her notes were reinforced by text encountered later in her task. For instance, seven
minutes and two resources after identifying “media you receive” as a contributor to individual
behavioral decision-making, the she confirmed it by a more reliable resource than the first time.
The first time she noted this issue, it was based on a blog titled Old man in a cave and the second
time it was embedded in a graphic titled A Model of Determinants of Behavior Relevant to
Climate Change in a sociology research paper from Purdue.
In another case Olivia read about how water rights might impact individual behavior
(detailed above from her review of Reddit comments) and connected this change in behavior
back to things that impact people personally by saying “so, again, this goes back to affecting
them…” This was about five minutes after reading in another resource (the Purdue study) about
how conditions that “affect them [people] personally” impact their willingness to act. Several
things happened here that demonstrated the emergence of a complex relationship among these
synthesis elements: (1) Olivia synthesized the idea of water rights for meaning in her notes; (2)
she had an insight about how water rights would affect people personally that was not explicit or
implied in the text, and (3); this insight reinforced the text read earlier. However, the difficulty
of designing models to fully deconstruct and explain this behavior is also demonstrated well
here. Is the insight Olivia had based on the text, since it was read relatively recently (i.e., are
these ideas still in working memory), or had she committed them to long-term memory, in which
case the insight would be based on background knowledge? That level of detail is not the goal of

187

this study, however, it is fair to say based on this data that the then current text itself did not
reinforce the idea of being affected personally, but her insight about it did.
In another instance, the reinforcement of an idea synthesized for meaning earlier
effectively clarified and simplified it for Olivia. When reading through a HuffingtonPost article
identifying reasons people don’t care about climate change, she read the first sentence of the
fourth idea, “…many of us don’t believe we can make a different in the our world,” and noted
“once person doesn’t matter ≠ enough make big difference.” This reinforced the idea of
collective vs. civilian impact she encountered earlier, but in a much simpler way. When reading
about this the first time, is was when she skipped from one comment to another on the Reddit
page, while the metaphor of the raindrops, a discussion of government-subsidized contributors,
and the example of 10% of water usage being civilian may have made the bigger idea more
difficult to understand. In fact, her notes based on the initial reading read “collective persons
(civilians) Δ.” This is somewhat ambiguous when compared to her “once person doesn’t matter
≠ enough make big difference.”
In that case, Olivia was able to make a complex interaction simpler by reinforcement. In
yet another, reinforcing an idea served to provide an additional level of useful complexity to the
first instance. When exploring an article about the opposition of Republican senators to
regulating CO2 emissions in Utah, Olivia read, “It’s not coming at us. If this is happening,
there’s very little scientific…” and commented, “…so again, they may not want to change
because they believe there is a lack of scientific evidence.” Given that the earlier interaction
about belief did not mention science or evidence, this instance may serve to reinforce her
background knowledge about science. However, her use of the phrase “so again” suggests that

188

this interaction with text was reinforcing for her, and did further explain the belief text she
synthesized for meaning earlier in the session.
Reinforcing insights. Olivia also experienced times when the text reinforced her own
insights. One of these insights occurred when reading about how some people hope new
technologies reduce the impact of climate change, when she stopped reading and said
“technology…I don’t know…its just…i don’t know if I’d rely on that.” As she continued to
read, the next paragraph stated “The study by Kempton and his colleagues found that most
people in all of the groups…rejected…” and then said “That’s right, other people agree with me,
not that my opinion is the only one that matters here, I’m trying to get a lot of people’s…but I
agree with that.” That the reinforcement happened so quickly may impact how significant the
effect of this interaction might be on long-term memory. However, this example provides two
important features about reinforcement. First, questioning the text can be considered a
generative form of synthesis in that it provides a value-added that is neither explicit nor implied
in the text. As such, the reinforcement that occurred here might have an impact on the
knowledge synthesis that Olivia experienced, but its secondary effect might be to encourage her
to question the text even more as she reads on. That is, the effect of questioning the text and
having the text (or even a later text) agree with your assertion might increase affect and be
motivating to continue to do so. If so, the affective and motivational component of
reinforcement may also be higher when generative forms of synthesis are reinforced, as opposed
to syntheses for meaning, when the reader is arguably less cognitively invested or responsible for
the ideas that are reinforced.
Another insight that was reinforced was that of money being influential in the individual
behavioral process related to climate change. As discussed above, the idea that money might be

189

a meaningful reason for people to change behavior was an insight Olivia had while synthesizing
a list of other reasons from the Purdue study for meaning. Thereafter, the text reinforced this
idea several times. First, it she encountered this idea again within the same article, in the context
of financial sacrifice, reading “most people are willing to make some financial sacrifices…yeah”
and then saying “that can be it, too, so not change behavior…money.” In this case, the
reinforcement of money as a motivator was not achieved until Olivia had repurposed money
from the current text to a reasons people might not make change, as discussed above. In this
way, the reinforcement occurred, but in a way that broadened her scope of application for
money. The issue was reinforced even more by reading about how “government-subsidized
contributors are more able to get special discounts...to externalize their environmental costs” and
then said “so then we go back to money…which definitely is an issue here.” The scope of
money was broadened yet again to include industrial financial implication on individuals. The
next reinforcement of money came later in the same resource, based on her reading of the
following comment, “I act again climate change because saving energy saves me money…” to
which she responded “ooh, saving money, why people change their behavior would be to save
money” while noting “save $” under why people Δ behavior. In this case, however, the
reinforcing idea lead her to elaborate even further. She said, “but I guess, others would say that
it will actually increase their costs because they have to pay more for these light bulbs, and so,
you know…for certain things.” The next instance both reinforced and further clarified how
money might be a reason for people not to make changes. In a HuffingtonPost article Olivia
read, “…it costs money…” and said “that’s right, that’s what other people have said.” She read
on

190

…so, I still drive and SUV, I bought it 130,000 miles ago when we lived were
snow could fall four feet in a weekend. I thought SUV…safest…now I live in a
place where it never snows…I have not reason to drive it, except I can’t afford a
hybrid and then said, “so, money, can’t afford a hybrid
as she wrote “$≠afford” in her notes. In this case, the text was actually reinforcing another
reinforcement (when she repurposed some text and said “that can be it, too, so not change
behavior…money,” noted above) making it more detailed. That is, the notes she took in the
latter case were a “$” in the why people no behavior section, with minimal connection to the
text. Where as, the second note was more specific and connected to a concrete example in the
text. Finally, when exploring government incentives for mitigation on greenfacts.org, Olivia
read “Climate change policies related to energy efficiency and renewable energy are often
economically beneficial,” commented “Oh, so they are often beneficial for money,” and noted
“long term beneficial to society – government.” In this case, the context in which the idea of
money was reinforced was at a much more sophisticated and global level than anything she had
yet encountered about money. As such, the text both reinforced and facilitated elaboration on
this important idea for her.

191

Appendix H
Olivia’s Creative Synthesis #1
When first presented with the reading-to-do task, Olivia started out with the following activity:
Right now, I’m just going to write down a few things I learned from the hour
before I go searching, or just things that I think I can come up with in
general…Some voluntary ways to get religious people to change their behavior
would be to…increase the value of mother nature as it relates to the bible or
whatever religion, Koran, or whatever religion it is. [noted ‘increase value mother
nature bible’]
I know they also suggested making some hymns, this one Web site talked about
making hymns…about nature and singing those types of hymns. [noted ‘hymns
about nature’]
Also, something religious people like to…a lot of times its a community…so we
can..they can do community things together that would persuade them to…so
community activities together that would persuaded them to clean up the
environment whether its to participate in recycling together…whether its to…just
do something as a community collectively, since they follow this group…it will
make them hopefully want to do it a little bit better…so there is that Web site that
said that. [noted ‘community: clean up enviro, recycling’]

Olivia continued this creative synthesis with ideas based on insights. She made several
insights, which appeared to be largely based on background knowledge that became the
foundation for her answer to the reading-to-do task. When first thinking out loud about how to

192

target college students, she said, “In terms of…um…the college aged students…a good way to
make some ideas to persuade them to volunteer to help lessen climate change would be to make
it cool and or put it on a social network,” and then noted “be cool, social networking.” She then
said
another way would be to make it fun, so, for example, it would be to take them to
other countries, possibly, maybe not by airplane to reduce that carbon footprint,
but I guess…somehow get them to other countries and to invest them…maybe by
traveling and volunteering in other countries to reduce the problems associated
and noted “fun → other countries travel volunteering reduce problems.” Finally, she said
So, college students, another way to get them would be to I guess, to make
different campus groups and I would still say that do fun things…allow them to
meet people…and allow them to feel good about themselves..because again I
learned that you may change if you feel it directly affects you…so maybe increase
their…them feeling good about themselves and that they are making a difference
collectively…
and noted “groups, fun, meet people, feel good about yourself.”
This was the first time that the ideas of social networking, fun, travel, or feeling good
about oneself emerged in either the reading-to-learn or reading-to-do session. Olivia did not read
about these in any text(s), or in this case, have insights about them based on text. However,
notice how the idea-play in this example emerged. She considered several different ideas
from the reading-to-learn session - cool was an insight in the first task, collective activity was
synthesized for meaning and then reinforced in the first task, being personally affected was
synthesized for meaning and then reinforced in the first task - which prompted several “new”

193

insights based on her background knowledge (social networking, fun, travel, feeling good about
oneself). As such, an argument can be made for another “downstream” influence of the Web
texts from the reading-to-learn session. That is, the ideas of “cool” and “collective activity”
together contributed to Olivia’s insight about social networking; the idea of “cool” contributed to
her insight about “fun;” and the idea of “being personally affected” contributed to her insight
about feeling good about oneself. As such, this creative synthesis about college students
demonstrated the complicated interaction between background knowledge, idea-play, and Web
text, and the difficulty parsing one from the other.

194

Appendix I
Olivia’s Creative Synthesis #2
This example from Olivia’s reading-to-do session is more explicitly scaffolded by her use
of the Web at that time. For example, just after the creative synthesis in Appendix H about
religion and college students, she said
and the high money making people, I think, how could we change their
behavior…I think this depends on if they care about spending their money or if
they don’t care about spending their money…because one way you could spend it
is…we’re going to do these things to try to save you money, so if you don’t need
to save money, I’m not really sure…I’m going to have to look that one up…
She then searched for [rich people change behavior global warming], selected an article
about global warming and social justice, and repurposed carbon taxes to tax benefits for the rich.
Thereafter, she continued reading the same article, “United States is not as vulnerable…to
serious losses from climate change…agriculture…United States might be a net loser from…a
world wide carbon tax…” and said “Or, one way you could make…I guess…um…let’s see…”
She skimmed more text segments from several paragraphs, “We accept the view that in many
domains resources should be redistributed from rich nations to poor nations…To understand the
issues of justice and motivations…United States…China…top emitters…” leading to the
following insight, “So…they care about money…because they are rich…they probably also care
about their family.” After skimming further, Olivia read the following
From the standpoint of distributive justice, it is temping to suggest that because of
its wealth, the United States should be contributing a great deal to the climate
change problem…far more that would otherwise be its fair share…Let us simply

195

stipulate that it would be a good idea for wealthy people in wealthy nations to
transfer resources to poor people in poor nations. Even if this is so, the claim runs
into three problems…the first problem is that emissions reductions would help
future poor people rather than current poor people.
and then said, repurposing this information, “Oh, so I guess…maybe…if rich people are
caring and giving they could give to future generations of the poor people,” and noted
“giving: future generations poor people.” She continued reading
If the goal is to help the poor, it is odd for the United States to spend significant
resources to help posterity while neglecting the present…The second problem is
that emissions reductions are an in-kind benefit. Poor people in poor nations
would in all likelihood prefer a cash transfer…perhaps India would prefer to
spend the money on education or AIDS prevention…The third problem is that
many of the beneficiaries of emissions reductions are wealthy, and many of the
losers are poor. Wealthy people…Global warming will also produce monetary
benefits in many places like productivity…
Olivia then verbalized an insight about the importance of determining investment
So, I guess just thinking about this then…the question to me comes if I want to
change the behavior what are rich people invested in? And, whatever they are
invested in we need to target that. So they are invested in money, so we can give
them a tax benefit…if they are invested in helping other people, they can give to
future poor people by trying to reduce their emissions now…
Her focus then shifted as she had the following insight

196

If they are invested in the typical Hollywood type ordeal…and…I
guess…fame…and recognition…we could say that if they give, there
will…depending on the amount, be a public plaque, or the bricks in a park…and
they would get their names on the bricks..depending on how much they give, or
on a fountain…or on a birdhouse or something, depending on how much they
give, whether its money, or whether its time…or a that they put in to trying to
voluntarily change global warming or climate change…so give them some sort of
incentive…
while noting “Hollywood/Fame/recog: Give→plaque - bricks...in a
park...name...fountain...birdhouse...$.” She then repurposed several ideas saying “so, college
students too...we could also reduce the cost of meals, for example if they learn to recycle…so I
would definitely say incentives,” while noting “reduce cost meals→recycle.” She continued by
repurposing this idea again “So, for religion, the reason why, the incentive would be to do it for
God or to do it for your community…more of a moral obligation for those reasons…and then
money...I’d say, yeah, depends, on what they’re interested in...” She had another insight “College
Students…well. They want to be cool, save money, have fun…and they also want to learn,”
while noting “be cool, save $, have fun, learn.” Olivia continued, saying “so I guess you could
make a class, which I have not taken, on the effects of global warming…or we can try to start
integrating concepts of global warming into multiple disciplines, such as math science,
sociology…” and noting “class on effects global warming...integrating global warming into
multiple disciplines.” Finally, Olivia read through her notes “So they can take a class…reduce
cost of meals…they can go someplace…” said, “actions that they can do that would be fun…”

197

continued reading her notes, “fun other countries…travel in other countries…groups, fun, meet
people…feel good about yourself…reduce cost…” and, said
Oh something cool would be for college students would be to design exercise
equipment that runs off of energy, your energy that your produce during exercise,
so like and engineering class…so they could make projects…in different
classes…that focus on global warming and reducing and changing behavior…
before noting “design exercise runs off of energy produced during exercise” and “projects in
class.”

198

Appendix J
Olivia’s Creative Synthesis #3
Another example from Olivia’s reading-to-do session demonstrated the multiplicity
creative syntheses exhibit in a different way. In this case, the different elements of synthesis
were mostly acting in concert as she processed the same Web text chunks.
Beginning just after creative synthesis #2, Olivia said to herself “I had read something
earlier that talked about how to get people to change behavior” which lead to a search for [how
to change a behavior voluntarily]. As she skimmed the results, she read a title “Public
Perception of Climate Change: Voluntary Mitigation,” commented, “so I know mitigation now
means lessening...” and selected the result. The ensuing page was an abstract from the American
Journal of Preventative Medicine. She read, “...mitigating…not only
government…consumers…population…awareness…behavior…data…subject…awareness
…vast majority reporting some level of concern…a multivariate analysis revealed significant
predictors of behavior change…individuals with heightened concern…” She continued reading,
and said
so we want to...so in order to get individuals to change behavior…you
know…voluntarily…again, we are going back to what I had read earlier…that
they need to be concerned…oh, it’s better if they have more education about
it…younger tends to do it…and let’s see…Of those who changed
behavior..reduced energy usage, reduced gasoline…oh, so if more people are
likely to reduce something…what they can do is reduce energy, reduce gas, and
recycle…are the tops ones…top EASY ways, this is easy, people like easy, easy
way to change behavior…

199

While doing so, she noted, (1) concerned (2) education (3) younger
recycling and labeled the last three as easy ways to

energy

gas

behavior. In so doing, Olivia

demonstrated three different elements of synthesis. First, she synthesized the ideas about
education and age from the current text for meaning. Then she had an insight about how the
ideas in this text were examples of easy ways to change behavior. Finally, she reinforced the
idea of caring, which had been originally from text in the reading-to-learn session, then
reinforced several times, and finally repurposed only minutes before during the reading-to-do
task. She continued reading, Government policy must eliminate economic, structural and
social barriers, and said yes there are barriers, we are aware of that...so the barriers. She then
noted barriers, reinforcing the idea of barriers to climate change behavior from Web text in her
reading-to-learn session (text from Trade Flows, Barriers and Market Drivers in Renewable
Energy Supply Goods which, at the time, prompted insights, and was later repurposed).
Before reading on, Olivia revisited this information in the context of her different groups
of individuals and said
so, I guess for religion…we can encourage [reading her notes] reduction in
gas…reduction in…encourage them to care more...So, I guess reinforcing the
Earth, and why God…or whoever…or a higher being…created
earth...paganism…if they like nature…that may be important
while she noted “earth why God/higher being earth” and “paganism → nature...” In this activity,
Olivia reinforced the idea of “creation care” from her reading-to-learn task and synthesized the
idea of pagan values about nature for meaning from her reading-to-learn notes. She then said
so for people who are rich…how to get the them to save…reduce energy…reduce
gas…ooh, hybrid cars…super cool...hybrid cars…convenient…and cool…and
200

unattainable to what others may have…assuming they want…assuming that they
are rich and they want something unique
while she noted “hybrid cars, convenient, cool, unattainable, unique” in her notes about high SES
individuals. These interactions demonstrated how Olivia had an insight about the impact of
unattainable or unique goods on this population, repurposed an idea about hybrid cars from the
first session (i.e., she read about how hybrid cars are too expensive for many and might have
been a reason for people not to change their behavior), repurposed cool from idea-play about
how to target college students earlier in the reading-to-do session, and synthesized the ideas of
reducing energy and reducing gas for meaning from the current Web text.
Before moving on, Olivia continued looking at her notes and said “and, for all of these
things, uh...I’d say you need to make it easy, same with...[rich] people...easy...I think religious
people would do it even if it weren’t easy because the are doing it out of different reason...” In
this case, Olivia elaborated on her original insight about “easy” from the text, and provided even
more value to it (i.e., that it may not be necessary for the religious group), and demonstrated
what might be considered an insight about her own insight. This generative synthesis about her
own insight was specifically prompted by a review of her notes.
Continuing on, Olivia visited a Google result about behavior change in Africa, she
decided, “no, no, I don’t want that” and revisited her search phrase. She adapted it and searched
for [how to change climate change behavior voluntarily], selecting a result titled Voluntary
versus Mandatory Approaches to Climate Change Mitigation. While skimming the first two
paragraphs, she read “...voluntary programs are best understood as a weak instrument...” and said
“so they are weak, so I wonder how to make them stronger...so if voluntary is generally
weak...compared to mandatory, I wonder how to make voluntary programs...to maybe enhance

201

the efficacy...but I’m going to keep reading anyway...” while noting “voluntary → weak, how to
make vol program, enhance efficacy.” In questioning the text in this way, Olivia essentially
made an insight, or generative synthesis of the text.
She read on in the article, and bounced from one header to the next for almost two
minutes, at one time saying, “This is not really what I thought it was.” However, one particular
section caught her attention, entitled Clinton, the Carbon Tax, and the Climate Change Action
Plan. She read, “When the administration presented its climate change action plan later in the
year, the focus had shifter from mandatory regulations to financial subsidies (including $200
million per year to stimulate the adoptions of more energy efficient technologies)” and said, “Oh,
so money…so you get money…that could be another benefit, get money…if adopt behavior, and
this is used in the government” and, noted “get $ if adopt behavior” in her college section. In
doing so, she repurposed the notion of government subsidies to individual behavioral payments,
at the same time reinforcing the multiple instances of money she encountered in the reading-tolearn session.
Continuing on in the same paragraph, Olivia read, “Among the voluntary programs were
Green Lights, Climate Wise, Motor Challenge, and Energy Star Buildings. Firms participating
in these programs received several benefits. They were given case studies detailing the cost
savings…” and said,
Yes…so that would be a good thing…I guess to continue to show the cost
savings…which they do at Michigan State…they show the cost savings in each dorm and
percent of saved, and for instance the amount of school busses each dorm or apartment
complex saves each year in trash, electricity…you could make it a competition…

202

and noted, “cost saving dorm/% saved among of school busses each dorm, apt, saves each yr,
trash, elect…” This particular text, combined with Olivia’s background knowledge about
Michigan State, facilitated three generative syntheses. First, she repurposed the notion of cost
savings from federal programs to dorm activities. Second, she made an insight about using
competition to make this work. Third, the entire interaction reinforced yet again the importance
of money in considering behavior regarding climate change.
After skimming the rest of this article without much interest, she immediately changed
her search phrase to [improving efficacy of voluntary behavior change], a direct result of her
questioning the text about weak voluntary program just prior. After looking at one result about
health behavior change, she returned to her search and said “um...improving...um...maybe take
out efficacy and put in climate change” and changed the search to [improving of voluntary
behavior change climate change]. She then selected a PPT about “empowering the public to
voluntarily change behaviors...” and it opened to the title slide “Water Stewardship Program: A
Success Story in Behavior Change and Public Participation.” She skimmed the first several
slides, reading keywords from each very quickly and making occasional comments,
systems point of view...empower the public to change behavior...25 year research
tools on community empowerment...15 year research applied to
environment...dozens...blah, blah, blah...tools addressed to address range of issues
including non-point source water, air pollution...empowerment model...social
marketing...where do I start...the barriers...carefully crafted actions...blah, blah,
blah
before slowing down on a results slide. She read this slide, “20,000 people achieved the
following savings, 40% reduction solid waste, 32%…” and said “using this program

203

methodology.” She backed up one slide to the “Household EcoTeam Program Methodology”
slide and read part of the first bullet point “distinct topic areas.” She then said, “we will need to
make each goal…maybe…distinct goals…and with distinct areas…easy actions…support group,
hold each other accountable…self-directed meeting…trained volunteer coach…they’ll have
somebody they can go to…” while reading text on the screen with the following bullet points:
•Carefully crafted actions in easy-to-use recipe format
•Peer support group of 5 to 8 household
•Self-directed meeting guides
•Trained volunteer coach
During this time, she also took notes in the margin of her paper for “goal dist, dist area,” easy
actio,” “support group hold account,” “self-dir,” and “coach.” In so doing, Olivia repurposed
these ideas from water stewardship behavior to climate change behavior, and also reinforced her
ideas about easy (an insight she had based on Web text earlier in this task). Finally, her
comments on this text provide further evidence of how inference and summary are incorporated
into the synthesis process. That is, her comments about how support groups hold each other
accountable was an inference she made about the meaning of “peer support.” Also, her
comments and notes in this example all represent an in-situ summary of the bullet points she was
reading. In these ways, inference and summary may also be used when repurposing Web text.
As she moved through the PPT, she continued reading small segments of text from
different slides, saying to herself “what is this program here, because I need ways to persuade
people, and want the most effective ways.” This commentary represents her continued and
consistent adherence to the task, but also how she has synthesized some of her new ideas to
guide the exploration. Specifically, she is making reference to an insight she made about eight

204

minutes prior when reading about the weakness of voluntary behavior change programs and how
to enhance their efficacy. Just thereafter, she read the selling points for neighborhoods from the
current slide “environmental improvement, neighborhood livability, community building,” and
repurposed the first idea for college students, “...so for college students, we would want to
improve environment,” noting that in the appropriate section. She then added “same with...I
guess really for everybody...we would want to improve the environment...no matter what’s the
group.”
She quickly clicked by a few more slides about social diffusion, research questions, and
funding of program, the program design, before stopping on a slide and reading the first bullet
about lifestyle assessment. She read “lifestyle assessment,” and said, “Oh, so, OK, so they give
everybody an assessment...so they know where to begin, where to go, “ and noted “assessment.”
In so doing, Olivia repurposed this idea from water stewardship behavior to climate change
behavior, and also reinforced the idea of empirical assessment that she synthesized for meaning
earlier in her reading-to-do session when reading about federal programs.
One slide later, Olivia read “A less measurable feature was the social capital
built…program participants met their neighbors,” and said “Ah…yes…so enhance
social…cohesiveness….so enhance cohesiveness…they were engaged...and this could work for
all three groups,” noting “enhance cohesiveness, engaged.” Again, she repurposed this idea from
water stewardship behavior to climate change behavior while reinforcing several ideas from
earlier (e.g., groups, fun, meet people, and social networking were all insights from her initial
reading-to-do idea-play and the idea of collective or community action had been introduced and
reinforced several times prior in both sessions).

205

Thereafter Olivia finished skimming the PPT, and said “Let’s see here...so I want to exit
out of this...and I’m going to go here [back to her Google results]...to...let’s see here...” and then
re-read the task instructions out loud. She continued “I’m going to just type in then, I guess...”
and looked at the same results again, before finally returning to the Google search box and
slowing saying and typing in “climate” before deleting is and then saying/typing “maybe...fun
projects for climate change.” In so doing, Olivia switched gears and effectively ended the
creative synthesis series of events related to her first search about changing behavior voluntarily
approximately 15 minutes prior.

206

Appendix K
Profile for Beth
Prior to the study, Beth provided the following information about herself in a survey of
Web learning. She reported using the Web for more than four hours per day on a typical day,
and for three hours per day for learning. On a likert scale of 1-5 (5 being the most), she reported
that she like to use the Web for learning tasks at a 4 level. She reported having recently used the
Web to learn about probability theory, the Fed’s intentions on quantitative easing, the Cholera
outbreak in Haiti, and the symptoms of a sinus infection. She reported using bookmarks,
emailing URLs, printing articles, posting articles to Facebook and Google news alerts to
organize and manager her Web resources and was unfamiliar with the concept of tagging. When
asked to provide possible search phrases to learn about the effects of climate change on global
poverty she listed: [climate change and global poverty], [UNDP climate change], [Millennium
Development Goals and global poverty], [UN global warming], [Al Gore global warming
poverty], and [climate change scientist global poverty]. She noted that choosing keyword search
phrases was easy for her and that she
[thinks] of things I associate with the main topic and use them to expand my
search. If I don’t know anything about the topic, I run the first search, look for
associated ideas, and then expand or narrow my search based on what I find.
She noted that “aha” or “light bulb” moments when learning on the Web usually came to her
when “finding solutions to math homework.” On a scale of 1 (beginner) to 5 (expert) she rated
her abilities in the following categories:
Finding information related to your studies
Organizing/managing the Web information you find that
is relevant to your studies
Determining the reliability of Web sites
207

4
3
4

Determining the relevance of Web sites to your specific
tasks
Determining the bias/stance of Web sites that you use
for learning
Synthesizing information from the Web into coherent
solutions
Using the Web to construct creative solutions to
problems
Using the Web to have “aha” moments of conceptual
breakthrough
Setting goals in your Web learning tasks
Staying focused when using the Web to learn

5
4
4
3
3
2
2

At the beginning of the first research session, Beth was asked to talk about everything she
knew, her background and experiences, on the topic of climate change. She responded:
I know it is a hot political topic, so I try to stay away from it. I know that there
was a big thing with scientists forging their results about it about a year ago. I also
know that a lot of the right, politically, does not think it is real…and the left
thinks that climate change is real and we’re all going to die. I know Al Gore is
really big in the climate change awareness stuff and I saw him at LiveEarth when
I was in Washington D.C. about three years ago…They run those commercials
about the polar bears not having any ice, and I had someone from Greenpeace
come up and tell me that the polar bears were going to run out of habitat unless I
donated money. I know back in the 14 and 1500’s there was a period of time
when it was warmer than normal, which lead to a lot of population growth, which
also I think might be part of the reason we had the industrial revolution, but I’m
not entirely sure…and…um…I think it would be silly to think that there is no
such thing as climate change and we are not effecting the environment, that we
can just spew crap into the environment and its not going to change anything. On

208

the other hand, I think the doomsday stuff might be a little overzealous. I also
think that the one good thing about awareness is that it is going to make us have
more green technologies, which will lead to economic growth…but that’s my
economist talking…You hear a lot about all of these animals that were going
extinct when I was little, like the bald eagle and stuff, and due to conservation
they’ve come back. So, hopefully, that’s something we can kind of keep doing,
even as climate change removes habitats, not let species go extinct…It’s one of
those things people will kind of like joke about…that sounds bad…but, its one of
those, it’s like politically polarizing…so, the extent to which I discuss climate
change is making…jabbing at other peoples political values...

209

Appendix L
Beth’s Synthesis for Meaning Examples
For instance, when first reading about recycling, Beth skimmed an Ohio DNR page. She
read“...benefits of recycling...reduces waste...recycling is a big business in Ohio...also supports
the many companies and employees doing this work,” said “so, they are actually making an
economic argument for recycling...” and noted “Recycling can be a business” while commenting
“although, I’m kind of disagreeing ‘cause a lot of stuff isn't cost effective to recycling.” Though
the synthesis for meaning is simple in this case, Beth, like Olivia, did engage in high level
reading skills. In this case, she utilized summary of the main ideas (i.e., that they are making an
economic argument) and questioned the text.
Beth also spent time skimming an article from the US News and World Report citing
research about why some people go green and others don’t. She read
some review out of Stanford said a third of US consumers are ready to buy green
products...um...they aren’t sure what makes someone a green consumer...let’s
see...book argues that political affiliation has very little to do with consumer
decisions...um...and its largely determined whether you consume green products
based on personality traits, not political beliefs...(she says “now that’s
interesting”)...um...now let’s see...people bought considered their personalities
and habits...each survey questions...(“they wrote a survey”)...they asked people
questions about whether or not they slow down for yellow light to see how risk
averse they are...blah, blah, blah...about personality...Idea consumer...green
consumers not only buy stuff, but they also buy ideas...most consumers daily
purchases aren’t motivated by ideology...um...let’s see...buying a Prius often says

210

something about the buyer that he or she is concerned about gasoline
consumption...for green consumers, that ideological benefit of the purchase often
outweighs the practical benefit...um all right...certain intangible parts of a
purchase are more important than the product practical use...huh...OK...so
altruism...people who donate blood are regularly give to charity are more likely to
be green consumers...um...time minders...they balance short-term and long-term
costs and benefits...yeah..all right...um...[clicks to next page]...green consumers,
they tend to think much more about the future.
after which she said “so reasons to behave green” and noted “more future oriented.” She read on
“Me tooers...busy restaurant...indicates...if other people in their network go green, they’re going
to do it too so they’re part of the team,” and noted “other people are also doing it.” In both cases
of note-taking, Beth synthesized meaning from the ideas directly stated in the Web text. Of
interest here is also what Beth did not explicitly synthesize for meaning. The idea of altruism is
a powerful reason for people to “go green” or act in ways that might lessen climate change, but
she did not directly indicate it was important to her or synthesized for meaning. Similarly, she
“overlooked” the notion that broader personality characteristics are indicative of green behavior.

211

Appendix M
Beth’s Insight Examples
First, Beth searched for [how to lessen climate change] and then selected an article titled
“Reduce Effects of Climate Change.” It was a paragraph of text followed by a list of twenty
items. Beth read
change light bulbs, already read that...use alternative methods of transportation,
such as walking riding a bike, purchasing a hybrid vehicle or car pooling...design
your own solar cooker...right...um oh, apparently my oven is horrible and energy
wasting, this web site sounds biased...stop purchasing bottled water...eat less
meat...give rid of the excess concrete and plant trees...move into a smaller
house...wash laundry in cold water and hang it to dry outside...that’s not practical
in Michigan in the winter...start a home business...pay your bills online
She then said, “well, people do that because it saves money and because its easier, so it’s easier,
pay bills online” while taking the notes “it’s easier (pay bills online).” Before returning to the
article, she looked over her notes and said “...lets see...reasons not to go green might be it
changes stuff too much, which isn’t a good way to put that, but people don’t like change” while
noting “it changes stuff too much (people don’t like change)” and saying “which wasn’t actually
on any of the Web sites, it’s just what my brain came up with.” Beth returned to the article and
read “purchase local food,” saying “that’s just yummy...it’s yummy, is that a reason to go green?
sometimes green products are higher quality...you can feel good about” while noting
“higher/quality/you can feel good about it.” She then said “personal opinion, not actually like
from the Web site is people do this kind of stuff because they feel good about it.” Beth returned
a final time to the article and read “Live a simpler life...be creative with your trash...turn soup

212

tins into pen holders...give green gifts...use white vinegar for some of the cleaning” and said “I
do that, I didn’t realize it was a green habit...but, reason to do it...less chemicals..safer for pets”
as she noted the latter two ideas.
In a similar way, a shorter list of ideas helped Beth make an insight later in the reading-to-learn
task. In this second example, Beth searched for [can I lessen climate change] then selected a
Nature.org page entitled “Climate Change: What You Can Do.” She read one sentence from the
first paragraph “small change in our everyday lives can make a big difference,” and then the first
sentence two paragraphs later, “climate change threatens natural communities and human-well
being.” Beth then scrolled and read a list of “Easy things you can do...travel
light...teleconference instead of flying...” before she switched back to her notes and said
Reasons people go green...is it reasons to go green or reasons people go green?
New thought...people may accidentally do things that help lessen climate
change without that being the motivation...i.e. hold teleconference because it is
cheaper than flying to meet customer...motive is still profit not being green.
After noting this insight, she continued reading through the list, “recycle and use recycled
products...inflate your tires...plant native trees...turn down the heat...act globally eat locally.”
Then, she switched to her notes and said “reasons people do not go green...they value diversity of
choices, i.e., buy imported buy imported non-local food over buy local.” Neither the notion of
accidental green behavior or valuing diversity was explicit or implied in the current or previous
Web text(s). Like the list discussed above, the one is not about “why” people make decisions to
do these activities, and that slight difference from her topic may have promoted insight more that
an article with “the answer” to her question.

213

In a third example of insight from Beth, she skimmed an opinion article on
associatedcontent.com from Yahoo, reading “going green meant growing vegetables, turning
off light when you left the room. Now you have to have cleaner burning fuels...” She scrolled
down and laughed, saying “Her first question is what will you use to line trash cans? She says
she switched to reusable grocery bags and then she realized she didn’t have small trash bags
for the bathroom trash can.” Beth read on, “How strong are you really? Ah, reusable shopping
bags are heavy...this article should be why you shouldn’t use reusable shopping bags...How
much laundry do you want to do? um...cloth napkins an placements.“ She then said “let’s
see...pros and cons...reasons not to go green...” and noted/said “unsure tradeoffs are worth the
hype (i.e., use more water by using cloth instead of paper may not be worth it if you only use
paper rarely).” Though this insight results from reading about cloth napkins, the concept of
tradeoffs was traced to several ideas from previous resources. When discussing and
repurposing ideas about money only ten minutes earlier, Beth identified both that money can
be saved by going green and be a reason people don’t (e.g. because organic foods are more
expensive). And, only a minutes prior, she had read the list of 20 ways to reduce the effects of
climate change, many of which could involve tradeoffs (e.g. eating less meat, living in a
smaller house, purchasing local food, taking shorter/colder showers).
Finally, similar to Olivia, Beth also experienced insights in the middle of synthesizing other
information for meaning. As noted above, as she read a Washington Post article, Beth
synthesized information about fads and morals from the Web text. At the same time, just after
taking her notes about the faddish nature of green behavior, Beth stayed on the notes page and
added a repurposed idea about economic growth. She returned to the Post article for one
second before saying “Oh, another reason to go green...for future generations...although

214

nothing’s actually said that, I don’t know where that came from...I think he’s going to argue
that, though,” as she added “for future generations” to her notes.

215

Appendix N
Beth’s Reinforcement Examples
When Beth first encountered the idea of changing light bulbs to save energy, it was as
part of a long list of ways to save energy. Her reading strategy on the page was primarily to read
the paragraph headers and then move on, such that in reading the text in Figure 4, she read “seal
up the house” made some comments about how she should do that in her own apartment, and
then read “avoid chores...change light bulb.” In reading only the headers, Beth reduced the
potential benefits of context, perspective, etc. To make things worse, the second time she saw
information about energy saving light bulbs, the text was in list form without any supporting
text. When she saw this idea the second time, Beth just said, “I’ve already read that.” The third
time she saw this idea, it was again structured like figure 5, with a short bolded title and text
following each bullet. Her response here was much the same as she read the headers “change 5
lights...look for energy star products...heat and cool smartly...” and then said, “these are all the
same.” However, upon closer examination of the text after the “change 5 light” header, it talked
about how if every household in the US changed five bulbs, greenhouse gasses equivalent to 10
million cars would be reduced. The new perspective of collective action was available and might
have lent more weight to the idea of energy saving light bulbs for Beth, but her reading strategy
did not allow for this.
She also reinforced her own insight about people going green for “future generations” in
an article later on when she read “to ensure the bright future of our children.” She commented,
“see, I already had that one” with a positive affect that may have been beneficial motivationally.
Similarly, she read the item “to reduce pollution” and said, “I already talked about chemicals.”
However, in both cases due to the list-like structure of the new text (see figure 5), very little in

216

the way of new context or perspective was immediately available. What is different about this
list is that each item is itself a hyperlink. Both the link about the bright future for our children
and reducing pollution linked to much longer articles about those topics that did provide new
contexts and perspective that could have provided that these opportunities for reinforcement that
was elaborative and generative.
Later, when reading about the top five reasons to go green besides climate change, she
read “you save water...water is no longer cheap” and commented, “they are making the economic
argument again.” This provided a reinforcement of the “economic argument” in a new context
of water prices. Similarly she read later on from the same page “recycling benefits...what are
the changes we will need...clear up your storage and make money off recycling...it also means
that the product is being constantly utilized reducing carbon footprint.” Thereafter, she
said/noted “can make money/decrease stress by getting rid of unwanted stuff.” In this latter case,
Beth’s ideas about recycling were reinforced, but also lead to an insight (about decreasing
stress). The structure of this page was the same as those in figure 4, bolded section titles
followed by a paragraph of text for each item. However, for both water and recycling on this
page, Beth read beyond the header and was able to facilitate elaborative reinforcement and even
an insight, by reading more deeply into the Web text.

217

Appendix O
Beth’s Creative Synthesis #1
One of the differences between Olivia and Beth’s reading-to-do tasks was how each
started out. While Olivia did not use the Web for almost five minutes, Beth started out by rereading the task instruction, saying “luckily I just read a whole bunch of ways to do that, for the
most part” and then transitioned right away to a search for [how to lessen effect of climate
change]. This was a search phrase she used in the reading-to-learn task, and led her to select an
article listing 20 things a person can do to reduce the effects of climate change that she had
visited earlier. She scrolled up and down this list for 10 seconds without saying anything,
returned to her notes, and then commented,
So, for college students, I personally think the best angle is something along the
lines of like ‘it’s good for you too.’ Going back to our earlier conversation, you
know, they’re most likely to accidentally do it, by saving money. It doesn’t have
to be expensive...(wrote “it doesn’t have to be expensive” in college student
section of reading-to-do notes)...individuals with strong religious affiliations...I
think it’s easier to go with (switched to notes from first session) the...see that’s
hard cause everyone already does that it’s God’s kingdom or whatever
stuff...OK...um...let’s see...(scrolled up and down in her notes from the reading-tolearn task and read)...for future generations...(switched back to notes from second
session and wrote “for future generations” in the religious section)...cause going
back to that one article I read...if I can find it (looking through history of
browser)...there was that one article about altruism...and (clicked on show all
history for browser)...um...(read titles from history)...why I won’t not go

218

green...peer pressure plays a key role...it was this one, the US news and world
report...why some people go green and others don’t...um...edit find...I don’t
remember if this actually talked about...it didn’t actually talk about (searched the
page for “relig” and it was not found) religion...(change page search to
“Christian”) nope it didn’t actually talk about Christianity either...and yes I know
I’m assuming by religious you mean Christian...um all right...let’s see...metoo’ers...I need to go back a page (history link in browser had taken her to the
second page of this article)...(as the first page loaded) so I feel like the religious
people...hurry up Internet...hurry up Internet...(after 15 seconds the page did not
lead and she returned to her reading-to-do notes)...OK...um...the future
generations...(read from the task instructions) hear any thoughts you have...your
ideas should be new and creative...they should be backed up with some
evidence...potential to be successful strategies...so evidence...this Web site that
won’t load...oh (as she recognized it is now loaded in another tab)...there we
go...all right...so it talks about why people decide to make the decisions they do
(while copying and pasting the URL from this article to her reading-to-do notes
under “evidence”)...um...all right...back to this article...OK...so it’s based more on
personality traits than political belief...so...idea consumers...green
consumers...altruism...I don’t know that religious people are necessarily more
altruistic than other people...all right...um...so I feel like that’s a
generalization...but...(returned to her reading-to-do notes) for future
generations...call on altruism...(added “call on altruism” to the same line as future
generations in religious section)...college students...it doesn’t have to be

219

expensive...making environmentally decisions can save money (said and added to
notes under college students)...I’m stuck on this high socioeconomic status
thing...all right...let’s just stick with the other two...see, I don’t know, I’m arguing
with myself...cause strong religious affiliation doesn’t have to mean Christian, so
you don’t want to, like, quote the bible or anything...but then there are all those
Bible verses that are good for this...but then, it’s like, I don’t know…all
right...back to college students, we’ll do that...um...future generations (noted
under college students)...chance to make a difference (noted under college
students)...could be this generation’s social movement (noted under college
students)...that idea came from the Washington Post civil rights type
article...which I need to find...history...show all history...um...Washington
Post...Washington Post...going off that because I can’t find it (clicked on a
different article from the reading-to-learn task about peer pressure)...peer pressure
plays a key role...(scrolled up and down)...so...um...greenwash, I didn’t see that
last time I looked at this...the idea of sustainability can be a remarkably effective
way of shifting patently unsustainable goods (from text)...what am I doing...(read
from task instructions) creative ideas about how to persuade...(clicked back to
session two notes)...so...(read from notes)...chance to make a difference...for those
with high socioeconomic status...going back to my pros and cons (switched to
notes from first session, scrolls to “reasons to go green” section)...OK...so...I’m
going to pull on something that I didn’t read on the Internet...which is like the
idea that people want to be like kind of near the first to adopt new stuff, and have
new stuff, if that makes sense...so...in some way you have to make it like be a

220

trendsetter type thing (noted ‘be a trendsetter’ in high SES section)...um...it’s
responsible (noted ‘it’s responsible’ in high SES section)...um...(read from
notes)...for future generations, call on altruism...(switched back to reading-tolearn notes)...all right...Google...(opened Google page)...

221

Appendix P
Beth’s Creative Synthesis #2
Beth’s second creative synthesis occurred during a period of time between minute 10 and minute
20 in her reading-to-do task. During this time she started by searching for [marketing green
products to upper class], but expressed her disappointed in the results. As she looked at the first
page of results she said “this isn’t helpful...plus I want creative ideas, I don’t want to read what
everyone else has done...” The first selection she made was based on a title “who are the green
consumers and what to they want?” which took her to essaymanina, a place that writes paper for
students, about which Beth was not happy. She then selected an article about “green branding,”
which was a research paper she deemed too long to read. Her final choice from this search was
for the article “How to Read Green Consumers - Using Psychographics to Define Your...” which
took a long time to load. After waiting 17 seconds, the page had still not loaded, and she said
“all right, going back to marketing plan” (her reading-to-do notes). The following creative
synthesis then occurred
All right, so (said/wrote) What I see for the religious people, is a poster of
something with beautiful landscape…and…something implying God’s planet and
then somehow connecting this poster with the idea of preservation and perhaps a
list of ways to make more earth conscious decisions...enforce the idea that making
sustainable decisions will be good... connect with possible bible verses and just
make making good decisions re: climate change “feel good”...um...college
students...what I see: not emphasizing that everyone is doing is, make it a
personal decision type thing...imply that its good, but focus more on economic
aspects, saving money, and ability to bring about change/help others...(read) a

222

change to make a difference...(said/wrote) help poor who will be most
affected...(read) be a trendsetter, be responsible for the high SES...(said/wrote)
emphasize plight of those in poverty...(said/wrote) Make it seem charitable and
visible...(said) my biases are showing here...so... (said/wrote) what I see, simple,
classy, emphasize how a small change can make a big difference...don’t
recommend moving into smaller house...(said) like that one Web site
suggested...emphasize health benefits...that could probably go up here in college,
too...emphasize health benefits...I don’t know why I don’t think religious people
want their health benefits emphasized, but I don’t...(switched to reading-to-learn
notes, read)...reasons people don’t go [green]...counteracting those...um...let’s
see...(looking over reading-to-learn notes)...it changes stuff too much...emphasize
small change can have a big impact...(said/wrote in SES section of reading-to-do
notes) small changes can have a big impact...(back to reading-to-learn notes,
read)...unsure if tradeoffs worth the hype...(said/wrote in reading-to-do notes
under high SES) if possible provide data...showing potential impact (but for what
actions?)...um...let’s see (looking over reading-to-do notes)...(read) individuals
with strong religious affiliations...(said/wrote) good for you, now...um...(read)
emphasize health benefits...doesn’t have to be expensive...(said/wrote) for
example insulting or turning off light saves money...eating less meat saves
money...(read)...future generations...(said/wrote after health benefits for college
students)...biking/walking versus driving...(said/wrote in high SES section)
emphasize helping others...

223

REFERENCES

224

REFERENCES

Adams, A. (2001). Decision making in multi-party regional water conflicts. In Marino, M. &
Simonovic, S. (Eds.) , Integrated water resources management. IAHS, Pub No. 272.
Afflerbach, P. (1990). The influence of prior knowledge on expert readers’ main idea
construction strategies. Reading Research Quarterly, 25, 31–46.
Afflerbach, P., & VanSledright, B. (2001). Hath! Doth! What? Middle graders reading
innovative history text. Journal of Adolescent & Adult Literacy, 44, 696-707.
Alexander P.A., Kulikowich J.M., & Jetton T.L. (1994). The role of subject-matter knowledge
and interest in the processing of linear and nonlinear texts. Review of Educational
Research, 64, 201–52
Amabile, T. M., Conti, R., Coon, H., Lazenby, J., & Herron, M. (1996). Assessing the work
environment for creativity. Academy of Management Journal, 39, 1154–1184.
Anderson, J. & Rainie, L. (2010). The future of the Internet. Pew Internet & American Life
Project. Retrieved March 8, 2010 from http://www.pewinternet.org/Reports/2010/Futureof-the-Internet-IV.aspx
Baker, L., & Brown, A. L. (1984). Metacognitive skills and reading. In P. D. Pearson, R. Barr,
M. L. Kamil & P. Mosenthal (Eds.), Handbook of Reading Research (pp. 353 - 394).
New York: Longman.
Bandura, A. (1994). Self-efficacy. In V. S. Ramachaudran (Ed.), Encyclopedia of human
behavior (Vol. 4, pp. 71-81). New York: Academic Press.
Barbour R.S. (2001) Checklists for improving rigour in qualitative research: a case of the tail
wagging the dog? British Medical Journal, 322, 1115–1117.
Bates, M. (1979). Information search tactics. Journal of the American Society for Information
Science, 30, 205–214.
Beghetto, R. A., & Kaufman, J. C. (2007). Toward a broader conception of creativity: A case for
mini-c creativity. Psychology of Aesthetics, Creativity, and the Arts, 1, 73–79.
Bereiter, C., & Scardamalia, M. (2003). Learning to Work Creatively with Knowledge. In: E. De
Corte, L. Verschaffel, N. Entwistle, & J. van Merriënboer (eds.), Unravelling Basic
Components and Dimensions of Powerful Learning Environments.EARLI Advances in
Learning and Instruction Series.
Bereiter, C., & Scardamalia, M. (1993). Surpassing ourselves: An inquiry into the nature and
implications of expertise. Chicago: Open Court.
225

Berliner, D. C. (1986). In pursuit of expert pedagogue. Educational Researcher, 5(7), 5-13.
Besemer, S.P., & O'Quin, K. (1989). The development, reliability and validity of the revised
creative product semantic scale. Creativity Research Journal, 2, 268-279.
Bilal, D. (2000). Children’s use of the Yahooligans! Web search engine: I. Cognitive, physical,
and affective behaviors on fact-based search tasks. Journal of the American Society for
Information Science, 51, 646–665.
Blaufus, K. and Ortlieb, R. (2009). Is simple better? A Conjoint analysis of the effects of tax
complexity on employee preferences concerning company pension plans. Schmalenbach
Business Review, 61(1), 60-83.
Bloom, Benjamin S. & David R. Krathwohl. (1956). Taxonomy of educational objectives: The
classification of educational goals, by a committee of college and university examiners.
Handbook 1: Cognitive domain. New York , Longmans.
Bogdan, R. C., & Biklen, S. K. (2003). Qualitative research for education: An introduction to
theories and methods, (5th ed.). Boston: Allyn & Bacon.
Bolter, J. D. (1991). Writing space: The computer, hypertext, and the history of writing.
Hillsdale, NJ: Erlbaum.
Bolter, J. D. (1998). Hypertext and the question of visual literacy. In D. Reinking, M. McKenna,
L. Labbo, & R. Kieffer (Eds.), Handbook of literacy and technology: Transformations in
a post-typographic world (pp. 3-14). Mahwah, NJ: Erlbaum.
Brand-Gruwel, S., Wopereis, I. G. J. H., & Vermetten, Y. (2005). Information problem solving
by experts and novices: Analysis of a complex cognitive skill. Computers in Human
Behavior, 21, 487-508.
Bromage, B. K., & Mayer, R. E. (1986). Quantitative and qualitative effects of repetition on
learning from technical text. Journal of Educational Psychology, 78, 271–278.
Brophy, J. (2004). Motivating students to learn. (2nd ed.). Mahwah, NJ: Erlbaum.
Brown, G. (2001). Locating categories & resources of information: How skilled are New
Zealand children? School Library Media Research, 4.
Brown, G.T.L. (2003, September/October). Searching informational texts: Text and task
characteristics that affect performance. Reading Online, 7(2).
Brown, A.L., Campione, J.C., & Day, J.D. (1981). Learning to learn: On training students to
learn from texts. Educational Researcher, 10, 14–21.

226

Bruce, B. (Ed.) (2003). Literacy in the information age: Inquiries into meaning making with new
technologies. Newark, DE: International Reading Association.
Bulger, M. (2006). Beyond search: A preliminary skill set for online literacy. Retrieved March
08, 2010 from http://transliteracies.english.ucsb.edu/post/research-project/researchclearinghouse-individual/research-papers/beyond-search-a-preliminary-skill-set-foronline- literacy.
Burbules, N. C., & Callister, T. A,,Jr. (2000). WatchIT: The risks and promise of information
technologies for education. Boulder, CO: Westview Press.
Calisir, F., & Gurel, Z. (2003). Influence of text structure and prior knowledge of the learner on
reading comprehension, browsing and perceived control. Computers in Human Behavior,
19, 135-145.
Carroll, J. B. (1999). Expert Internet information access. Journal of Educational Computing
Research, 20(3), 209-222.
Chen, H., Wigand, R.T., Nilan, M.S., 1999. Optimal experience of Web activities. Computers in
Human Behavior,15(5), 585–608.
Cho, B.Y (2011). Adolescents' constructively responsive reading strategy use in a critical
Internet reading task. (Unpublished doctoral dissertation). University of Maryland.
Chow, S.C., & Chang, M. (2008). Adaptive design methods in clinical trials - a review.
Orphanet Journal of Rare Diseases, 3,11.
Coiro, J. (2003). Reading Comprehension on the Internet: Expanding Our Understanding of
Reading Comprehension to Encompass New Literacies. The Reading Teacher, 56, 458465.
Coiro, J. (2007). Exploring changes to reading comprehension on the Internet: Paradoxes and
possibilities for diverse adolescent readers. (Unpublished doctoral dissertation).
University of Connecticut.
Coiro, J., & Dobler, E. (2004). Investigating how less-skilled and skilled readers use cognitive
reading strategies while reading on the Internet. Paper presented at the 54th annual
meeting of the National Reading Conference, San Antonio, TX.
Coiro, J. (2009, December). Purposeful, critical and flexible: New standards for online thinking,
teaching and theory. In The near future for literacy in an age of rapid technological
change: Points of substantial disjuncture form traditional view of literacy and most
urgently needed new directions for research. Symposium conducted at the annual
meeting of the National Reading Conference, Albuquerque, NM.

227

Coiro, J., & Dobler, E. (2007). Exploring the comprehension strategies used by sixth-grade
skilled readers as they search for and locate information on the Internet. Reading
Research Quarterly, 42, 214-257.
Coiro, J., Knobel, M., Lankshear, C., & Leu, D. (Eds.). (2008). Handbook of research on new
literacies. Mahwah, NJ: Erlbaum.
Coiro, J., Knobelobel, M., Lankshear, C., & Leu, D.J. (2008). Central issues in new literacies and
new literacies research. In J. Coiro, M. Knobel, C. Lankshear, & D.J. Leu (Eds.),
Handbook of research on new literacies (pp. 1–21). New York: Erlbaum.
Craik, F.M, & Lockhart, R.S. (1972). Levels of processing: A framework for memory research.
Journal of Verbal Learning and Verbal Behavior, 11, 671-684.
Csikszentmihalyi, M. (1990). Flow: The psychology of optimal experience. New York: Harper &
Row.
Csikszentmihalyi, M. (1997). Creativity: Flow and the Psychology of Discovery and Invention.
New York: Harper Collins.
Davis, F. B. (1944). Fundamental factors of comprehension in reading. Psychometrika, 9, 185197.
Deci, E. L., & Ryan, R. M. (1985). Intrinsic motivation and self-determination in human
behavior. New York: Plenum Press.
Deci, E., Vallerand, R., Pelletier, L., & Ryan, R. (1991). Motivation and education: the self
determination perspective. Educational Psychologist, 26, 321-346.
DeSchryver, M. & Mishra, P. (2008). Googling creativity: An investigation of how pre-service
mathematics teachers use the web to generate creative ways to teach. Paper presented at
the annual meeting of the American Educational Research Association, New York, NY.
DeSchryver, M. & Spiro, R. (2008). New forms of deep learning on the Web: Meeting the
challenge of cognitive load in conditions of unfettered exploration. In R. Zheng (Ed.),
Cognitive effects of multimedia learning. Hershey, PA: IGI Global, Inc.
DeSchryver, M. & Spiro, R. (in preparation). CFT on the Web: Illustrated theoretical cognitive
and metacognitive processes for advanced nonlinear knowledge acquisition of illstructured topics.
Desjarlais, M., & Willoughby, T. (2007). Supporting learners with low domain knowledge when
using the Internet. Journal of Educational Computing Research, 37(1), 1–17.
Dole, J.A., Duffy, G.G., Roehler, L.R., & Pearson, P.D. (1991). Moving From the Old to the
New: Research on Reading Comprehension Instruction. Review of Educational Research,
61, 239-264.
228

Dreher, M.J. (1992). Searching for information in textbooks. Journal of Reading, 35(5), 364-371.
Dreher, M.J. (1993). Reading to locate information: Societal and educational perspectives.
Contemporary Educational Psychology, 18, 129-138.
Duke, N. K., & Pearson, D. (2002). Effective practices for developing reading comprehension. In
A. E. Farstrup & S. J. Samuels (Eds.), What research has to say about reading
instruction (pp. 205–242). Newark, DE: International Reading Association.
Duke, N. (2010). Informational text and young children: when, why, what, where, and how. In
National Geographic Best Practices and Research Base.
Dye, J. F., Schatz, I. M., Rosenberg, B. A., & Coleman, S. T. (2000). Constant comparison
method: A kaleidoscope of data. The Qualitative Report, 4(1/2). Retrieved December
15th, 2009, from http://www.nova.edu/ssss/QR/QR4-1/dye.html
Eagleton, M.B. (2003, December). Essential literacies in the digital age: Adolescents’ Internet
inquiry strategies. Paper presented at the 53rd annual meeting of the National Reading
Conference, Scottsdale, AZ.
Eastebrook, L., Witt, E., & Rainie, L. (2007). Information searches that solve problems. Pew
Internet & American Life Project. Retrieved March 8, 2010 from
http://www.pewinternet.org/~/media//Files/Reports/2007/Pew_UI_LibrariesReport.pdf.p
df
Eastin, M.S. & LaRose, R. (2000). Internet self-efficacy and the psychology of the digital divide.
JCMC, 6(1).
Eagleton, M.B., & Guinee, K. (2002). Strategies for supporting student Internet inquiry. New
England Reading Association Journal, 38(2), 39-47.
Eagleton, M.B., Guinee, K., & Langlais, K. (2003). Teaching Internet literacy strategies: The
hero inquiry project. Voices From the Middle, 10, 28–35.
Ertmer, P. A., & Newby, T. J. (1996). The expert learner: Strategic, self-regulated, and
reflective. Instructional Science, 24, 124.
Educational Testing Service (2005). ICT literacy: Equipping students to succeed in an
information-rich, technology-based society. Princeton, NJ: Educational Testing Service.
Retrieved March 5th, 2010 from
http://www.ets.org/Media/Tests/ICT_Literacy/pdf/ICT_Equipping_Students_to_Succeed.
pdf
Evaluating Web pages: Techniques to apply & questions to ask. Retrieved from
http://www.lib.berkeley.edu/TeachingLib/Guides/Internet/Evaluate.html
229

Fidel, R. (1985). Moves in online searching. Online Review, 9(1), 61–74.
Florida, R. (2002). The rise of the creative class. New York: Basic Books.
Freidman, T. (2005). The world is flat. New York: Farrar, Straus and Giroux.
Gallo P, Chuang-Stein C, Dragalin V, Gaydos B, Krams M, Pinheiro J. (2006). Adaptive design
in clinical drug development – an executive summary of the PhRMA Working Group
(with discussions). Journal of Biopharmaceutical Statistics, 16 (3), 275-283.
Gee, J. (2003). What video games have to teach us about learning and literacy. New York:
Palgrave.
Gehlbach, H., Brown, S.W., Ioannou, A., Boyer, M.A., Hudson, N., Niv-Solomon, A.,
Maneggia, D. & Janik, L. (2008). Increasing interest in social studies: Social perspective
taking and self-efficacy in stimulating simulations. Contemporary Educational
Psychology 33(4), 894-914.
Glister, P. (2000). Digital literacy. In Pea, R. (Ed.). The Jossey-Bass reader on technology and
learning. San Francisco, CA: Jossey-Bass, Inc.
Graesser, A. C., & Britton, B. K. (1996). Five metaphors for text understanding. In B. K. Britton
& A. C. Graesser (Eds.), Models of understanding text (pp. 341–351). Mahwah, NJ:
Lawrence Erlbaum.
Graesser, A. C., Mills, K. K., & Zwaan, R. A. (1997). Discourse comprehension. Annual Review
of Psychology, 48, 163-189.
Graesser, A.C., Swamer, S.S., Baggett, W.B., and Sell, M.A. (1996). New Models of deep
comprehension. In B. K. Britton & A. C. Graesser (Eds.), Models of understanding text
(pp. 1-32). Mahwah, NJ: Lawrence Erlbaum.
Graesser, A. C., & Kreuz, R. J. (1993). A theory of inference generation during text
comprehension. Discourse Processes, 16, 145-160.
Graesser, A.C., Singer, M. & Trabasso, T. (1994). Constructing inferences during narrative text
comprehension. Psychological Review, 101, 371-395.
Griffiths, J. R., & Brophy, P. (2005). Student searching behavior and the web: use of academic
resources and Google. Library Trends, 53, 539-554
Gunther, R.E. (2011). The truth about your emotions when making decisions. Pearson Education
Inc., FT Press: New Jersey

230

Guthrie, J.T., & Kirsch, I.S. (1987). Distinctions between reading comprehension and locating
information in text. Journal of Educational Psychology, 79(3) 220-227.
Hannafin, M. J., Hannafin, K. M., & Gabbitas, B. (2009). Re-examining cognition during
student-centered, web-based learning. Educational Technology Research and
Development, 57(6), 767–785.
Hanoch, Y., Wood, S., Barnes, A., Liu, P., & Rice, T. (2011). Choosing the right Medicare
prescription drug program: The effect of age, strategy selection, and choice set size.
Health Psychology, 30, 719–727.
Hartman, D.K. (1994). The intertextual links of readers using multiple passages: A
postmodern/semiotic/cognitive view of meaning making. In R.B. Ruddell, M.R. Ruddell,
& H. Singer (Eds.), Theoretical models and processes of reading (pp. 616-636). Newark,
DE: International Reading Association.
Hartman, D. K. (1995). Eight readers reading: The intertextual links of proficient readers reading
multiple passages. Reading Research Quarterly, 30, 520–61.
Hartman, D. K. (2000). What will be the influences of media on literacy in the next millennium?
Reading Research Quarterly, 35, 280–2.
Hartman, D. K., Morsink, P. M., & Zheng, J. (2010). From print to pixels: The evolution of
cognitive conceptions of reading comprehension. In E. A. Baker (Ed.), The new
literacies: Multiple perspectives on research and practice (pp. 131-164). New York, NY:
Guilford.
Henry, L.A. (2006). SEARCHing for an answer: The critical role of new literacies while reading
on the Internet. The Reading Teacher, 59, 614–627.
Hidi, S., & Renninger, K. A. (2006). The four-phase model of interest development.
Educational Psychologist, 41, 111-127.
Hobson, J. (2009, November 4). Danish pupils use web in exams. BBC. Retrieved March 5th,
2010 from http://news.bbc.co.uk/2/hi/education/8341886.stm
Hofstadter, D. R. (1985). Variations on a theme as the crux of creativity. In D. R. Hofstadter,
Metamagical themas (pp. 232-259). New York: Basic Books.
Holscher, C., & Strube, G. (2000) Web search behaviour of Internet experts and Newbies.
Proceedings of WWW9. Available at http://www9.org/w9cdrom/81/81.html
Hollander, R. (2004). Social genomics: Genomic inventions in society. The nature of what’s to
come. Science and Engineering Ethics, 8 (4), 485-496.

231

Howe, M. J. A. (1970). Repeated presentation and recall of meaningful prose. Journal of
Educational Psychology, 61, 214 –219.
Hsieh-Yee, I. (1993). Effects of search experience and subject knowledge on the search tactics of
novice and experienced searchers. Journal of the American Society for Information
Science, 44(3), 161-174.
International Reading Association. (2009). New literacies and 21st-century technologies: A
position statement. Newark, DE: International Reading Association. [On-line]. Retrieved,
March 5th, 2010 from
http://www.reading.org/General/AboutIRA/PositionStatements/21stCenturyLiteracies.as
px
International Society for Technology in Education. (2008). National educational technology
standards for teachers.Retrieved from
http://www.iste.org/Content/NavigationMenu/NETS/ForTeachers/NETS_for_Teachers.ht
m
International Society for Technology in Education. (2009). NETS for students: Global learning
in a digital age. Retrieved from
http://www.iste.org/Content/NavigationMenu/NETS/ForStudents/NETS_for_Students.ht
m
Jacobson, M. J., & Spiro, R. J. (1995). Hypertext learning environments, cognitive flexibility,
and the transfer of complex knowledge: An empirical investigation. Journal of
Educational Computing Research,12, 301-333.
Jackson, N. J. (2006) Imagining a different world, in N.J. Jackson et. al. (eds.) Developing
Creativity in Higher Education: an Imaginative Curriculum, London: Routledge-Falmer.
Jenkins, H. (2006). Confronting the challenges of participatory culture: Media education for the
21st century. Chicago, IL: The John D. & Catherine T. Macarthur Foundation.
Johnson, H. M., & Seifert, C. M. (1999). Modifying mental representations: Comprehending
corrections. In H. Oostendorp & S. R. Goldman (Eds.), The construction of mental
representations during reading (pp. 303-318). Mahwah, NJ: Erlbaum.
Junivo, 1. (2006). Development of a cognitive model for navigating on the web (Unpublished
doctoral dissertation). Utrecht University.
Desjarlais, M (2010). Novices' Learning from the Internet: An Exploration of Navigation
Behaviours, Learner-related Factors, and Mental Effort. Retreived from
http://dr.library.brocku.ca/bitstream/handle/10464/3160/Brock_Desjarlais_Malinda_201
0.pdf?sequence=1

232

Karchmer, R. (2001). The journey ahead: Thirteen teachers report how the Internet influences
literacy and literacy instruction in their K–12 classrooms. Reading Research Quarterly,
36, 4, 442-466.
Kardash, C. M., & Howell, K. L. (2000). Effects of epistemological beliefs and topic- specific
beliefs on undergraduates’ cognitive and strategic processing of dual-positional text.
Journal of Educational Psychology, 92, 524–535.
Keene, E. & Zimmerman, S. (1997). Mosaic of thought. Portsmouth, NH: Heinemann.
Kendeou, P., & van den Broek, P. (2005). The effects of readers’ misconceptions on
comprehension of scientific text. Journal of Educational Psychology, 97, 235–245.
Kiewra, K. (1989). A review of note-taking. The encoding-storage paradigm and beyond.
Educational Psychology Reivew, 1 (2), 147-172.
Kintsch, W. (1998). Comprehension. New York: Cambridge University Press.
Kintsch, E. (1990). Macroprocesses and microprocesses in the development of summarization
skill. Cognition and Instruction, 7(3), 161-195.
Kintsch, E., & Kintsch, W. (1996). Learning from text. In E. DeCorte & F. E. Weinert (Eds.),
International encyclopedia of developmental and instructional psychology (pp. 519–524).
Kidlington, Oxford: Pergamon Press.
Kintsch, W., & Rawson, K. A. (2005). Comprehension. In M. J. Snowling & C. Hulme (Eds.),
The science of reading: A handbook (pp. 209–226). Oxford, UK: Blackwell.
Kress, G. (2000). Design and Transformation: New Theories of Meaning. In B. Cope and M.
Kalantzis Multiliteracies: Literacy Learning and the Design of Social Futures (pp. 155156). London: Routledge.
Kuiper, E., Volman, M., & Terwel, J. (2005). The Web as an information resources in K-12
education: Strategies for supporting students in searching and processing information.
Review of Educational Research, 75(3), 285-328.
Kuiper, E., Volman, M., & Terwel, J. (2009). Developing web literacy in collaborative inquiry
activities. Computers & Education, 52, 668-680.
Kumbruck, C. (1998). Hypertext reading: Novice vs. expert reading. Journal of Research in
Reading, 21, 160–172.
Lebo, H. (2004). The digital future report: Surveying the digital future, year three. Los Angeles:
USC Annenberg School Center for the Digital Future. Retrieved March 8, 2010 from
http://www.digitalcenter.org/pdf/DigitalFutureReport-Year4-2004.PDF

233

Leu, D. Castek, J., Hartman, D., Coiro, J., Henry, L., Kulikowich, J., Lyver, S. (2005).
Evaluating the development of scientific knowledge and new forms of reading
comprehension during online learning. Final report presented to the North Central
Regional Educational Laboratory/Learning Point Associates. Retrieved March 5th 2010
15, 2006 http://www.newliteracies.uconn.edu/ncrel.html.
Leu, D. J., Coiro, J., Castek, J., Hartman, D., Henry, L. A., & Reinking, D. (2008). Research on
instruction and assessment in the new literacies of online reading comprehension. C. C.
Block & S. Parris, (Eds.). Comprehension instruction: Research-based best practices
(2nd ed.). (pp 321-346). New York: Guilford Press.
Leu, D.J., JR., Kinzer, C.K., Coiro, J., & Cammack, D.W. (2004). Toward a theory of new
literacies emerging from the Internet and other information and communication
technologies. In R.B. Ruddell & N. Unrau (Eds.), Theoretical models and processes of
reading (5th ed., pp 1570-1613). Newark, DE:International Reading Association.
Retrieved Dedcember 15th, 2009 from
http://www.readingonline.org/newliteracies/lit_index.asp?HREF=/newliteracies/leu
Leu, D. J., McVerry, J. G., O’Bryne, W. I., Zawilinski, L., Castek, J., & Hartman, D. K. (2009).
The new literacies of online reading comprehension and the irony of no child left
behind: Students who require our assistance the most, actually receive it the least. In: L.
M. Morrow, R. Rueda, & D. Lapp (Eds.), Handbook of research on literacy and diversity
(pp.173-194). New York: Guilford Press.
Leu, D. J., Reinking, D., Carter, A., Castek, J., Coiro, J., Henry, L. A., et al. (2007, April).
Defining online reading comprehension: Using think aloud verbal protocols to refine a
preliminary model of Internet reading comprehension processes. Paper presented at the
annual meeting of the American Educational Research Association, Chicago. Retrieved
December 15th, 2009, from http://docs.google.com/Doc?id=dcbjhrtq_10djqrhz
Leu, D. J., Zawilinski, L., Castek, J., Banerjee, M., Housand, B., Liu, Y., et al. (2007). What is
new about the new literacies of online reading comprehension? In L. Rush, J. Eakle, & A.
Berger, (Eds.). Secondary school literacy: What research reveals for
classroom practices. (37-68). Urbana, IL: National Council of Teachers of
English. Retrieved December 15th, 2009 from
http://www.newliteracies.uconn.edu/docs/whatsnewaboutonlinereading
comprehension.pdf
Leu, D. J., McVerry, J. G., O’Byrne, W.I ., Kiili, C., Zawilinski, L., Everett-Cacopardo, H., &
Forzani, E. (2011). The new literacies of online reading comprehension: Expanding the
literacy and learning curriculum. Journal of Adolescent & Adult Literacy, 55, 5–14.
Litwin, M. S. (1995). How to measure survey reliability and validity. Thousand Oaks, CA: Sage
Publications.

234

Lorenzo, G., Oblinger, D. & Dziuban, C. (2006). How choice, co-creation, and culture are
changing what it means to be net savvy. EDUCAUSE Quarterly, 30(1). Retrieved April
17th, 2008 from
http://connect.educause.edu/Library/EDUCAUSE+Quarterly/HowChoiceCoCreationand
Cul/40008
Lubart, T. I. & Sternberg, R.J. (1995). An investment approach to creativity: theory and data, In
S. M. Smith, T. B. Ward and R. A. Finke (Eds), The creative cognition approach (pp.
271-302), Cambridge, MA: The MIT Press.
Madden, M. & Jones. S. (2008). Networked workers: Most workers use the Internet or email at
their jobs, but they say these technologies are a mixed blessing for them. Pew Internet &
American Life Project. Retrieved March 8, 2010 from
http://www.pewinternet.org/~/media//Files/Reports/2008/PIP_Networked_Workers_FIN
AL.pdf.pdf
M. McKenna, L. Labbo, R. Keiffer, & D. Reinking, D. (Eds.). (2006) International handbook of
literacy and technology (Vol. 2). Mahwah, NJ: Erlbaum.
McWilliam, E. (2009, June). The knight’s move: Its relevance for educational research and
development. Keynote presented at the 3rd Redesigning Pedagogy International
Conference, Singapore.
Mishra, P., & DeSchryver, M. (2009, May). Creative teaching. Co-presenter at half-day
workshop conducted for university faculty at Michigan State University.
Mishra, P. & Koehler, M.J. (2007). Technological Pedagogical Content Knowledge (TPCK):
Confronting the Wicked Problems of Teaching with Technology. In R. Carlsen et al.
(Eds.), Proceedings of Society for Information Technology & Teacher Education
International Conference 2007 (pp. 2214-2226). Chesapeake, VA: AACE.
Mishra, P., Spiro, R.J., & Feltovich, P.J. (1996). Technology, representation, and cognition: The
prefiguring of knowledge in Cognitive Flexibility Hypertexts. In H. van Oostendorp & A.
de Mul (Eds.), Cognitive aspects of electronic text processing (pp. 287-305). Norwood,
N.J.: Ablex.
Mitroff, I., Alpasian, M., & Green, S. (2004). Managing crises in the twenty-first century.
International Studies Review, 6 (6), 165-194.
Monereo, C., Fuentes, M. & Sanches, S. (2000). Internet search and navigation strategies used by
experts and beginners. Interactive Educational Multimedia, 2000, 1, 24-34.
Morgan, D. L. (2007). Paradigms lost and pragmatism regained: Methodological implications
of combining qualitative and quantitative methods. Journal of Mixed Methods Research,
1(1), 48-76.

235

Mosenthal, P.B., & Kirsch, I.S. (1991). Toward an explanatory model of document literacy.
Discourse Processes, 14, 147-180.
National Assessment Governing Board. (2008). Reading framework for the 2009 National
Assessment of Educational Progress. Retrieved June 29, 2011, from
http://www.nagb.org/publications/frameworks/reading09.pdf
New London Group. (1996). A pedagogy of multiliteracies: Designing social futures. Harvard
Educational Review, 66, 60-92.
Netemeyer, R.G., Bearden, W.O., & Sharma, S. (2003). Scaling procedures: Issues and
applications. Thousand Oaks, CA: Sage.
Organisation for Economic Co-operation and Development. PISA 2009 assessment framework:
Key competencies in reading, mathematics and science. Retrieved October 13, 2011,
from
http://www.oecd.org/document44/0,3343,en_2649_35845621_44455276_1_1_1_1,00.ht
ml
Osborn, A. (1953). Applied Imagination. New York: Scribner’s.
Pace, S. (2004). A grounded theory of the flow experiences. Int. J. Human-Computer Studies,
60, 327–363.
Paerson, P.D., Roehler, L.R., Dole, J.A., & Duffy, G.G. (1992). Developing expertise in reading
comprehension. In S.J. Samuels & A.E. Farstrup (Eds.), What research has to say about
reading instruction (pp. 145–199). Newark, DE: International Reading Association.
Pask, G. (1976). Conventional techniques in the study and practice of education. British Journal
of Educational Psychology, 46, 12-25.
Perreault, W.D. & L.E. Leigh. (1989), Reliability of Nominal Data Based on Qualitative
Judgments. Journal of Marketing Research, 26 (May), 135–48.
Peshkin, A. (1988). In search of subjectivity—One’s own. Educational Researcher, 17(7), 17–
21.
Pink, D. (2006). A whole new mind: Why right-brainers will rule the future. New York, NY:
Penguin Group.
Potelle, H., & Rouet, J.-F. (2003). Effects of content representation and readers’ prior knowledge
on the comprehension of hypertext. International Journal of Human-Computer Studies,
58, 327-345.
Pressley, M., & Afflerbach, P. (1995). Verbal protocols of reading: The nature of constructively
responsive reading. Hillsdale NJ: Erlbaum.

236

QCA. (2004). Creativity: Find it, promote it. Retrieved November 5th, 2006, 2005. from
http://www.ncaction.org.uk/creativity
RAND Reading Study Group. (2002). Reading for understanding: Toward an R&D program in
reading comprehension. Santa Monica, CA: RAND.
Rawson, K. A., Dunlosky, J., & Thiede, K. W. (2000). The rereading effect: Metacomprehension
accuracy improves across reading trials. Memory & Cognition, 28, 1004 –1010.
Reinking, D., McKenna, M.C., Labbo, L.D., & Kieffer, R.F. (Eds.). (1998). Handbook of literacy
and technology: Transformations in a post-typographic world. Mahwah, NJ:Erlbaum.
Rouet, J-F. (2006). The skills of document use. Mahwah, NJ: Erlbaum.
Rouet, J.F., & Levonen, J.J. (1996). Studying and learning with hypertext: Empirical studies and
their implications. In J.-F. Rouet, J.J. Levonen, A. Dillon, & R.J. Spiro (Eds.), Hypertext
and cognition (pp.9–23). Hillsdale, NJ: Erlbaum.
Sawyer, K. (2012). Explaining Creativity: The Science of Human Innovation. New York, NY:
Oxford University Press, Inc.
Schrum, L. (2005). Technology as a tool to support instruction. Retrieved from
http://www.educationworld.com/a_tech/tech/tech004.shtml
Selwyn, N. (2008) An investigation of differences in undergraduates' academic use of the
internet. Active Learning in Higher Education, 2008, 9(11), 11-22.
Schraw, G. (1997). Situational interest in literary text. Contemporary Educational Psychology,
22(4), 436-456.
Slotte, V., & Lonka, K. (1999). Review and process effects of spontaneous note taking on text
comprehension. Contemporary Educational Psychology, 24(1), 1–20.
Song, H. (2005). Motivating ill-structured problem solving in a Web-based peer-group learning
environment: A learning-goal perspective. Journal of Educational Computing Research,
33 (4), 351-367.
Spiro, R.J. (2006a). The “New Gutenberg Revolution”: Radical new learning, thinking,
teaching, and training with Technology…bringing the future near. Educational
Technology, 46 (1), 3-4.
Spiro, R.J. (2006b). The post-Gutenberg world of the mind: The shape of the new learning.
Educational Technology, 46 (2), 3-4.
Spiro, R.J. (2006c). Old ways die hard. Educational Technology, 46 (3), 3-4.

237

Spiro, R.J. (2006d). Approaching the post-gutenberg mind: The revolution is in progress.
Educational Technology, 46 (4), 3-4.
Spiro, R.J. (2006e). The “New Gutenberg Revolution”: Radical new learning, thinking,
teaching, and training with technology…bringing the future near. Educational
Technology, 46 (6), 3-5.
Spiro, R. J., Collins, B. P., Thota, J. J., & Feltovich, P. J. (2003). Cognitive flexibility theory:
Hypermedia for complex learning, adaptive knowledge application, and experience
acceleration. Educational Technology, 44, 5–10
Spiro, R.J. & DeSchryver, M. (2008). Constructivism: When it’s the wrong idea and when it’s
the only idea. In T. Duffy & S. Tobais (Eds.) Constructivist instruction: Success or
failure? Mahwah, NJ: Lawrence Erlbaum.
Spiro, R., DeSchryver, M. (2010). Advanced cognitive and metacognitive processes for deep
learning on the web. In At the intersection of literacy & technology: Research
developments from the literacy achievement research center. Symposium conducted at
the annual meeting of the National Reading Conference, Forth Worth, TX.
Spiro, R. J., & Jehng, J. C. (1990). Cognitive flexibility and hypertext: Theory and technology
for the nonlinear and multidimensional traversal of complex subject matter. In D. Nix &
R. J. Spiro (Eds.), Cognition, education, and multimedia: Explorations in high
technology (pp. 163-205). Hillsdale, NJ: Lawrence Erlbaum.
Spiro, R. J., Feltovich, P. J., Jacobson, M. J., & Coulson, R. L. (1992). Cognitive flexibility,
constructivism, and hypertext: Random access instruction for advanced knowledge
acquisition in ill-structured domains. In T. Duffy & D. Jonassen (Eds.), Constructivism
and the technology of instruction (pp. 57–75). Hillsdale, NJ: Erlbaum. [Reprinted from a
special issue of the journal Educational Technology on Constructivism, 1991.]
Spiro, R. J., Collins, B. P., & Ramchandran, A. R. (2006). Modes of openness and flexibility in
“Cognitive Flexibility Hypertext” learning environments. In B. Khan (Ed.), Flexible
learning in an information society. (pp. 18-25). Hershey, PA: Information Science
Publishing.
Stake, R. (1995). The art of case study research. Thousand Oaks, CA: Sage.
Sternberg, R. J. (1988). The Nature of Creativity: Contemporary Psychological Perspectives.
Cambridge, U.K.: Cambridge University Press.
Sticht, T. (1979). Applications of the AUDREAD model to reading evaluation and instruction.
In: L. Resnick and P. Weaver (Eds.), Theory and Practice in Early Reading: Vol. 1,
Hillsdale, N.J., Lawrence Erlbaum and Associates.

238

Sticht, T. G. (2002). Teaching reading with adults. Retrieved November 17, 2011, from National
Adult Literacy Database, Inc. Web site http://www.nald.ca/fulltext/sticht/jan02/cover.htm
Strømsø, H. I., & Bråten, I. (2002). Norwegian law students’ use of multiple sources while
reading expository texts. Reading Research Quarterly, 37, 208–227.
Strømsø, H. I., Bråten, I., & Samuelstuen, M. S. (2003). Students’ strategic use of multiple
sources during expository text reading: A longitudinal think-aloud study. Cognition and
Instruction, 21, 113–147.
Stromso, H. I., Braten, I., & Samuelstuen, M. S. (2003). Students' strategic use of multiple
sources during expository text reading: A longitudinal think-aloud study. Cognition and
Instruction, 21, 113-147.
Sutcliffe, A. & Ennis, M. (1998). Towards a Cognitive theory of information retrieval.
Interacting with Computers, 10, 321-351.
Sutherland-Smith, W. (2002). Weaving the literacy Web: Changes in reading from page to
screen. The Reading Teacher, 55, 662–669.
Sweller, J., van Merrienboer, J. J. G., & Paas, F. (1998). Cognitive architecture and instructional
design. Educational Psychology Review 10(3), 251–296.
Tang, H., Ng, J. H. (2006) Googling for a diagnosis—use of Google as a diagnostic aid: Internet
based study. BMJ, 333,1143-1145.
Tarchi, C. (2010). Reading comprehension of informative texts in secondary school: A focus on
direct and indirect effects of reader's prior knowledge. Learning and Individual
Differences, 20(5), 415-420.
Thaler, R.H., Sunstein, C.R. (2008). Nudge: Improving Decisions About Health, Wealth, and
Happiness. New Haven, CT: Yale Univ. Press
Thieman, G. (2008). Using Technology as a Tool for Learning and Developing 21st Century
Skills: an Examination of Technology Use by Pre-Service Teachers with their K-12
Students. Contemporary Issues in Technology and Teacher Education, 8(4), 342-366.
AACE.
Tarchi, C. (2010). Reading comprehension of informative texts in secondary school: A focus on
direct and indirect effects of reader's prior knowledge. Learning and Individual
Differences, 20(5), 415-420.
Taxonomy of Internet reading comprehension skills and strategies. Retrieved from University of
Connecticut New Literacies Web site:
http://www.newliteracies.uconn.edu/iesproject/documents/TICA_Taxonomy_of_Skills.p
df

239

TICA Phase II Checklist. Retrieved from University of Connecticut New Literacies Web site:
http://www.newliteracies.uconn.edu/iesproject/documents/TICA_Phase_II_Checklist.doc
Tesch, R. (1990). Qualitative research: Analysis types and software tools. Bristol, PA: Falmer.
van dijk, T.A., & Kintsch, W. (1983). Strategies of discourse comprehension. New York:
Academic.
Van Scoyoc, A. (2006). The electronic academic library: Undergraduate research behavior in a
library without books. Libraries and the Academy, 6(1), 47–58.
Vakkari, P. (2001). A Theory of the task-based information retrieval process: A summary and
generalisation of a longitudinal study. Journal of Documentation, 57(1), 44-60.
Wallas, G. (1926). The art of thought. New York: Harcourt, Brace.
Ward, T.B., Smith, S.M., & Finke, R.A. (1999). Creative Cognition. In R. J. Sternberg
(Ed). Handbook of Creativity (pp. 189-212). Cambridge, U.K.; New York, NY:
Cambridge University Press.
Warwick, C., Galina, I., Rimmer, J., Terras, M., Blandford, A., Gow, J., & Buchanan, G. (2009)
Documentation and the users of digital resources in the humanities. Journal of
Documentation, 65(1), 33-57.
Weaver, C.A., III & Kintsch, W. (1991). Expository text. In R. Barr, M.L. Kamil, P. Mosenthal,
& P.D. Pearson (Eds.), Handbook of reading research (Vol. 2, pp. 230–244). White
Plains, NY: Longman.
Weisberg, R. W. (2006). Creativity: Understanding innovation in problem solving, science,
invention, and the arts. Hoboken, NJ: Wiley.
Weisberg, R.W. & J.W. Alba. (1981). An exammination of the alleged role of 'fixation' in the
solution of several ‘insight' problems. Journal of Experimental Psychology: General, 110
(June), 169-192.
Weisberg, R.W. (1995). Prolegomena to theories of insight in problem solving: A taxonomy of
problems. In R.J. Sternberg & J.E. Davidson (Eds.), The nature of insight (pp. 157–196).
Cambridge, MA: MIT Press
Wen, J. (2003). Post-Valued Recall Web Pages: User Disorientation Hits the big time.
IT&Society, 1, 184-194.
Wenger, M. J., & Payne, D. G. (1996). Comprehension and retention of nonlinear text:
considerations of working memory and material-appropriate processing. American
Journal of Psychology, 109(1), 93–130.

240

White, R.W. and Drucker, S.M. (2007) Investigating behavioral variability in web search. Proc.
WWW 2007, 21-30.
Willoughby, T., Anderson, A., Wood, E., Mueller, J., & Ross, C. (2009). Fast searching for
information on the Internet to use in a learning context: The impact of prior knowledge.
Computers & Education, 52, 640-648.
Wolfe, M. B. W., & Goldman, S. R. (2005). Relations between adolescents' text processing and
reasoning. Cognition and Instruction, 23, 467-502.
Wilson, M. (2009). Established information-seeking models. Journal of the American Society for
Information Science, 60(7), 1407-1422.
Wilson M.L., Schraefel, M.C., White R.W. (2009). Evaluating advanced search interfaces using
established information-seeking models. Journal of the American Society for
Information Science & Technology. 60(7), 1407-1422.
Wyatt, D., Pressley, M., El-Dinary, P. B., Stein, S., Evans, P., & Brown, R. (1993).
Comprehension strategies, worth and credibility monitoring and evaluations: Cold and
hot cognition when experts read professional articles that are important to them. Learning
and Individual Differences, 5, 49-72.
Wu, Y. & Tsai, C. (2006). University students' Internet attitudes and Internet self-efficacy: A
study at three universities in Taiwan. CyberPsychology & Behavior, 9(4), 441-450.
Yin, R.K. (1989). Case study research: Design and methods (2nd ed.). Thousand Oaks, CA:
Sage.
Young, K. (2005). Direct from the source: the value of ‘think-aloud’ data in understanding
learning. Journal of Educational Inquiry, 6(1), 19-33.
U. S. Department of Commerce: National Telecommunications and Information Administration
(2002). A nation online: How Americans are expanding their use of the Internet.
Washington, DC: U.S. Department of Commerce.
Zerubavel, E. (1995). The rigid, the fuzzy, and the flexible: Notes on the mental sculpting of
academic identity. Social Research, 62,1093–1106.
Zhang, S., & Duke, N. K. (2008). Strategies for Internet reading with different reading purposes:
A descriptive study of twelve good Internet readers. Journal of Literacy Research, 40,
128-162.

241