mm “W“ gm 1; MU} um; um um \I

 

 

ABSTRACT

THE EFFECT OF PHOTOGRAPHIC PRODUCTION ERRORS
IN THE VISUAL TRACK.OF A MULTI-SENSORY

PROGRAM FORMAT ON COGNITIVE LEARNING

BY

Bruce Leslie Miles

An important activity of many media specialists and virtually
every person involved in the process of teaching or the management of
learning, is the design, production, and delivery of instructional
messages. Frequently such messages are multi-sensory in design.

Precisely how to consistently deliver a viable instructional
product is a dilemma which suffers from a recognized paucity of
empirically derived guidelines. Thus, the design and subsequent
production of such messages is accomplished largely through procedures
suggested by tradition, intuition, or experience.

One production aspect which seemingly remains unresolved
concerns the relationship between technical quality in the visual
channel of multi-sensory modalities and cognitive learning outcomes.
This is especially true when the audio channel is the primary message
source. Traditionally, the most frequently used guideline suggests

that poor technical quality interferes with cognitive learning.

Bruce Leslie Miles
However, other research dealing with channel relationships, information
processing, and visual information levels, suggests that such a
position might not be intrinsically true, but rather conditioned by
other factors including audience characteristics, the learning task,
and subject matter.

This study was undertaken to investigate the effect of commonly
observed production-type errors on cognitive learning. The test
vehicle was a slide tape program rendered in cartoon style and the
content, largely unfamiliar to the subjects, was an introduction to
an instructional development model. A population of undergraduate
education students at Michigan State University viewed one of four
versions of the program which varied only in the number of errors
present. The experimental control was A no-error verSion. It was
hypothesized that production errors would be observed and that
interference from such errors would result in decreased learning as
the number of errors increased.

Analysis of variance was performed on the data and yielded a sig-
nificant difference between groups on error perception. Although all
treatment groups scored lower than the no-error control group and there
was a trend toward positive correlation between the number of errors and
recall scores, differences were not statistically significant.

The obviousness of the errors undoubtedly contributed to the
highly significant finding on error perception. Population size,
statistical design, channel configuration, and the reliability

of the cognitive instrument may have influenced the finding of no

Bruce Leslie Miles

significant difference on the learning hypothesis.

Suggestions for future research include replication of
the study using more subjects, refined instrumentation, and exposure
of treatment groups to increased numbers of errors. Additional
suggestions include a version of the study to investigate common
presentation (projection) errors; one using different subject matter;
and a covariate study investigating audience characteristics, error
perception, and recall. It is strongly recommended that research

be continued in the general area.

THE EFFECT OF PHOTOGRAPHIC PRODUCTION ERmRS
IN THE VISUAL TRACK OF A MULTI-SENSORY

PROGRAM FORMAT ON COGNITIVE LEARNING

BY

Bruce Leslie Miles

A DISSERTATION

Submitted to
Michigan State University
in partial fulfillment of the requirements
for the degree of

DOCTOR OF PHILOSOPHY

Department of Instructional Development and Technology

1974

, ACKNOWLEDGMENTS

The author wishes to express his deepest thanks and appreciation
to the members of his doctoral committee for their advice, suggestions,
generous attitude, patience, and understanding.

To Dr. Stanley E. Hecker for his searching questions when the
research was in the formative stage; to Dr. Dale Alam who sensed unique
conditions and allowed much freedom of action; to Dr. Paul W. F. Witt
who prodded with kindness and counseled with unusual scholorship; and
to Dr. Charles F. Schuller who opened many doors, the most important of
which was a faculty position in the Instructional Media Department of
this great University.

Additional and sincere thanks go to wonderful colleagues and
friends including Somnath Chatterjee, Drs. Kent Gustafson, Curtis
McCarty, Elwood Miller, Professor Wilfred Veenendaal, Sharon Fatura,
and Ruth Murphy for their professional advice, concerns, and, above
all else, their moral support.

Finally, special appreciation is extended to the author's family.
To mywife Joan who truly made the realization of a lifelong dream a
reality; to Lynda, Dirk, and Kelly, my children who through their
willingness to change friends and life styles helped to make the years
of study a pleasure; and to my father whose own contribution to society
and his family can never be equaled and shall forever be a model to

strive toward.

ii

TABLE OF CONTENTS

LIST OF TABLES O O O O O O O O O O O O O O 0

Chapter

I 0 INTRODUCTION 0 O O O O C O O O O O 0

Purpose of the Study . . . . . . .
The Nature of the Problem . . . . .
Need for the Study . . . . . . . .
Summary . . . . . . . . . . . . . .

II. REVIEW OF THE LITERATURE . . . . . . .

Communication Noise and Visual Channel Interference
Relevant-~Irre1evant Attention Gaining Devices .

Communication Properties of Pictures

Legibility and Technical Quality .
smary I O O O O O O O O O O O O 0

III. PROCEDURES AND METHODOLOGY . . . . . .

Statement of the Problem . . . . .
Hypotheses . . . . . . . . . . . .
Population .. . . . . . . . . . . .

. Instrmentation . C O O O O O O O 0

Message Design . . . . . . . . . .
Audio Channel Revision . . . . . .
Visual Channel Preparation . . . .
Channel Relationship . . . . . . .
Equipment and Facilities . . . . .
Technical Assistance . . . . . . .
Summary . . . . . . . . . . . . . .

IV. ANALYSIS OF DATA . . . . . . . . . . .

Review of Experimental Procedure .
Hypotheses . . . . . . . . . . . .
Perception of Errors Hypothesis and
Additional Findings . . . . . . . .

Data Analysis

Cognitive Learning Hypothesis and Data Analysis .

iii

Page

20
25
28
32
35

37

-37
37
38
39
40
43
44
48
49
50
53

55_

55
56
56
63
64

Chapter

Error-Performance Relationship Hypothesis and
Data Analysis . . . . . . . . . . . . . . . .
Summary . . . . . . . . . . . . . . . . . . . .

V. SUMMARY, CONCLUSION AND DISCUSSION . . . . . . . .

Summary . . . . . . . . . . . . . . . . . . . .
Conclusions . . . . . . . . . . . . . . . . . .
Discussion . . . . . . . . . . . . . . . . . . .
Implications for Future Research . . . . . . . .

”PENDICES O O O O O C O O O O I O O O O O O C O O O O

. LETTER TO PROFESSOR BREHM . . . . . . . . . . . .
RECALL INSTRUMENT . . . . . . . . . . . . . . .
. ERROR PERCEPTION INSTRUMENT AND RESPONSE FORM . .
CREDITS FOR "A FABLE REVISITED" . . . . . . . . .
. INSTRUCTIONAL DEVELOPMENT MODEL . . . . . . . . .
. AUDIO CHANNEL CONTENT DETERMINATION FORM . . . .
. SCRIPT FOR "A FABLE REVISITED" WITH ERROR LOCATION
TYPE AND CHANNEL RELATIONSHIPS . . . . . . . . .
H. INSTRUCTION-GROUP ASSIGNMENT FORM . . . . . . . .
I. SCRIPT OF TAPE RECORDED INTRODUCTION . . . . . . .

QWNUOUJID‘

BIBLIOGMPEIY O O O O O O O O O O O O O O O O I O O O O 0

iv

Page

65
66

67

67
69
71
78

82

82
84
91
94
95
96

97
112
113

115

Table

3.1.

3.3.

4.1.

4.5.

4.6.

4.8.

4.9.

4.10.

4.11.

5.1.

LIST OF TABLES

Population demographics. . . . . . . . . . . . . . . .
Location of errors and error-types by slide number . .

Corrected location of errors and error-types by
slide number . . . . . . . . . . . . . . . . . . . .

Responses to error perception Questionslq 2, and 3 .
Estimate of errors observed--Question 4. . . . . . . .

Analysis of variance table for estimate of errors
observed--Question 4 . . . . . . . . . . . . . . . .

Estimate of spelling errors observed--Question 5 . . .

Analysis of variance table for estimate of spelling
errors observed--Question 5. . . . . . . . . . . . .

Determination of 0, 4, 8, or 12 error version--
Question 6 O O O O O I O O O I O O O O O O O I O O 0

Analysis of variance table for determining error
version--Question 6. . . . . . . . . . . . . . . . .

Retrospective estimate of errors observed--
Question 10. O O O O O O O O O O O O O O O O O 0 0

Analysis of variance table for retrospective estimate
of errors observed--Question 10. . . . . . . . . . .

Type of error causing most concern--Question 7 and 8 .
Analysis of variance table for effect on recall. . . .

Achievement scores-~cognitive test . . . . . . . . . .

Page
38

46

47
57

58

59

59

6O

60

61

62

63
64
65

70

CHAPTER I

INTRODUCTION

Purpose of the Study

 

The purpose of this study is to determine if photographic produc-
tion mistakes in the visual information track of a slide tape program
have any effect on cognitive learning as measured by immediate recall
of audio and visual channel information.

The visual track of a professionally produced cartoon style slide
tape presentation was degraded to include common mistakes--hereinafter
referred to as production errors or simply errors--which may occur
during production of conventional 35 mm (2){2) color slides. The four
error types selected included misspelled words, poor or incorrect ex-
posure, focus, and copy techniques. Three experimental groups were
exposed to one of three versions of the presentation which varied only
in the number of production errors involved. These were compared to a
control group which viewed a version without errors. Each treatment
group was subsequently compared to each other treatment group.

The study specifically tested the following hypotheses:

1. Viewers will show evidence of having observed produc-
tion errors in the treatment versions.’

2. The effect of production errors will be to lower cog-
nitive learning as measured by immediate recall.

2

3. The effect will be inversely related to the number
of errors observed.

This study was performed on a group of 59 undergraduate teacher
education students enrolled in Education 101A, Introduction to Ele-
mentary Education, at Michigan State University during fall quarter,

1973.

The Nature of the Problem

 

The essence of the problem is whether or not photographic pro-
duction errors in multi-sensory instructional materials negatively '
influence the amount of cognitive learning derived from such materials
and, if so, to what extent. In effect, the entire instructional
materials industry has Operated on the principle that productions had
to be completely accurate and technically perfect if satisfactory
learning results were to occur.

The distinct possibility that this cornerstone of instructional
media has been accepted under circumstances which are empirically sus-
pect motivated the investigation. It seems quite clear that important
pedagogical and economic considerations are involved.

From an historical perspective, instructional media--the large
and expanding collection of sensory stimuli including photographic
materials--owes much of its existence to the photographic technology--
a technology less than 150 years old. unsure of the value of their
discoveries, certainly unaware of future applications, early
technologists could only speculate that the photograph might become

useful and popular (Life,,l966).

3

Clearly the photograph did become useful-~and popular-—and some-
what of an enigma for as applications were discovered and proliferated,
people came to attach great significance and value to the visual image.
While in many ways this was justifiable, man's predisposition to assign
pictures an intrinsic goodness in no small measure contributed to his
parallel success in endowing photographs and pictures with values which
came dangerously close to constituting a mythology--a mythology which
in many ways persists to this day and in essence held that pictures
were good because they were pictures. Characterized some years later
by those many persons who strongly believed that "a picture is worth
a thousand words," it was simply assumed that pictures would be
even more powerful if combined with other information channels.

For whatever reasons, there were those who found it impossible
to accept the attributes assigned to photographic materials from such
an apparently naive and illogical information base. While admittedly
speculative, the motivation to question the nature and values of
the photographic medium may have resulted from the realization that
research efforts to provide badly needed factual information were
being restrained by tradition and even emotion. Thus principles of
design and use slowly emerged to become part of the body of knowledge
associated with instructional media.

It is unfortunate that tradition and assumption have played such
major roles in the fOrmative years of visual communication for certain
residual effects may remain as deterrents to progress and needed
research. There is also the probability that many of the principles

which have evolved are not principles at all but rather are, according

4

to Travers [1967] "based largely on practical experience rather than
on research" [p. 17]. There is no intent to deny the value of
experience as an adjunct to research. However, in a real sense many
of the new "principles" have combined to form*what is essentially a
new mythology and thus within recent history can be found the inter—
esting paradox of one mythology evolving to counter the effects of
another.

For whatever reasons, a research effort evolved and some important
issues have been resolved. Representative of those topics which have
not received adequate attention and thus remain largely unresolved is
the concern--or lack of concern--for the effect of the technical
quality of visual materials used in multi-sensory formats on cognitive
learning outcomes. Traditionally, the field has voiced support for
the seemingly logical production and use guidelines integral to a
point of view often--but not exclusively--associated with Eastman
Kodak [1967]. Essentially this position holds that:

Visuals should always be technically flawless with
professional caliber art and photography. Sloppy

art work, improper exposure, glare, poor color

balance, finger prints on the slides, all detract

from the effectiveness of your message [p. 20].

Be accurate. If your audience catches you on one

small error, they will be looking for mistakes in

each visual that follows, and will not be giving

full attention to the presentation [p. 16].

It is virtually impossible to estimate accurately the number of
people who actually practice this philosophy in their own production
activities or recommend it to others via publications and teaching.

Conceivably it may constitute a majority of those considered to be

media specialists for seldom does one find public advice to the

5
contrary. It can be speculated with some degree of assurance that
adherents either believe or assume that the position is based on
scientific research. In fact, however, little research evidence has
been found which supports the Kodak position and thus there is
reason to suspect that what is held by many to be a principle of
visual design and use is not established in research but rather in
assumption based largely on opinions and past practices. This suspicion
has been reinforced in conversations with leading visual communication
authorities including John Debes who speculates that what might be
called the "Kodak position" is probably the well intended result of the
cumulative consequences of practitioners reporting their own practices--
what things worked and what things did not work.

The foregoing is important in that it establishes reasonable
cause to question the validity of the position. There remains the
possibility that the position is correct.

A key concern of this study undoubtedly involves communication
or stimulus noise--production errors which Kornog and Rose [1967]
define as "any condition in the stimulus or subject that tends to
degrade viewing performance (e.g., blur, glare, stimulus errors . . .
etc.)" [p. 456]. As such, it has long been known and subsequently
reinforced that this phenomenon interferes with the communication
process [Shannon and Weaver, 1949; Berlo, 1960]. Although of a
somewhat different nature, the importance of physical clarity of
visual materials has much support in research [Seibert, Kasten, and
Potter, 1959; Spooner, 1973], and adds to the believable nature of

the guidelines suggested by Kbdak.

6

In this event, the present study will serve to reinforce what
many already believe and practice. If incorrect, appropriate behavioral
modifications can be suggested. In either event, research evidence
will substitute for tradition. There is, however, another important
aspect involving the "other" group-~unknown numbers of teachers and
students who are not even aware of the existence of this question or
of its implications.

While at one time interest in the production of visual materials
was primarily the province of the professional producer or media expert,
this has changed. Recently there has been an explosive growth in the
number of institutions and individuals who, at one level or another
and with varying degrees of skill, produce and use visual materials of
some complexity in the communication, teaching, and learning processes.
This group of producers may include thousands of instructors and
students. In no small way great increases in production and the
number of producerstunnabeen motivated by the proliferation of instruc-
tional media programs, the growing visual literacy movement, the
ubiquitous camera and cassette tape recorder, and, of great importance,
a realization on the part of many professional educators that visual
communication should be incorporated in programs which have historically
focused primarily or exclusively on verbal materials.

Something of the implications resulting from the "Kodak position"
may now be perceived. As perception grows, some of the consequences
are also revealed and it becomes painfully clear that the problem has
many crucial dimensions--both pedagogical and economic. The problem

is a product of history and has been perpetuated by tradition. Whether

7
or not the "Kodak position" is "right or wrong" is of far less
consequence than determining what is empirically defensible for it
seems all too apparent that a significantly large segment of the
educational community is following erroneous and perhaps harmful

educational practices.

Need for the Study

 

Essential to the present investigation is the concept that the
design and production of instructional materials may be perceived as
a syntax of critical interacting elements which serve to enhance or
lessen the probability that a given message will be encoded, trans-
mitted, decoded, and perceived as the communicator intended. It has
long been accepted that "The design of messages (audiovisual) implies
certain aspects of production" [Ely, 1963, p. 28]. Thus defined.
technical matters, including production errors, are design considera-
tions and as such this study will contribute to documented needs at
several levels as suggested below.

1. Much of the existing visual design and communication
theory has been experientially derived and is of
questionable validity. General principles must be
research-based and a concerted effort made to quanti-
tatively expand the base.

2. Instructors, administrators and media specialists
lack research-based guidelines which are needed to
help them in day-to-day production and use of instruc-
tional materials.

3. In view of growing recognition that each visual medium
has unique characteristics, values, and limitations,

there is need to expand research efforts within speci-
fic modalities.

4. Within the specific modality under investigation are
economic and pedagogical ramifications of some imme-
diate consequence.

Agonizing as it may seem to those who plan, design, produce or
otherwise use instructional materials, it must be observed that far
too few decisions and procedures are based on experimentally derived
guidelines. As reported earlier, Travers [1967] offers evidence to
suggest that tradition and experience often substitute for research
in deriving audiovisual theory. Much the same situation may be
observed in practice. While attempting to derive guidelines which
could be systematically applied to the design of instructional
materials, Travers [1967] visited four producers of acknowledged
professional competence and found evidence which "made it absolutely
clear that there was no agreement among the four concerning principles
that should be followed in the design of such materials" [p. 9]. It
may be speculated that personal or corporate experience prevails.

Many other investigators acknowledge a similar paucity of
research evidence. For example, Bourisseau, Davis, and Yamamota [1967]
observe that:

The comfortable and respectable notion that pictures have

a greater sensory appeal than verbal stimuli [represented

by printed words] has long suffered from an absence of

evidence from solid empirical research. Nevertheless,

this belief has persisted and served as a guide to theory

and practice concerning the use of audiovisual materials

[p. 259].

Even such a relatively contemporary subject as channel capacity
is not immune from the influences of tradition and experience.

Severin [1967] notes that "over the decades, the combination of

channels has been accomplished with no theory and few rules to

9
guide it, and the products have varied greatly" [p. 400].

Unquestionably, the process by which theory is derived has
improved markedly in recent years. However, such comments are
relevant in that they provide a general historical referent for a
problem which has resulted from principle formulation procedures
considered to be something less than rigorous.

Another aspect of the needs assessment involves the general
theoretical foundation which must exist to provide guidelines for
decision making at all levels of media design and use. Simply stated,
the present base is weak and incomplete-~a fact attested to by a host
of writers including such respected persons as Miller [1957],
Lumsdaine [1963], Travers [1967] and Dwyer [1970]. The importance
of basic research is clearly outlined by Miller [1957] in discussion
involving the nature of graphic communication in education.

The problem of discovering fundamental principles

of graphic communication is a part of the general

problem of the psychology of perception, motivation

and learning. Discovering more about fundamental

principles will help us to improve all media and should

tell us under what conditions one medium will be

superior to another [p. 61].

Conceptually, graphic communication is probably viewed somewhat
differently today, but the underlying meaning and intent of Miller's
.observation is no less viable.

Fundamental theory is of great importance and relevance not
only to visual communication but to the instructional media field in
general. It seems clear that much remains to be accomplished.

However, from a different perspective--that of those who must regularly

design, produce and use instructional media--research has limited

10

influence and results in very few helpful guidelines.

A highly representative example of the lack of consistent
guidelines can be distilled from the contributions of Hsia [1968,
1969, 1971] directed at channel effectiveness. The discussions and
controversies surrounding the topic of channel effectiveness are not
atypical and demonstrates the changing and vacillatory nature of
research findings.

Which channel is most effective--or does a combination of
channels work best? Over relatively short time spans the literature
suggests that the visual channel exceeds the audio channel in
instructional potency, the audio channel is superior to the visual
channel, a combination of channels is superior to both and finally,
channel combination is often inferior to either channel alone.

It may be fairly said that the resulting indecision produces
frustration which invades the senses of media specialists, teachers
and administrators for at one moment it appears that consensus nears
and then new information is discovered which causes the original
position to be suspect. Progress is frustratingly slow and the
resultant collection of new and stable information becomes even
slower as perceived by those who wish to use visual comunication
efficiently and effectively. In essence, practical guidelines have not
materialized. Why something works or does not work, that something is
correct or incorrect is more relevant and usable.

In a frequently quoted supplement to Audiovisual Communication

 

Review devoted to graphic communication, Miller [1957] states that

"Learning is not inevitable. In order to produce efficient learning,

11
good material must be used in the right way" [p. 61].

While "good" and "right" may be perceived as somewhat.platitu-
dinous terminology, in the context of discussing the need for research
in graphic communication Miller [1957] exposes in a fairly straight-
forward manner concerns which continue to plague media professionals
in their search for a prescriptive science of media design and use.
What is good material? How is good material recognized? How can good
material be consistently produced and, what constitutes proper or
correct use?

Implicit in the above statement is the suggestion that there are
answers to what is "good" and what constitutes correctness of delivery.
Progress has been made since 1957 but it becomes increasingly obvious
that added impetus must be given to the quest for design and use
guidelines for educators, media specialists, even commercial
producers are forced to make critical pedagogical and economic
decisions in the absence of guidelines or on evidence which, at best,
may be of questionable value.

From the preceding discussion it may be speculated that there is
need to provide real-world guidelines for application in teaching and
learning within the walls of thousands of educational institutions.
Meierhenry's [1962] exhaustive analysis of research into the educaf
tional applications of media tends to support both the lack and the
need for practical guidelines. The review indicates "that to give
teachers and administrators better recommendations for the use of

specific media much more needs to be known" [p. 307]-

12

From the writings of Moore and Sasse [1971] comes additional
support. Discussing the relatively new concept of visual literacy
they write:

It appears that the limited research available in visual

literacy is insufficient to satisfy the increased interest

of educators in the subject. Teachers are seeking

practical information concerning projected still images

as well as challenging traditionally accepted audiovisual

approaches in educational presentation [p. 438].

There has been no intent in the preceding discussion to suggest
that research into theory is of questionable value. Rather, the
argument underlies the belief that current needs are such that research
must be pursued at different levels simultaneously. In light of the »
fact that multi—sensory materials exist in vast quantities and
because such materials are used in ever growing numbers and frequency
by both teachers and students, there is pressing need for accurate
guidelines to be developed with dispatch.

Accomplishment of this goal may be partially realized through
efforts to expand the information base as it relates to specific
modalities such as the slide tape format. There is a long and
thoroughly documented history of research which has focused attention
and attempted to compare the effects of different modalities-~motion
pictures and still pictures, for example--on a given task. While
many such studies are valuable, others have been ill advised for,
as Dwyer [1972] suggests, investigations on this level often fail to

consider the probability that simple comparisons across modalities

are inadequate and inappropriate.

13

Many studies attempt to compare the effectiveness of

the various media in presenting the same information,

but do not give adequate consideration to inherent

capabilities and limitations (of a given format).

In some instances, it would seem inappropriate to

compare the effectiveness of a motion picture and a

series of slides abstracted from the film in

presenting the same information [pp. 79-80].

Early recognition of modality uniqueness is suggested in Miller's
[1957] observation that:

Completely empirical comparisons of different media in

a given situation can yield results which are largely

an artifact of unspecified characteristics of the

particular representatives of the media which the

experimenter happens to use. Such results tell us

little about the inherent prOperties of different

means of representation [p. 61].

However gradual, there now exists what Spooner [1973] describes
as an "increasing awareness by instructional technologists that
each individual medium has unique characteristics and limitations
which influence its role in instruction" [p. 12]. As attention is
focused on the desirability of learning more about specific modalities
under clearly‘identified conditions, needs and interests become
increasingly specific. Popham [1969] reports that research attention
is being directed at the isolation of treatment variables within
specific modalities and their effects on learning "of most interest
are those that are l) readily manipulable, that is, relatively easy
to modify, and 2) significantly correlated with learner accomplishment"
[p' 28] o

It would appear that this position is essentially what Dwyer
[1970] is alluding to when he proposes systematic evaluation among

various types and complexities of visual treatments in order to provide

users with selection criteria based on specific objectives.

14

Much has been written about the uses and advantages of

the visual medium for instructional purposes. Little

has been done, however, to identify which physical

characteristics contained within the various types

of visual illustrations are most efficient in facili-

tating student achievement of different types of

educational objectives. Since there are no definitive

rules or principles that enable a teacher to select the

types of visual illustrations best suited for instruc-

tion, it seems evident that the instructional media that

depend primarily upon the visual channel would profit

from a program of systematic evaluation. This

evaluation should be focused on relative effectiveness

varying amounts of realistic detail, facilitate

student achievement of specific educational objectives

[p. 235].

Campeau [1972] brings the need for the present study into sharp
definition by suggesting that one important research area in need
of attention relates to the isolation of those factors which inter-
fere with learning when a specific medium or a combination of media

are used under a given set of physical and pedagogical conditions.

Summary

At the present time there is a paucity of research concerning
many important aspects of visual design, production, and use,
including research on treatment variables within a particular modality
such as the slide tape. One such variable is the effect of visual
interference--production errors-—on cognitive learning. Presently
there are few data-based guidelines in this area which address the
problem in terms of multi-channel modalities. Those guidelines which
exist derive in large measure from the suggestions of Eastman Kodak,
which.apparent1y, are based largely on experience rather than research.

Intuitively it makes a certain amount of sense to accept the

15

Kbdak guidelines. There is a modicum of logic involved and at least
a suggestion of support in research directed at the effects of
communication noise as well as studies which investigate the influence
of physical clarity and legibility on learning.

If the Eastman position on the importance of high quality is
defensible, mere subjective analysis will suggest that many mediated
learning strategies involving visual information channels are less
effective than instructors suppose for errors may frequently be
observed especially in locally produced materials. There is an
affective aspect, also, for it is suggested [Eastman Kodak, 1967] that
in the presence of low quality visuals, message importance is lowered
and it can be speculated that source credibility suffers. Implicit,
also, is the possibility that the attitude of students toward the
topic and perhaps even toward the media has been negatively affected
because of the effects of repeated experiences involving error-ridden
media productions.

On the other hand, if quality is not actually a problem, vast
sums of money, energy and human resources have perhaps been expended
needlessly for it is difficult to disprove the position [Eastman
Kodak, 1967] "good visuals cost money and more than money, they take
time, thought, and imagination to design" [p. 10]. And to these
costs can be added those associated with quality production.

It has been suggested that the present study is needed to make
contributions at several research levels including fundamental theory
development and inter-modality understanding. Of greater priority,

however, is the effect the investigation may have on the need to

16

provide guidelines which may be applied to issues of immediate and
practical concern--the day-to-day production and use of multi-sensory

instructional materials.

CHAPTER II

REVIEW OF THE LITERATURE

In Chapter I the nature of this study is shown to be an inquiry
into the effects of visual interference on recall of cognitive
information. The test vehicle is a slide tape presentation and
interference is induced through production errors in the visual
channel.

As evidence by such compendia as Traver's Research and Theory
Related to Audiovisual Information Transmissions [1967], Chu and
Schramm's Learning from Television: What the Research Says [1967],
Ball and Byrnes's Research Principles and Practices in Visual

Communication [1960], Saul‘s A Review of the Literature Pertinent to

 

the Design and Use of Effective Grgphic Training Aids [1956], and
Hoban and van Ormer's Instructional Film Research: 1918-1950 [1951],
there is no dearth of research and literature pertaining to the
several discrete information transmission formats involving multiple
channel communication. What can be synthesized from these and other
works is the efficacy of audiovisual communication devices to function
effectively in teaching and learning. Not so clear, however, are

the variables or conditions which tend to influence the efficiency

and effectiveness of multi-sensory materials. In reviewing the

literature on instructional television, Allen, [1971] states:

17

18

That students learn from televised teaching cannot be

doubted, but the conditions under which such learning

takes place and the specific characteristics of

televised presentations that bring this about are yet

to be determined, . . . [p. 10].

There is no apparent reason to doubt the applicability of this
statement to motion pictures as well as other multiple channel
modalities involving visual communication.

Given this base, a good deal of research effort has been fOCused
on experiments which attempt to isolate test variables in what might
be termed the conditions of multiple channel communication.

In Chapter II the literature most closely associated with this
study is reviewed. While a number of areas are of related interest,
four have been identified as particularly pertinent to the purpose
of this study:

1. Communication noise and visual channel interference

2. Relevant and irrelevant attention-gaining devices

3. Communication properties of pictures

4. Legibility and technical quality

At this point it must be noted that the relationship between
experimental research and media applications impose important limita-
tions on the relevance of much of the literature reviewed. Specifi-
cally, the review process clearly exposes the fact that a majority of
the research studies in multi-sensory formats assumes a condition of
visual primacy. Experiments tend toward conditions in which the
visual channel is the primary message source.

Given that this level of research is needed, there is much

evidence to support the contention that in the every day production

19

and use of instructional television, motion pictures, slide tapes, or
sound film strips, audio primacy has and continues to represent the
norm. In an evaluation of a study conducted by vanderMeer [1950],
Hoban and van Ormer [1951] report that the investigator found:

a growing tendency to put the greater part of the

learning material in the verbal element of the film-

strip (the titles and the teacher's notes or guide to

explain the material), while the pictorial materials

serves merely as an approximate illustration of the

verbal content [p. 8-22].

From a review of film studies, Travers [1967] confirms this
tendency. "In present-day instructional movies the sound track
contributes a greater share of the transmitted information than does
the visual" [p- 58]. More recently, Anderson [1972], in commenting
on contemporary instructional television research, writes that "A
majority of the studies have used television not as a visual medium,
but as a pipeline for whatever incidental activities accompany a
primarily verbal communication" [p. 58].

Thus, while much of the data reported herein have relevance, they
are somewhat modified and conditioned by the fact that the communica—
tion vehicle used in this investigation was purposely selected
because of its "real" nature--the presentation is typical in that a
close examination of both channels indicates that by far the greater
portion of the message is embedded in the audio channel with the visual

channel serving an illustrative or attention-attracting function

[Hartman, 1961].

20

Communication Noise and Visual Channel Interference

Both communication and learning involve two fundamental elements—-
stimulus and response. In discussing communication models, Ball and
Byrnes [1960] note that:

whenever humans learn or humans communicate some stimulus

is present . . . We can define the term "stimulus" in

rather broad terms to take in anything which we can

perceive through one of our five senses . . . We can also

point out that whenever there is a stimulus there is also

some response. By response, we mean simply that the

individual perceiving the stimulus does something, i.e.,

votes, runs away, talks, thinks, changes an attitude,
etc. [p. 33]

It is given that a communicator normally attempts to produce
messages of the highest possible fidelity--another way of saying that
the stimuli will be received as intended. To insure fidelity of
reception, theory has dictated that communication noise must be
reduced for, as Berlo [1960] indicates, "noise and fidelity are
two sides of the same coin. Eliminating noise increases fidelity;
the production of noise reduces fidelity" [P- 40-41]- Hsia [1968]
describes the importance of noise in communication thusly:

Another factor of fundamental importance in communication

is noise which, in a broad sense, is unwanted information,

an inevitable occurrence in any I.P. (information

processing) situation [p. 246].

In Chapter I the equivalence of production errors and noise is
suggested. It is speculated that statements which predict a negative
impact on learning as a result of technical imperfections such as
production errors are either largely unsupported by research or at

least subject to important qualifications. For example, an Eastman

Kodak publication [1967] suggests that:

21

Ours is a visually sophisticated age. Magazines, motion

pictures, television, advertising . . . all have created

an audience who take good visuals for granted. It is

also an audience which has been trained since infancy in

the fine art of tuning out amateurish, poorly done, or

annoying visual messages [p. 18].

Although the topic will be discussed in some detail later, Levie's
[1973] summary statement dealing with technical quality of the visual
channel in training films is relevant to and somewhat unsupportive of
the Kodak position. While few specific studies are cited, he
concludes that:

Researchers have demonstrated that variation in

technical quality do not affect the acquisition of

information so long as the relevant cues are

available to the learner [p. 31].

This would suggest that under conditions of audio primacy, the
effects of visual channel interference would be negligible. However,
in summarizing even more limited numbers of audiovisual studies in—
volving audio channel noise, Chu and Schramm [1967] state that:

Noise will reduce the effectiveness of learning from

film and television so far as part of the learning

comes from the auditory medium [p. 72].

Support for this conclusion is found in two recent studies which
include, among other treatments, visual channel interference.

A study conducted by Hsia [1969] which investigated intelligence
as related to auditory, visual, and audiovisual information processing
is highly relevant. A group of 192 seventh grade students were pre-
sented verses of English poetry under three channel conditions-~audio
only, visual only, and audiovisual. While all conditions were exposed

to noise, of most immediate interest was the introduction of noise in

the treatments involving visual presentation. This interference

22

"consisted of random black dots scattered over the visual stimuli
blackening about 25% of the visual space" [p. 27]. Audio channel
interference was 25% "white" noise. Recall was tested immediately
following each presentation by having the subjects check whether or
not they observed or heard a certain word in the presentation.

In terms of the effects of noise, the investigator concluded
that "it seems to be conclusive that noise was a detrimental factor
that hindered both HI (high intelligence) and LI (low intelligence)
83 with much more severity on LI 35" [p. 279]. Interestingly enough,
the channel which was affected least under noise conditions was the
visual channel. This was attributed to the type of noise (scattered
dots) which "afforded comparatively easy discrimination" [p. 280].
A logical implication to be drawn from this finding is that extreme
care should be taken to insure high fidelity channels in situations
where the average intelligence level of an audience is either unknown
or recognized as low.

The above point is reinforced by Severin [1967] in discussing
the importance of using relevant pictures (as Opposed to irrelevant)
in multiple channel communications. He states:

If interference is accidentally introduced between

two channels, then much effort, time, and money is

wasted, for one channel could then communicate more

efficiently. If, on the other hand, single-channel

communications are not effective enough for a

specific task, the combination of channels with-cue

summation could prove to be the solution to more

effective communication. This could be the solution

in situations where the literacy or intelligence

level of the members of the intended audience is

known to be low, where presentations may have to be

made under conditions of high "noise" (interference

in the audio or visual channels from sources either

within or outside the message source), or where

the material is relatively difficult for a given

audience [p. 400].

l

23
In his conclusion, Hsia [1969] seems to say much the same thing:

In order to increase the IP (information processing) rate

of LI 85 it seems imperative to reduce noise and to

employ the AV channel in which BCR (between channel

redundancy) should be made as high as possible [p. 281].

A second experimental study which incorporates one highly
relevant aspect was conducted by Schlater [1966]. That part of the
two part experiment which investigated the effects of irrelevant
cues on recall of audio channel information is particularly interesting
for at least some of the cues--superimposure (credits), boom shadow,
ornate frame, letters/numbers in grease pencil, shadow of hand or
finger pointing--can be considered production errors.

Schlater's study dealt with speed of presentation and inter-
ference variables in the video channel of a series of short presenta-
tions on Greek temple architecture. Presentation rates included one
to nine visuals per 30 seconds and interference variables were measured
by Ss viewing 45 visuals while varying the number (0-45) and rate (0-5
per 30 seconds) of irrelevant inclusions. In each case there were six
treatment groups.

Using 910 high school seniors as subjects, standard 21(2 slides
of the material selected were simultaneously transmitted via television
to Ss located in 30 rooms on the Michigan State University campus. The
audio channel was held constant throughout both experiments. The nar-
ration was related to but not redundant with the pictures as presented.
Recall of information was tested with multiple choice pictorial video
questions and verbal audio questions.

Schlater [1966] reported that interference from increasing speed

of presentation (from one visual per 30 seconds to nine per 30 seconds)

24
produced a significant difference (p‘<.05) when tested using verbal
questions, no significant difference under pictorial questioning and
no significant effect on comprehension of information presented in the
audio channel.

More closely allied is the second part of the experiment. The
six treatment groups viewed 45 relevant visuals for six seconds each
and were exposed to either 0, 9, 18, 27, 36 or 45 irrelevant inclusions.
These in turn were presented at rates up to five per 30 seconds of pre-
sentation. Boom shadow, for example, was seen once by the first treat-
ment group and five times by the fifth. Results indicated that inter-
ference did not significantly affect recall of video information when
tested by pictorial video question or verbal video questions, but did
significantly affect (E'- .01) recall of audio information. However,
the expected pattern--poorer recall as the number of irrelevancies
increased--did not emerge. By way of explanation, the investigator
suggested that some of the irrelevant cues were probably not obvious
enough and thus may have gone unobserved. Such a lack of interference
was at least implied in the preceding study by Hsia [1969] who also
found the visual channel to be more resistant to the effects of
interference.

Whether or not audiences generally perceive technical flaws or
errors such as misspelled words or poorly focused visuals has been a
matter of conjecture--prhmarily among professionals in the field of
design and production. No audiovisual studies were found which tested
this specific point, but two relevant print studies were found which
appear to relate.

Sencer [1965] conducted an experiment on 350 male college

25

freshman to test the effects of incorrect grammar on attitude and
comprehension. Five degrees of error--zero to 51 per 500 words of
message--were introduced into a basic 1200 word essay. The finding of
no significant difference in comprehension elicited the conclusion
"that noise in the communication system can be tolerated in terms
of comprehension, at least, to quite an extensive degree" [p. 50].
The experiment did indicate that errors were observed and perception
of errors was positively related to 85 measured level of verbal ability.

Razinsky [1967] also found that 55 perceived errors in an experi-
ment to determine the effects of message quality deterioration. The
finding of no significant difference in comprehension supports Sencer's
[1965] findings although Razinsky does note, in contrast to Schlater
[1966], a linear decrease in comprehension as the number of errors
increased. He concludes that " . . . errors influenced learning only

to the extent that they are actually perceived" [p. 28].

Relevant--Irre1evant Attention Gaining Devices

While it is yet to be established that production errors or other
forms of visual interference are consistently perceived, it seems
fruitful to pursue a course of inquiry based on the assumption that
errors are observed--at least under certain conditions. If errors are
noticed, do they gain attention to the degree that interference
results? Fleming [1967] states that:

Attention is drawn to changes in stimulation, changes

from immediate-past experience and changes from long-

term experiences. In a sense, the more the new stimula-

tion differs from the prevailing, the more distinctive
or attention-gaining it is [p. 23].

26

The effects of relevant and irrelevant attention-gaining devices
on channel effectiveness have been the subject of numerous studies and.
offer interesting and relevant data. In an early study, Neu [1950]
found:

No evidence that the relevant attention-gaining devices
as used in these experimental films (closeups, spot-
lighting, pointing finger, unusual camera angles, voice
clues, sound effects) added to the instructional effec-
tiveness of the film [p. 8-24].

HOban and van Ormer [1951] agreed with Neu [1950] and suggested
that "Devices which call attention to irrelevant materials may inter-
fere with the learning of more important items" [p. 8-33].

The topic of relevant-irrelevant materials or devices is also
discussed by Chu and Schramm [1967] with respect to research in instruc-
tional television. They conclude that "attention-gaining devices can
add to learning." Miller [1957] speculates that it is possible for cues
or attention-gaining devices to both contribute and detract from
message fidelity.

When cues from different modalities (or different cues
within the same modality) are used simultaneously, they
may either facilitate or interfere with each other.
When the cues elicit the same responses simultaneously
or different responses in the proper succession, they
should summate to yield increased effectiveness. When
the cues solicit incompatible responses they should
produce conflict and interference [p. 78].

Hartman [1961] agrees with the essence of this statement.

It may be argued that additional cues are excellent
attention-gaining devices. So they are, if carefully
handled. But a one which has gained attention is
likely to hold it. Too often the attention-gaining
device is allowed to continue in competition with

the attention-getting device. Attention may be
gained but communication of the message may not
result [p. 255-56].

27

A number of investigators have moved beyond inquiries into pure
attention-gaining devices to considerations of the effects of irrele-
vancies in general. Ketcham and Heath [1962] report on an investiga-
tion involving pictures which are unrelated to the material being
presented. A major finding of this study suggests interference re-
sulting from irrelevant visual materials. The investigators reported
that:

the results suggest . . . that the projection of

abstract forms on the screen distracts attention more

than it improves concentration, and that the distrac-

tion is worsened by the increased detail, or the incon-

gruity, of an irrelevant picture [p. 92].

Shamo and Meador [1969] support this general conclusion. In an
experiment in which subjects were shown a number of irrelevant slides
while they listened to short persuasive messages, they found that the
irrelevancies produced a significant reduction in the amount of recall
of the content of the audio message.

Severin [1967] investigated the interaction of relevant and
irrelevant pictures in multiple channel communications and concludes
that pictures which are unrelated to the message content--therefore
irrelevant--produce the least amount of learning. Accounting for the
finding, Severin [1967] indicates that it is "because of the irrelevant
cues introduced . . ." [p. 392]. From an investigation of circum-
stances involved in the effective transmission of information in
multiple sensory presentation Travers [1964] concludes:

Devices which have been used to draw attention to the

information transmitted through one sense modality

tend to depress information received through the
other [p. 378].

28

One finding which differed significantly is reported by Severin
[1967] in the experiment previously described. In the treatment which
varied the number of irrelevant cues presented, the investigator found
". . . that more recognition resulted from the treatment with the
more irrelevant cues, rather than the less irrelevant ones, . . ."
[p. 399]. He suggests that blocking may have taken place which effec-
tively reduced the level of interference. The explanation would be
_consistent with that of other investigators such as Travers and
Alvarado [1970] who report that:

Living organisms are systems with a limited capacity for

handling information and have to be selective in what

they attend to. If they were not, they would Constantly

find themselves swamped with large volumes of useless

information [p. 57].

Hsia [1969] suggests a related phenomena when he postulates that
there is a "positive correlation between intelligence and noise-
filtering capacity; . . ." [p. 273].

On the basis of the evidence it appears that visual irrelevancies
do tend to affect learning negatively but not under all conditions. At

what point and under what conditions the human information processing

capacity is seriously affected is not completely known.

Communication Properties of Pictures

Introduced by Schlanger [1966], the term "visual task factor" is
defined by Perrin [1969] as relating to "the work the viewer must do to
extract the necessary information (from the picture)" [p. 375]. Con-
ceptually the term seems highly relevant to this discussion for

attention is focused on the fact there is more to visual communication

29
"than meets the eye." In other words, a visual is not good because it
is a visual, but rather because it accommodates certain critical
requirements of visual perception and thereby increases the efficiency
of the communication.

That there is a definable level of necessary information is
interesting in its own right and is supported by numerous researchers.
Arnheim [1962] suggests that the basis of visual education is found in
the understanding that "every picture is a statement" [p. 20].- As such,
the statement is not the thing or object but rather "a set of proposi-
tions about the object; . . ." [p. 20]. He concludes that "these
propositions are stated in a visual language" [p. 20].

While the existence or level of a visual language has been
questioned [Travers, 1967], there is support for the premise that
communicable elements must be present in visual materials at some
minimum level of recognition before communication (and therefore
learning) can be achieved. Arnheim [1962] refers to such elements as
"visual properties" [p. 20]. From the research evidence in information
processing which suggests that humans have limited infOrmation process-
ing capacity, Travers [1967] postulates that "the design of audiovisual
teaching materials requires that he concentrate on essential informa-
tion" [p. 97]. Here, then, is the additional suggestion that a
maximum level of information may also be identified.

The implications to be drawn from Travers' statement are clear.
First, the amount of information to be communicated in any given
situation must receive close scrutiny by the message designer. For
example, Spaulding [1961] indicates that "the amount of information

communicated by a pictorial illustration is inversely proportionate

30
to the complexity of the drawing (or otherwise stated, the amount of
information in the drawing)" [p. 61].

rOptimum information levels have been considered in depth by
several researchers. Dwyer [1970] tested critical information levels
by manipulating amounts of detail in visuals. His findings suggest
that :

The effectiveness of a particular visual in facilitating

student achievement of a specific objective depends on

the type of information needed by the student to achieve

that objective [p. 247].

Fonseca and Kearl [1960] tested a similar concept and came to essen-
tially the same conclusions. However, they offer a note of caution by
suggesting that, under certain conditions:

Excessive deletion of detail, by requiring the viewer

to mentally complete the detail of a schematic drawing,

reduces comprehensability [47:6].

This is entirely consistent with the impact of the visual task factor
discussed by Perrin [1969] who also indicates that under certain cir-
cumstances visual cues can become so abbreviated or compressed as to

increase the visual task factor.

A second implication suggests that once the apprOpriate informa-
tion level is known, information must be encoded and transmitted with
as much fidelity as possible. Stated another way, the visual prOper-
ties of the message must arrive in a readable condition for, as Arnheim
[1962] infers, "The picture itself must steer perception" [p. 19]. The
directing or steering mechanism apparently derives from the presence of
the original referent are portrayed in the picture.

Gibson [1954] suggests that an object has many visual properties

which are of concern in producing pictorial surrogates and concludes

31
that "the most important ones (for most human purposes) are the form,
shape and proportiOn of its edges and surfaces" [p. 21]. The picture,
he writes "like a model, has many dimensions of fidelity to the
object, . . ." [p. 21].

These "dimensions of fidelity" are not unlike the visual
properties which have been delineated by others, including Arnheim
[1962] and Harrison [1964]. Discussing the perceptual properties
of pictures, Arnheim [1962] draws on Gestalt psychology in stating that
"The shape of contours, the contrast of brightness, the structure of
the over-all pattern will determine what is seen" [p. 19]. Additional
dimensions are added by Harrison [1964] in commenting on the importance
of figure-ground relationships. He observes that:

We might also predict that contrast between symbol and

background will be important. If the original referent

has color, texture, or shading, it may be important to

retain this in the symbol, particularly to the degree it

increases the contrast with background [p. 39].

In his work on the classification of illustrations, Fleming [1967]
includes the term "pictorial elements" as a descriptor for critical
perceptual attributes and describes them as:

Those configurations of line, dot, or area and any

combination of these three that resembled events or

objects (persons, places, and/or things) either as

perceived or as generally conceived [p. 247].

Harrison [1964] indicates that color may be a perceptual cue if
color is important in the original referent. There are many studies
which attempt to evaluate color as an instructional variable and
include, among others, McLean [1930], Long [1946], VandefiMeer [1948

and 1954], Fullerton [1956] and McCoy [1962]. While findings have

varied with the balance clearly in favor of color not being a

32

significant learning factor, research and discussions continue and there

is growing support for a modified position. While agreeing with the
general conclusion, Chu and Schramm [1967] are of the opinion that:

We must not overlook the possibility that in certain

particular learning tasks, color may play an essential

part. For instance, certain substances, like blood,

have no definite shape and can only be identified by

color [p. 25].

Based on his own research, Dwyer [1970] concurs and adds broader
dimensions to the possible value of color.

For specific objectives the addition of color in certain

types of visuals, for students in specific grade levels,

appears to be an important instructional variable in

improving student achievement [p. 247].

On the basis of the evidence it seems conclusive that there exist
certain critical visual properties which, once the optimal information

level of a picture is stated, are responsible for successful

communication.

Legibility and Technical Qualiqi

The retention and transmission of visual properties are closely
related to discussions of legibility and technical quality of media.
Again, channel primacy influences the degree of relevance and most
studies deal with situations where the visual channel is considered by
assumption or design to be primary.

The literature on legibility includes numerous studies on such
topics as the type and size of characters, viewing conditions, viewing
angle and distance, and resolution of displays. Although of interest

from the stand-point of simple visibility, standards and guidelines

33
resulting from such studies have little relevance to the present
study other than the obvious premise by Spooner [1973] that, "If a
student is to learn from visual materials, he must be able to see them;
that is, physical clarity of visuals is fundamental to a student
learning from them" [p. 10].

This, again, suggests that under conditions of visual channel
primacy the critical visual properties of a picture must be visible,
above the visual threshold as discussed by numerous writers, including
Siebert, Casten, and Potter [1959] and Spooner [1973]. Early research
was not always so conclusive, however, and led Hoban and van Ormer
[1951] to observe that:

Physical surroundings and conditions of projection,

except, of course, when the room is fully illuminated,

are not likely to be critical factors in effecting

factual learning from motion pictures [p. 8-43].

Later, Miller [1956] suggested that this position might be true
but added an important condition.

With highly motivated students who are experienced

viewers of pictures, one might expect such factors as

size of screen, distance from screen, angle of view,

and illumination to be critical only if they interfere

seriously with the presentation of the relevant cues [p. 82].

The importance of this condition is apparent in research on
instructional television. Summarizing the available research, Chu
and Schramm [1967] conclude that such factors as angle and distance can
interfere with learning in a situation "where accurate perception of
images is an important part of learning, . . ." [p. 68]. Blackwell
[1968] and Perrin [1966] concur on the importance of illumination.

Perrin [1966] concludes that "different visual tasks require full

screen illumination in a glare-free visual field" [p. 374]. He further

34

suggests important relationships between image resolution, density of
materials, and visual acuity.

A review of the literature pertaining to technical quality shows
that quality is largely associated with production variables--slickness
and embellishments. The history of this type of research indicates
relatively consistent results. Hoban and von Ormer's [1951] survey of
the literuature led to the conclusion that the visual influence of
motion pictures is "relatively unaffected by 'slickness' of production
as long as meaning is clear" [p. 9-5]. Similarly, Chu and Schramm
[1967] point out that with the possible exception of motor skill learn-
ing where research indicates the use of a subjective camera angle:

There is no clear evidence on the kind of variations

in production techniques that significantly contri-

bute to learning from instructional television [p. 48].

As a production variable related to quality, the use of color, as
reported earlier, makes little or no difference except where cOlor
represents a critical perceptual property.

Numerous additional references such as Lumsdaine and Gladstone
[1958], Baker and Popham [1965], and Levie [1973], support the general
conclusion that slick or highly embellished visuals are not significantly
more effective in learning than simple, unembellished versions.
However, a study by Baker and POpham [1965] exposed an affect on
attitude of viewers. They conclude that it may be unsafe to assume
that embellishments serve no real purpose.

Chu and Schramm [1967] extend a similar caution based on the lack
of knowledge as to what influence vastly increased exposure to quality

visual materials will have on audiences in the future.

35
On the basis of the latest evidence, Levie [1973] has concluded
that "when the attainment of cognitive learning objectives is the

criterion, insistence upon flawless technique is unjustifiable" [p. 31].

Summagy

A careful review of the literature spanning a period of at least
50 years failed to disclose any studies which could be characterized
as being "highly similar." There is a long history of studies which
attest to the instructional value of media, but the conditions under
which instructional materials function most effectively remain
unclear.

Most research dealing with variables within modalities has been
carried out under conditions of visual channel primacy whereas audio
channel primaCy is more often representative of the norm in multi-
senscry designs. The present study is based upon the use of visual
auditory messages in which there is audio channel primacy and therefore
most reviews are only tangentially related.

Four related areas were explored. Most literature confirms the
importance of communication noise per se. When applied to interference
phenomena in such areas as relevant-irrelevant attention-gaining
devices, legibility, and technical quality, noise theory is modified.
For example, studies within these areas of inquiry report noise
effects ranging from interference to enhancement of learning.

Increased learning under noise conditions--usua11y characterized
as a "surprising finding"-- is currently explained as the result of

blocking or filtering, the limited information processing capacity of

36
humans, and high intelligence.

Numerous studies in human information processing have undoubtedly
motivated related but limited research directed toward determining
optimal information levels in pictorial surrogates via manipulation of
the known visual properties of pictures. Findings are varied but
useful in that evidence grows to support the emerging knowledge that
pictorial content and optimal information level is a function not of
the media but of the audience and instructional task. Once an optimum
information level is reached, it may be speculated that visual inter-
ference exerts a negative effect on learning. That an individual must
be able to see a visual to learn appears as a given in most of the
literature. Yet, the most recent review of studies dealing with
legibility and technical quality suggests that technical perfection is

unjustifiable when cognitive learning is the goal.

CHAPTER III

PROCEDURES AND METHODOLOGY

Statement of the Problem

 

The principal concern of this study is to determine whether or not
photographic production errors in the visual channel of a multi-sensory
program exhibiting audio-primacy message design characteristics will
exert any influence on cognitive learning as measured by immediate
recall of information presented within the program. Stated somewhat
differently, it is postulated that interference may result from the
recognition of technical imperfections in pictorial material and
reduction in learning will occur which will be inversely proportionate

to the number of imperfections observed.

Hypotheses

 

Stated in null form, the hypotheses are:

1. There will be no difference between groups in their
perception of production errors.

2. There will be no difference between groups in cognitive
learning due to the presence of production errors.

3. In the event of significant effect, there will be no

relationship between the number of errors perceived and
scores on the cognitive measure.

37

38

Population

 

The experiment was conducted with a population of 59 students
enrolled in Education 101A, Introduction to Elementary Education, at
Michigan State University on November 19, 1973. Relevant data concern-

ing the population are shown in Table 3.1.

Table 3.1. Population demographics.

 

 

 

 

Sex College level

_M _F_ Fresh . ' 80ph . g}; . g; . Grad.
Control 1 12 l 6 5 l 0
T1 4 10 0 6 7 0 1
T2 1 l6 0 7 8 0 2
T3 3 12 O 5 7 2 1
Totals 9 50 l 24 27 3 4
Percentage 15.3 84.7 1.7 40.7 45.7 5.1 6.8

 

 

 

 

 

 

 

 

 

Selection of the population was based on the following criteria.
First, it was desirable that the population be related to the audience
for which the original test vehicle was designed--instructional or
administrative personnel in grades K-12. Secondly, it was important
that the population be largely unfamiliar with systems theory in
general and with the three-stage, nine-function Instructional
Development Institute model in particular. Finally for statistical

reasons, it was necessary to have a population of approximately 60

39

persons. Permission to utilize this population was obtained from
the course instructor and details of the experiment and procedures
were communicated to her in advance of the actual experiment

(see Appendix A).

Instrumentation

 

'IwO evaluation instruments were constructed to accommodate the
design of the study--a 35 item multiple choice recall test (see
Appendix B) and a 10 item error perception questionnaire (see
Appendix C).

Preliminary discussions with colleagues and members of the
Department of Counseling Personnel Services and Education Psychology
resulted in the decision not to use a pretest since the procedure
would necessarily expose subjects to the precise content and language
they would receive in the test presentation material which was
purposely selected because of its unfamiliarity to the subjects.

An initial pool of 45 questions was prepared. Each consisted of
five possible choices including a constant option of "I really don't
know." Designed to provide another measure of the possible effects of
production errors, it was hypothesized that the latter option would
remove the necessity for guessing where subjects truly did not know
the answer. Subjects were instructed not to guess and encouraged to
use this option when it represented the best answer.

Following the first validation (n==22), individual questions
were analyzed and the number reduced to 35. Ambiguities and

inconsistencies were eliminated and questions with more than one

40
possible response were reworked. A second validation (n==20), produced,
with minor changes, a final instrument with Kuder Richardson reliability
of .694 on the control group. V

In the final form the instrument consisted of 23 audio and
12 visual channel questions. Of these, 24 specifically tested
content at error locations, a requirement of the design. The order
of questions was randomized.

A second instrument was designed to elicit information concerning
the perception of production errors. Given the constraints of clouding
the real purpose of the instrument and not suggesting information about
errors--existence, number, type, and effect--until appropriate,
conventional paper-pencil tests were discarded and a different design
was sought. Valuable advice was offered by colleagues and testing
specialists and a prototype six item tape recorded questionnaire
resulted.

A tryout on 20 students was used to test the instrument and
suggested the addition of questions which would refine the probing
steps and also produce additional information. The final version,
consisting of 10 questions and a response form, was tested on a
second group (ns=22) and appeared satisfactory in every way. Discri-
mination between the control and experimental group (T3--the group

subjected to the most errors) was obvious.

Message Design

 

The study was implemented by using a modified version of a slide

tape presentation entitled "A Fable Revisited" which was designed

41

and produced for the National Special Media Institutes, Instructional
[DevelOpment Institute (IDI) program (see Appendix D). [As originally
designed,the purpose of "A Fable Revisited" was to teach the Component
elements of a particular instructional development system (IDI) in an
entertaining manner and to motivate acceptance of using a systematic
approach to problem solving. It contained 71 visuals and had a
running time of eight minutes and ten seconds. The program is rendered
in cartoon style and employs a conversational narration style.

Modification of "A Fable Revisited" was required for purpose of
the study. The need for introducing errors in the visual channel
necessitated the replacement of certain slides with slides containing
injected errors. This in turn necessitated rephotographing all art work
so that every slide would be of tie same quality. Secondly, a detailed
audition of the original audio channel provided evidence that some of
the system's terminology which identifies elements and activities
within the IDI model were either overused or completely missing from
the narration. Some potentially troublesome ambiguities and inconsis-
tencies were also identified. As modified, the test-presentation
contains the same 71 visuals as the original, but has 79 seconds
additional running time. The increase is largely attributable to
inherent differences in speaking rate and slight changes in the timing
of slide changes. The original narrative style remains essentially
unchanged.

The selection of "A Fable Revisited" was based on several
criteria. An important condition was established by Debes [1973] who
indicated that the vehicle selected for a study of this nature should

meet the qualification of "goodness." This, in effect, suggested that

42
the material selected exhibit high initial technical quality, some
degree of validation in terms of stated objectives, and acceptance
based onmultiple use. "A Fable Revisited" is recognized for its
high quality, has been used several hundred times and validated
under field conditions. This investigator has personally used the
material on as many as 50 occasions and has found the attitude of
audiences toward the material to be extremely positive. Administra-
tion of semantic differential instruments has confirmed this
observation.

A second criterion suggested that the material selected be
typical in terms of channel conditions. As indicated earlier, much
multi-sensory material is designed and/or produced in such a way that
the audio channel transmits most of the information. Analysis of
"A Fable Revisited" demonstrated this condition. A third consideration
was that both channels be completely open to inspection so that
channel relationships could be objectively determined as suggested by
Hartman [1961]. Fourthly, a content was needed which would tend to
negate the effect of past experience--scmething relatively if not
completely unfamiliar to the subjects. Past experience of this
investigator with students enrolled in the introductory level teacher
education courses suggested that an in-depth familiarity with "systems"
was unlikely and that there was virtually no chance of previous
exposure to the specific model presented in this presentation. Finally,
in terms of relevance and generalizability, the final criterion was to
use material of demonstrated "real world" applicability rather than

something specially created fOr empirical study.

43

Audio Channel Revision

 

A typewritten c0py of the narration was prepared and analyzed for
the presence and use of key and related terms which represent the
language of the IDI model, applications of the concepts presented,
and any other information of an explanatory nature. Key terms are
those which identify stage and function names; related terms are those
which represent sub-concepts and activities under function headings
(see Appendix E).

The analysis exposed some deletions as well as a certain amount
of redundancy. For example, the key term "develop" was used twice in
the narration while the key term "evaluate" was totally missing.
Following a minor rewrite, each of the key or related terms was used
once in the narration thereby allowing greater control and discrimina-
tion in preparing the cognitive evaluation instrument and subsequent
analysis of data should significant differences by found.

A check of the audio channel also exposed some inconsistencies.
Representative of this type of inconsistency is the following
statement which, in the original version, stated:

This pig--we'll call him "Straw Pig"--never really saw

the wolf. He concerned himself mainly with the problem

of protecting himself from the elements of the

weather [slide #9].

A few seconds later the narration includes the following

information.

His solution strategy was simple: build a house for
protection from the wolf and the weather [slide #13].

The inconsistency of the two narratives derives from the addition

of a further dimension to the problem--the wolf. Should a random error

44
be associated with either frame it would be difficult to determine
whether a viewers response was due to the unintended ambiguity or to
the inaccuracy of one or the other. As revised, the initial statement
is left intact, but the second if modified to maintain consistency:

His solution strategy was simple: just build a house
for protection from the wind and rain [slide #13].

Following the revision process which resulted in so little
recognizable change that persons intimately acquainted with the program
were unaware of the modification, the new narration was recorded in a
professional sound production studio. Prior to recording, the narrators
listened to the original audio channel and rehearsed several times in
order that the original style, inflection, humor, and overall effect
would be faithfully reproduced in the new version. The finished
narration was pulsed for synchronized presentation on a 3-M Model 2550

cassette tape recorder.

Visual Channel Preparation

 

From a preliminary list of 10 possible production related errors,
four were selected for use in this study, namely errors in copy
technique, exposure, focus,.and spelling.

Selection criteria for these four types of errors were based
upon conversations with photographers, graphics technicians, and
persons regularly involved in producing, viewing, and evaluating
materials of the type used in the study. Selection was also influenced
by the opinions, observations, and experiences of the investigator as

an instructor of photography and message design. Further, the

45
investigator was closely associated with the design of the Instructional
DeveloPment Institute and with production of the materials used in this
study and personally observed all of the selected error types during
the process of developing and testing the prototype of "A Fable
Revisited" as well as other slide tape presentations used in the
Institute.

Determination of specific locations for each of the errors
injected was made according to procedures which accommodated constraints
imposed by the nature of the visual channel and of the experimental
design.

An obvious constraint, for example, was that a Spelling error
had to be placed on a slide that included words. It was necessary, as
well, that the words were an essential part of the message design, that
the words were large enough to be read, and that projection time
was sufficient for the words to be read. Finally, in order to avoid
questions of relevance, it was determined that the title slide would
not be used for any error.

A major condition applicable to the location of all errors was
that the narration opposite a slide containing a visual error provide
sufficient content to allow a correct response to a logical and
answerable question.

Given the above constraints, 51 possible error locations were
identified. Othhese, only eight could be considered as spelling
error locations. The remaining 43 locations were satisfactory for
any other error type. All locations were numbered on 1” x 1" card

stock and a random selection procedure was used to first determine

46

the location of Spelling errors in the three treatment groups. This
procedure was necessitated by the chance that selection of the other
three error types would exhaust the limited number of spelling error
locations.

Procedurally,tjm:eight numbered squares were placed in a container
and a number was drawn. This number became the location of the single
spelling error for T1, one of the two locations for T2 and one of the
three locations for T3. A second number was drawn which represented
the second of two locations for T2 and the second of three for T3.

A final number was drawn which determined the third location for T3.
The unused spelling location numbers were then placed with the others
and in a similar manner the locations of the remaining error types
were selected. Other than spelling errors, the order of error

type selection was random. Error locations are shown in Table 3.2.

Table 3.2. Location of errors and error-types by slide number.

 

 

 

Error type T1 T2 T3
Spelling 27 27, 50 . 27, 40, 50
Focus 37 37, 45 37, 45, 68
C0py 9 9, 47 9, 47, 62
Exposure 49 43, 49 14, 43, 49
TOtals 4 8 12

 

 

 

 

During initial data analysis a mistake was discovered which

resulted from human error. While loading slide trays, the focus error

47

in T1 (location 37) was missed and an extra copy error was inserted
(location 62). In T2, one of the two focus errors intended for
location 45 was missed and an extra copy error was inserted. The
number of errors remained correct, but in these two instances the
error types were incorrect. The actual error locations are shown in
Table 3.3.

Table 3.3. Corrected location of errors and error-types by
slide number.

 

 

 

Error type T1 T2 T3
Spelling 27 27, 50 27, 40, 50
Focus -- 37 37, 45, 68
Copy 9, 62 9, 47, 62 9, 47, 62
Exposure 49 43, 49 14, 43, 49
TOtal - 4 8 . 12

 

 

 

 

The photographic production of the revised version of "A Fable
Revisited" was done by the researcher from the original art work. All
exposures were made on Ektachrome X film with standard copy equipment
and lighting facilities. A gray card reading was used to determine
correct exposure. Error slides were handled in the same manner with
the exception that various degrees of error were photographed. Final
selection was based on a single criteria that the error was obvious
enough to be seen by most observers in a series of tryouts. From these

masters, four duplicate copies were prepared by a professional

_48
photographic technician who was also given instructions not to alter
or attempt to correct any of the error location slides. The duplicate
copies were subsequently evaluated for consistency and utilized in the

study as the Control, T1, T2, and T3 treatments.

Channel Relationship

 

Hartman [1961] suggests that in studies involving multiple
channels ". . . the relation of the information in each channel to
the information of other channels should be specified" [p. 242].

In an attempt to determine channel relationship and thereby provide
evidence concerning the source or sources of the message (audio,
visual, or audio visual primacy), the 71 slides representing the audio
channel were shown to a group of 21 undergraduate education students
at Michigan State University.

The first effort at content identification involved the students
recording their impression of what the slide "said" by writing a
single word or short description of each. The following directions
preceded the presentation of the slides. In no case was the narration
used.

"These slides illustrate a modern version of the old

three-pigs fable and their trouble with the wolf.

They are used to illustrate visually (with and/or

without an accompanying narration) a problem-solving

system which we wish to teach to others. As I show

you the slide, please write in the space provided a

simple observation about what is happening--one word

could be enough or it may take a short statement."

Following the initial presentation, the slides were shown a second

time under somewhat different conditions. Each person was provided with

49

a cOpy of the instructional development model which is illustrated by
"A Fable Revisited" (see Appendix F) containing only the stage and
function names. The following directions were given:

"As a slide is shown, write the number of that slide

under the concept--such as "Identify Problem" or

"Specify Methods"--which you feel is most closely

related to the picture. If you feel that a picture

is not related to any concept, write the number of that

slide at the bottom of the page. Pictures may or may

not relate to one of the nine concepts. You'll be given

each slide number twice, once at the beginning and once

before the next slide is shown.

Each of the pairs of tests were then compared for commonalities
and on the basis of a lengthy analysis and comparison to previously
determined audio channel content, the channel relationship at each
slide location was estimated. This information is recorded in the
script (see Appendix G). Utilizing Hartman's [1961] channel relationship
categories of redundant, related, unrelated, and contradictory, the
analysis indicates that all channel conditions except contradictory
are represented. For the most part, the channel relationship may be
characterized as being related, that is to say that the narration
tends to describe either the object or the situation shown in the

visual channel at any given point. On this basis "A Fable Revisited"

can be said to be "audio primary" in design.

Equipment and Facilities

 

Extreme care was taken to ensure high quality message delivery
for it was of the utmost importance that any errors reported, or the
possible effects of such errors, be attributable to the experimental

design--not the result of accidental projection errors, distortion

50
resulting from( keystoning, lack of light control, poor audio quality,
or viewing conditions.

‘ Four manual focus Kodak Ektagraphic Model B slide projectors
equipped with new lamps and F/3.5 4" to 6" zoom lens were used for
projection. The projectors were placed behind and slightly above the
subjects at a 90 degree angle to 6' X 6' matched matte white screens.
The audio channel was delivered by a 3-M Model 2550 tape recorder
located at the center of the room and synchronized with the projectors
for slide changes with a specially designed wire set. The sound was
amplified and transmitted through the internal public address system
of the facility.

The Erickson Kiva, a large-capacity, multiple-tiered, circular
classroom with sunken center section, was used for the experiment.
Light control was achieved with heavy lined window drapes and conditions
during the experiment may be described, by consensus of the subjects,
as somewhat too dark for comfortable note taking. Because the
investigator regularly teaches in this facility, it can be stated that
lighting conditions were "characteristic" at the time of the experiment

and consistent over the several groups tested.

Technical Assistance

 

Eight technical assistants were used during the experiment to
assure consistent projection and sound. One person operated the tape
recorder, four were assigned to projectors, two manned the distribution
table at the entrance of the room, and one person served as an overall

monitor.

51

9

Prior to the presentation, all technicians and assistants were
briefed on the nature of the study, on what was going to take place
during the experiment, and on how to react to anticipated emergencies
or problems. This included a written set of instructions including
procedures to follow in the event of such emergencies as a failure
of the tape recorder, blown projector lamps, failure of a slide to
drop, loss of synchronization, and the like. All appropriate personnel
were given scripts to follow and instructed to note any irregularities
observed during the presentations.

During the pretest rehearsal, the control projector failed to
respond to the tape signal and thus this unit was Operated manually
following the script which was provided for such an emergency. No
apparent problem resulted from this emergency. Insofar as the
investigator and the monitor could discern, there were no distractions
or deviation from.what was planned as a normal environment.

Upon arrival, subjects received a sheet containing instructions
and group assignment which had been previously determined using a
table of random numbers (see Appendix H). Assistants helped direct
subjects to clearly identified group locations and to randomly selected
seat locations within these areas. Inidividual seats were identified
by name and by an identification code on test packets.

Four nearly identical viewing areas were arranged around the
floor and first two tiers of the Kiva type classroom. Seating was
arranged so that the projection screen and 16 chairs were located
within a rectangular area-—approximately 10 X 16 feet.

While all groups were located in the same room, the circular

52

design of the facility and placement of groups and equipment precluded
any person from viewing any other image.

Immediately priOr to the presentation of "A Fable Revisited,"
subjects were shown a test slide which was similar in design, style,
and lettering to those used in the experiment. Subjects were instructed
to exchange seats if they were unable to easily read the lettering
on the test slide. No person was observed to move from his assigned
position.

The experiment was conducted during a regularly scheduled 50
minute class period. Prior to the presentation, subjects were given
a brief verbal introduction which included a caution against any
conversation for the remainder of the period.

A two-minute tape recorded message (see Appendix I) served as
the formal introduction to the activity which was explained to be
a validation procedure. The message was designed to provide a
motivational element and also to serve as an "advance organizer."
Specific activities which would take place during the course of the
experiment were described.

The above introduction was followed immediately by the presenta-
tion of the slide tape program. Verbal instructions were then given
for the cognitive test. The subjects were asked to Open their test
booklets to the first page, read the directions, complete the test,
and to close the booklet after completing the last question. An
opportunity was given for questions concerning the test or the machine
scored response form, but none were asked. Seventeen minutes were

devoted to the cognitive test. Upon completion of the 35 item

53

cognitive test, subjects were asked to open the test packets to the
last page which was the response form for the error perception test.
Because of the need to hide the purpose of this instrument, it was
necessary to select a system for exposing questions one at a time.
Therefore, both directions and questions were presented by tape.
This procedure took four and one half minutes.

The period was ended on a brief note of appreciation from the
investigator and assistants then held a short debriefing session for
the purpose Of determining if any irregularities were noted. Other
than the synchronization problem reported previously, no further

irregularities were revealed.

Summagy

A population of 59 elementary education students were the subjects
of this experiment to attempt to determine if cognitive learning is
affected by the presence and recognition of common production errors in
the visual track of a slide tape program. Errors included poor COpy
technique, incorrect exposure, poor focus, and misspelled words.
Subjects were randomly assigned to a Control and three treatment groups
and were presented a professionally rendered cartoon slide tape presen-
tation designed to teach a particular instructional development paradigm
and related system's terminology.

With the exception of the number of errors present (Control = 0,
Treatment 1 - 4, Treatment 2 - 8, Treatment 3 = 12) the message and
viewing conditions were virtually identical. Following the presentation,

subjects completed a previously validated 35-item multiple choice

54
instrument designed to test recall and a lO-item error perception
test which was designed to determine if errors were actually seen.
The experiment‘was completed as planned and the results are reported

in the following chapter.

CHAPTER IV

ANALYSIS OF DATA

Review of Experimental Procedure

The purpose of this study was to determine if photographic
production errors in the visual information track of a slide-tape pro-
gram had any effect on the recall of audio and/or visual channel
information.

A population of appropriate subjects was drawn from an under-
graduate Elementary Education class and randomly assigned to four
treatment groups.

The message, an introduction to the systems concept and a
specific instructional development paradigm, was presented via an
existing professionally rendered cartoon style slide-tape presentation.

The experimental design involved four treatment groups and
four degrees of visual interference ranging from a no error version in
the control to four, eight, and twelve errors respectively in T1, T2,
and T3. The four versions were shown simultaneously and used a
common audio source. Immediately following the presentations, subjects
completed two evaluation instruments. The first attempted to measure
recall of information contained in the presentation. The second,

a tape recorded questionnaire, was designed to determine if errors

were perceived and, if so, to what extent.

55

56

Hypotheses

Three hypotheses were tested. In null form, they are:

1. There will be no difference between groups in their
perception Of production errors.

2. There will be no difference between groups in cognitive
learning as measured by immediate recall.

3. In the event of significant effect, there will be no
relationship between groups in the number of errors perceived

and scores on the cognitive measure.

Perception of Errors Hypothesis and Data Analysis

 

Null Hypothesis: There will be no difference between groups in
their perception of production errors.

The design of the error perception instrument (see Appendix C)
suggested two analytical procedures. Questions 1, 2, and 3 elicited
yes-no answers and thus were simply tabularized.

One way analysis Of variance (ANOVA) was used to test for effects
between groups on questions 4, 5, 6, and 10. The F statistic was used
to determine significance.

For convenience in interpreting the data, the questions from
the instrument are restated below and the results displayed in sub- .
sequent tables.

1. Did you Observe any deviation from the generally high

quality of the visual material? Circle yes or no.

2. Did you Observe any instances of what might be called

57

production mistakes or errors in photography? Circle yes or
no.‘

3. Production mistakes include such matters as misspelled
words, poorly focused or "fuzzy-unclear" slides, imprOperly
exposed slides which are excessively light or dark, and
poor copy techniques which allow things to show in the slide
which shouldn't be there. Did you notice any instances of
these kinds of mistakes? Circle yes or no.

Responses to these questions are shown in Table 4.1.

Table 4.1. Responses to error perception Questions 1, 2, and 3.

 

 

 

 

Question 1 Question 2 Question 3
Treatment YES NO YES NO YES NO
Control 2 ll 0 l3 2 11
T1 1 13 6 8 8 6
T2 3 l4 7 10 17 0
T3 13 2 l4 1 15 0
Totals 19 40 27 32 42 17

 

 

 

 

Statistical analysis was not performed on these data. However,
inspection indicates a pronounced tendency for treatment groups to
differ and clear evidence that all three experimental groups experienced
and perceived errors at a level obviously greater than the control
group. This is especially evident in T2 and T3.

Computer analysis was performed on the data resulting from

58

questions 4, 5, 6, and 10. In each case, and at the .05 level of

confidence with 3 and 55 degrees of freedom, an F value of 2.76 would

allow rejection of the hypothesis of no difference. The questions are:
4. If your answer to question three was yes, how many mistakes

do you think you observed? Write the estimated number.

Table 4.2. Estimate of errors Observed--Question 4.

 

 

 

Frequency Control T1 T2 T3

0 ll 7 O 0

l O 4 6 O

2 2 2 6 l

3 l 3 3

4 O 2

5 2 4

6 2

7 l

8 0

9 0

‘10 2

TOtal observed 4 ll 37 78

Total errors present 0 56 136 188
Mean no. observed , .308 .785 2.176 5.200

 

 

 

 

 

Analysis of the data is displayed in Table 4.3.

59

Table 4.3. Analysis of variance table for estimate of errors
observed-~Question 4.

 

 

 

Sum Of Degrees of Mean
Source of variance squares freedom square F value
Between groups 209.562 3 69.854 30.984 8
Within groups 123.997 55 2.254
Total 333.559 58

 

 

 

 

 

There is an obvious statistical difference between the groups in
their ability to perceive errors. Therefore, the null hypothesis of
no difference is rejected.
5. With the exception of the word "tunnel" which was intentionally
misspelled, did you notice any misspelled words? If yes,

write the estimated number.

Table 4.4. Estimate of spelling errors observed--Question 5.

 

 

 

Frequency Control T1 T2 T3

0 13 9 ’ 8 4

l 3 7 1

2 l 2 7

3 1

4 l l

5 1

TOtal observed 0 9 ll 27
Total errors present 0 14 34 45
Mean no. observed .000 .643 ‘.647 1.800

 

 

 

 

 

60

Analysis of the above data is presented in Table 4.5.

Table 4.5. Analysis of variance table for estimate of spelling errors
observed--Question 5.

 

 

 

. Sum of Degrees of Mean

Source of variance squares freedom square F value
Between groups 24.063 3 8.021 7.949 S
Within groups 55.497 55 1.009

Total 79.560 58

 

 

 

 

 

There is statistical difference and thus the null hypothesis is
rejected.
6. Four different versions of "A Fable Revisited" were shown
today. The one you saw contained either pg_mistakes,
four mistakes, eight mistakes or twelve mistakes. Which
did you see--none, four, eight or twelve? Write the number.

Table 4.6. Determination of 0, 4, 8, or 12 error version--
Question 6.

 

 

 

Version Control T1 T2 T3
0 ll 5 0 O
4 2 9 l7 5
8 10
12
No. actual errors 0 4 8 12
Mean of estimate .615 2.574 4.000 6.667

 

 

 

 

 

61

Analysis of the data is shown in Table 4.7.

Table 4.7. Analysis of variance table for determining error version--
Question 6. -

 

 

 

Sum of Degrees of ‘ Mean
Source of variance squares freedom square F value
Between groups 274.398 3 91.466 38.157 S
Within groups 131.839 55 2.397
Total - 406.237 58

 

 

 

 

 

A significant difference is clear and the null hypothesis is

rejected.

10. Upon reflection, how many of each of the possible mistakes
do you think you saw? Write the estimated number Opposite
the letter standing for copy, exposure, focus, and spelling.

Totals of all error types Observed are summed in Table 4.8 to

provide a further measure of the ability of subjects to detect errors.

62

 

 

 

 

 

 

Table 4.8. RetrOSpective estimate of errors observed--Question 10.
Frequency Control T1 T2” T3

0 ll 5 2 0

l O 4 3 O

2 O l 4 l

3 O l 5 l

4 2 3 2 3

5 l l

6 2

7 3

8 2

9 l

10 0

ll 1

Total observed 8 21 39 91
Total errors present 0 56 136 188
Mean no. observed .615 1.500 2.294 6.061

 

 

Analysis for these data is shown

in Table 4.9.

63

Table 4.9. Analysis of variance table for retrOSpective estimate Of
errors observed--Question 10.

 

 

 

Sum of Degrees of Mean
Source of variance squares freedom square F value
Between groups 250.312 3 83.437 25.990 S
Within groups 176.569 55 3.210
Total 426.881 58

 

 

 

 

 

The null hypothesis was rejected.

Additional Findings

 

Data which may have relevance for future studies were gained from
responses to error perception question 7 and 8.
7. Assuming you saw a version with mistakes, did these bother you
in any way? Yes or no.
8. If your answer to the last question was yes, did any one
type bother you more than the others? If yes, indicate
which--copy, exposure, focus, or spelling by circling
c, e, f, or s.
The combined results of these question are reported in Table
4.10. Responses tabulated under the "none" column are equivalent to
a "no" answer to question 7. The remainder are "yes" responses.
Among those subjects responding to question 7 in the affirmative,
by far the greatest number indicate focus errors as most bothersome.

The notable exception is in T1 where, as previously indicated

64

Table 4.10. Type of error causing most concern--Questions 7 and 8.

 

 

 

Treatment None C0py Exposure Focus Spelling
Control 11 0 0 2 0
T1 11 3 0 O . 0
T2 1 0 2 13 1
T3 0 0 l 14 0
TOtal 23 3 3 29 l

 

 

 

 

 

 

(see page 47), the single focus error intended for that version was
inadvertently omitted.

On the basis of the evidence, it seems quite clear that subjects
were able to perceive production errors. Moreover, inspection of the
tabularized data clearly indicates that the number of errors perceived
increases as the incidence Of increased numbers of errors present in a

version becomes greater.

Cognitive LearninggHypothesis and Data Analysis

Null Hypothesis: There will be no difference between groups in
cognitive learning as measured by immediate recall.

The statistical design for testing this hypothesis anticipated
several analyses in the event of significance. These included one-way
analysis of variance on the cognitive test, on classes of question
within the total test and on types of errors. Finally, the Tukey

(if subjects were equally distributed between groups) or Scheffé

65

(if unequally distributed) test for multiple comparison of means would
be applied post hoc.

Computer analysis was performed on the data. At the .05
level of confidence with 3 and 55 degrees of freedom, an F value of
2.76 would allow rejection of the hypothesis Of no difference

(see Table 4.11).

Table 4.11. Analysis of variance table for effect on recall.

 

 

 

_ Sum of Degrees of Mean
Source of variance squares freedom square F value
Between groups 63.231 3 21.077 1.688 NS
Within groups 686.905 55 12.489
Total 750.136 58

 

 

 

 

 

On the basis of this analysis there is no statistical difference
between the groups, and the null hypothesis cannot be rejected. Due to

this finding no further analysis was performed on the data.

Error-Performance Relationship Hypothesis and Data Analysis

Null Hypothesis: In the event of significant effect, there will
be no relationship between groups in the number of errors perceived and
scores on the cognitive measure.

Statistical analysis of this hypothesis was dependent on a
finding of significance in the preceding hypothesis. Since there was

no significant difference in cognitive learning, statistical

66

investigation is inapprOpriate.

Summagy

The error perception hypothesis was first tested with three yes-
no questions. A strong tendency toward differences between groups was
noted and four subsequent error-estimate questions all provided
statistically significant (p==.05) evidence that errors were perceived
and the null hypothesis was rejected.

Although all three experimental groups exhibited lower mean
achievement scores on the cognitive instrument as compared with the
control group, such differences were not significant (p==.05). The
null hypothesis of no difference due to production errors was therefore
accepted.

Because of the finding of no significant difference for the
learning hypothesis, the error-performance relationship hypothesis
was not analyzed statistically. However, a trend was noted between
C, T1, and T2. While T3 consistently reported higher mean scores
on error Observations, the mean achievement score failed to follow
the downward trend established between control, T1, and T2. That is,
C’T1> T2<T3.

These findings will be reviewed and further discussed in Chapter
V; implications of the results of the study will be analyzed and

certain suggestions Offered for future research.

CHAPTER V

SUMMARY, CONCLUSION AND DISCUSSION

Summary

The literature dealing with inter-modality variables such as
color, legibility, technical quality, channel effectiveness, and infor-
mation level, clearly indicates that far too few prescriptive guide—
lines have been developed empirically. Likewise, there is much
evidence that such information is sorely needed by producers and
users alike.

One such factor which has not received adequate attention is that
of the effect of low quality visual materials on cognitive learning.
The most frequently applied guidelines, possibly resulting from a
combination of early communication noise research and practical
experience, mandates virtual technical perfection in the production
of visual materials. This topic is not to be confused with earlier
embellished-unembellished materials studies for unembellished
materials are not necessarily of poor technical quality.

An extensive exploration of the topic yielded practically no
research information which would help resolve the issue. Thus an
experiment was undertaken whereby a population of undergraduate
education students, unfamiliar with the subject matter of instructional

development, were exposed to four versions of a professionally

67

68

prepared cartoon style slide tape pmogram in which the visual
channel varied only in the number of technical imperfections--production.
related errors including poor copy techniques, under or over exposure,
poorly focused slides, and misspelled words. The number of errors
ranged from none in the Control group to 4, 8, and 12 in the three
treatment groups. The presentations were made simultaneously under
virtually identical physical conditions in a large, circular, Kiva-
type classroom and utilized an identical audio channel delivered from
a common audio source. Following the presentations, two evaluation
devices were admininstered--a 35 item multiple choice test designed
to measure cognitive learning and a 10 item error perception
instrument constructed to determine if subjects actually perceived
the induced errors.

Three related hypotheses were tested.

1. There will be no difference between groups in their
perception of production errors.

2. There will be no difference between groups in cognitive
learning as measured by immediate recall.

3. In the event Of significant effect in the second
hypothesis, there will be no relationship between
groups in the number of errors perceived and scores on
the cognitive measure.
Subsequent analysis of the data produced the following results.
In all cases, simple one way analysis of variance (ANOVA) was the
statistic Of choice and significance was calculated at the .05
level of confidence using the F statistic.

1. Hypothesis 1 - A significant effect was found, pp..01.

2. Hypothesis 2 - A finding of no significant difference
resulted, p < .10.

69

3. Hypothesis 3 - Not analyzed statistical because of
NSD in the second hypothesis. However, a trend toward
positive correlation was noted.

Conclusions

 

On the basis of the data reported above, several observations
are offered.

On the perception hypothesis a pronounced error recognition
effect was noticed with a significance level exceeding the .01 level
of confidence on all questions designed to be evaluated. On the basis
of this finding, several conclusions may be drawn.

1. Subjects did observe errors in the treatment versions.

2. Such observations were in direct proportion to the number
of errors present in a given version. That is to say, the
group exposed to the 12 error version reported seeing more
errors than the group experiencing 8 errors, and they saw
more than the group subjected to the 4 error version.

3. Treatment groups consistently underestimated the number
of errors present in a given version. However, estimates
were more accurate as the number of actual errors present
in a version increased.

4. Of the four error types used in the study, subjects
overwhelmingly reported focus errors as most bothersome.

It therefore seems apparent that in the present study errors
were seen by those groups who were exposed to errors in direct
proportion to the number of actual errors present. It is also

suggested that not all error types are perceived in the same

7O

proportion nor do they appear to be equal in influence.

NO significant difference was found for the major hypothesis of
effect on cognitive learning (p‘<.10). Though not statistically sig-
nificant, the following Observations are of interest (see Table 5.1).

l. The control group consistently displayed a higher mean score

on the recall instrument than any of the treatment group.

2. Mean recall scores demonstrated a tendency toward pr0portional
lowering between the Control, T1, and T2 groups--a direct re-
lationship between lower scores and increased incidence of
errors present and/or observed. This tendency was not main-
tained in T3, the group receiving the largest number of errors,

for the mean T3 groups recall score was higher than T2.

Table 5.1. Achievement scores--cognitive test.

 

 

 

Control T1 T2 T3
14 7 9 15
17 16 ll 14
18 17 12 17
15 13 ll ll
17 10 12 12
15 L 10 17 10
12 l8 17 20
l6 17 22 18
l4 l7 16 13
19 13 8 15
l4 16 ll 12
14 9 10 23
l4 l3 14 12
13 ll 7
12 14
5
14
Total 199 189 212 213
Mean 15.308 13.500 12.471 14.200

 

 

 

 

71

No absolute conclusions can be drawn, but there is a small,
statistically insignificant trend toward a possible interference
effect as well as a similar relationship between scores and errors.
The case of T3 could be due to chance or it may substantiate like find-
ings in a few previous studies. In this event, several theories are
of interest and will be considered later in this chapter.

As indicated, the third hypothesis was not analyzed and therefore
no conclusions are offered.

Given the conditions and restraints imposed on the present
study--population, subject matter, design characteristics, vehicle
(slide tape), style (color cartoon), channel design and relationship,
type of errors employed and type of learning involved--two conclusions
may be offered.

First, there is nearly unmistakable evidence that subjects did
perceive production errors when, in fact, errors were present. More-
over, perception was based on and directly related to the number of
errors present in a version.

Lastly,there is no statistically significant evidence that
production errors interfere with the cognitive learning process. On
the basis of this finding, it is probably unjustifiable to
automatically concentrate human and financial resources on attaining

technical perfection given the conditions outlined above.

Discussion

 

Whether or not audiences are generally aware of technical

imperfections in media presentations has long been a matter of

72

conjecture among media professionals. Direct personal observation of
audiences noticing-~or apprently not noticing--misspelled words or
poOrly focused slides, perhaps has influenced how a given media
producer looks at the question. Some persons have maintained that
audience characteristics undoubtedly influence error perception.

For example, media specialists or photographers see technical errors,
and English teachers are particularly sensitive to spelling errors.

Studies which have specifically tested these perception/non-
perception positions are rare and, in fact, most information is by
implication. Schlater [1966], in a previously reported study, felt
that errors were not perceived. While he found a significant lowering
of learning in material presented in the audio channel in the presence
of production errors, this effect may well have been due to a second
variable--speed of presentation. The explanation for lack of effect
otherwise was that the errors probably were not obvious enough.

Sencer [1965] and Razinsky [1967] did test perception of errors
in printed material and determined that errors were definitely Observed.
Hsia [1969], in a study using 25% black dots as visual channel
interference, reported a significant effect, that is, lowering of
learning. Interference was said to be the probable cause. However,
perception of the interference source was apparently not singled out
for testing. Thus perception of the interference source is by
implication. The present study has shown that errors were seen. A
possible explanation rests in the fact that the four error types were
made visable--as Obvious as possible without exceeding the limits of

reality and common sense.

73

There existed an early suspicion that errors might be tabulated--
counted--as the program was presented. If so, logic suggested that
the very process of remembering the number of errors might cause
interference. Estimates Of errors, before and after the subjects had
been given an indication of the number Of errors they might have seen,
were consistently less than the actual number present. Response to a
direct question of whether subjects found themselves overtly counting
errors brought an overwhelming "no" reSponse. Thus the tabulation
theory seems, on the surface at least, nonexistent. However, not all
error types were equally obvious.

Copy and exposure errors were reported by 37% and 32% respectively
of those subjects exposed to errors while spelling errors were reported
by 60% of the subjects. By far the most frequently reported error type,
and that which was judged most bothersome (see Table 4.10), was focus.
Reported by almost 100% of the subjects in treatment groups having
focus errors, this overwhelming response is interesting and deserving
of further investigation simply because this probably represents the
most frequently occurring error in day-to-day presentations. The
present study did not answer the question Of ghy_focus errors were
most bothersome, but it may be speculated that subjects felt that
something important was being withheld. Focus was the only error type
which actually distorted an image beyond recognition. The distortion
hypothesis would be consistent with Levie [1973], Arnheim [1962], and
Harrison [1962], all of whom maintain that there is a point beyond
which compression (or withdrawal) Of relevant cues--the minimum

information level at which desired visual perception can occur--becomes

74

dysfunctional and impedes or makes visual learning impossible. Even
though there was no apparent effect on learning, the feeling among
subjects may still persist.

While there are no data upon which to speculate about the
relatively low incidence of reported copy and exposure errors, one
possible explanation lies in the fact that these error types were
not perceived by most subjects as errors at all. Rather, such
imperfections are representative of everyday occurrences among amateur
photographers. Over and under exposed slides and snapshots are common
and extraneous materials, hands, heads, and other unintended items are
all too frequently seen and accepted in the work of amateur
photographers.

Thus there is the distinct probability that error types are not
equally dysfunctional or observable by general audiences and that
audience characteristics may well exert an influence on how different
types Of errors are perceived.

There are at least two interference theories or positions which
are relevant to the outcomes of the present study. The literature
on communication noise as characterized by Shannon and Weaver [1949],
Berlo [1960], and others, postulates the negative effects of noise
on communication while a comparatively small conditional representation
summarized by Levie [1973] suggests that noise (as represented by
technical quality) is not dysfunctional per se--only when relevant
(necessary) visual cues are obliterated or otherwise masked. The
assumption of noise/technical quality equivalency has apparently been

assumed. Yet, there is evidence that under certain previously assumed

75

noise conditions interference does not occur. Thus an intriguing
possibility exists that the concept ”noise” is not a general concept
but rather specific to the situation. Moreover there is a suggestion
that interference effects are again conditioned by audience character-
istics--the intelligence relationship suggested by Hsia [1969] or the
limited information processing capacity of humans reported by numerous
writers, including Broadbent [1958], Travers [1964], Severin [1967],
Hsia [1968], and Travers and Alvarado [1970].

Severin [1967] tested the effects of irrelevant cues and found
greater recognition learning coming from the treatment with the more
irrelevant cues and speculated that "It may be that subjects ignored the
obviously irrelevant cues, . . ." [p. 399]. The term "blocking" was
also used to explain the finding. That is, the subjects blocked the
channel with the irrelevancies and relied on the other for information.

The finding of no significant difference in the present study may
indeed be explained in part by Severin's [1967] comments. A second
and closely related possibility may be attributed to information
processing capability. Travers and.Alvarado [1970] write that

Living organisms are systems with a limited capacity

for handling information and have to be selective

in what they attend to. If they were not, they

would constantly find themselves swamped with large

volumes of useless information [p. 57].

Closely related to this explanation is that of Hsia [1968].

It has been suggested that man is not motivated toward

hearing or seeing all stimuli surrounding him; he either

subconsciously refuses to process all the information--his

defensive mechanism in IP (information processing) performs
homeostatic functions to fend off excessive information--or

he selects only the information which has "surprise value,"

or "predictive value" [Cherry, 1961], or is essential for
his survival [pp. 255-56]. '

76

Hsia [1968] offers a further explanation in the possibility that
when informatiOn overloads the information processing system the
amount of information serves to overextend the processing capacity
of the central nervous system.

The finding of no significant difference on the major hypothesis
is, in a real sense, significant if nOt generalized too far beyond the
conditions of the present experiment. As indicated previously, there
is at least a hint that technical perfection is unnecessary and
possibly a waste of resources.. However, the consistency of lower
scores in all treatment groups as compared to the control group is
of interest and quite logically motivates a nagging feeling that
some effect is present, that something perhaps has been missed.

The finding that the group subjected to and observing the most
errors (T3) failed to follow a pattern of lower scores as the number
of errors increased is interesting but not too surprising for the data
tend to confirm, to some extent, previous findings of Schlater [1966]
and Severin [1967].

The question of channel relationship must necessarily be raised.
As reported earlier, the message vehicle used in the present study
was purposely selected because it was representative of contemporary
channel design--audio primacy. If a vehicle had been selected which
displayed visual primacy, would the interference effects have been
greater? The answer can only be speculative, but there is evidence
[Schlater, 1966], Severin [1967], and Hsia [1969] that the visual
channel is more resistant to interference than the audio channel and

thus similar findings would be a distinct possibility.

77

There remains the possibility that the results of this present
study may be attributed to chance or population characteristics.
It is, however, interesting to note that the results of pre and post
experiment testing of the media and evaluation instruments were
extremely similar even though audiences were older on the average
and established in elementary and secondary education curricula
rather than just beginning. Thus it is felt that the population
was representative.

A further consideration involves the instrumentation used in
the study. While the reliability of the error perception was assumed,
all tryouts, before and after the experiment, yielded highly consistent
results. Reliability of the cognitive instrument was determined to
be nearly .70 on the Kuder Richardson scale and again pre and post
experimental uses showed consistency. However, an interesting--and
thus far unexplained--event occurred during the actual experiment.
When the Kuder Richardson reliability coefficient was calculated for the
control group following the experiment, it was found to be -.69, a
highly unusual measurement. The data were re-analyzed by computer and
eventually by hand, but the coefficient remained the same. Numerous
consultations and work by testing specialists failed to determine the
cause and the only plausible explanation offered is related to the
unusually narrow range of scores for the Control group (see Table 4.12)
as compared to the treatment groups. This chance distribution may be
responsible for the negative reliability of the instrument.

A final comment involves the level of confidence (p==.05) used to
measure significance and suggests caution--along with the reliability

aspect--in accepting the finding of no significant difference

78

on the major hypothesis. If a confidence level of p = .25 is used,
there is significance. Although the incidence of chance error is
obviously greater, this combined with other factors suggests future

research direction.

Implications for Future Research

 

There are numerous research directions either suggested by
this or related studies. Most obvious, perhaps, is a replication of the
present study in which several modifications should be considered.

1. The cognitive instrument should be further refined to
achieve higher reliability. It is likely that the
negative reliability coefficient would disappear in such
a replication.

2. A large population (n>'25 subjects per group) should be
used in any replication and an attempt made to isolate a
second variable related to audience characteristics--
intelligence for example.

3. A fourth and possibly fifth treatment group (T4) should
be added and be exposed to 16 and 20 errors, respectively,
to maintain the established linear progression. This
would serve to determine whether the upward swing in
mean score demonstrated by T3 would continue upward,
stabilize, or regress. In effect, this condition would
be a check against current information processing theory
discussed previously.

4. Consideration should be given to a different subject

79

matter and the possible use of a pre-test.

Other follow-up studies might include design variables such as

the following:

1.

Using measured error perception data and recall scores
as covariables, data might be obtained to demonstrate
whether the presence of errors per se has any effect or
whether such errors must be overtly perceived before
possible interference results.

A second multi-variable design could use attitude, per-
ception, and recall data. The attitude element becomes
important whether an effect is noted or not for, as in
the case of color vs. black and white film studies, there
is growing speculation that attitude toward the message,
source, or indeed, the medium itself, may be highly
important. A logical application of this concern

might be in teacher education where poor attitude

toward a medium could conceivably influence use at

a later date.

In view of the reported "bothersome" nature of focus
errors, a study which used only that type of error

seems logical for, as mentioned previously, focus

errors are perhaps the most common error in day-to-day
applications. This is more directly related to presenta-
tion errors-~those which occur during projection-~since
the combination of different types of film emulsions and

slide mounts naturally produce out-of-focus images and

80

very few schools currently own or use so called "auto-
focus" slide projectors.

4. The preceding study suggests an entire experiment de-
voted to presentation errors--focus problems, blown lamps,
the failure Of a slide to drop, and the like. Such errors
are clearly more obvious and thus may produce interference--
if only from delays or disruption of message continuity. If
negative effects are found, a logical application of the data
would be to develop corrective procedures such that inform-
tion from the message would not be lost. The implication,
therefore, is that the experiment would be designed such that
presentation errors would be corrected in one version,
uncorrected in another.

5. No attempt was made in the present study to specifically
measure the influence of a given error or error type on
information processing at the point of the error or
on information which immediately preceded or fOllowed that
error.

6. A further area of interest involves audience expectations
and technical quality. What are the effects, for example,
when an audience expects high quality but experiences poor
quality or expects low quality and experiences low quality.

The preceding suggestions are by no means exhaustive for so

little is known and so much is needed. For example, much is known about
comparative learning--both simultaneous and sequential--under ideal

conditions. What happens under conditions of noise or interference

81

when, for example, a sequential comparison strategy is used? The
~ answer, it is speculated, is very little, and yet comparative learning
strategies are frequently used in media design--Often times without
realization and recognition. It is entirely possible that the value
of the strategy is negated under such possible interference conditions.
It is hoped that the present study has made some small contribution
to the instructional media field in general and to message design and
delivery in particular. Certainly one of the most important efforts
that present and future scholars can make is in the direction of
developing a prescriptive message design and delivery technology, a
technology which, given Objectives, audience or individual characteris-
tics, message content, and other necessary inputs can deliver a

product which will be both effective and efficient educationally.

APPENDICES

APPENDIX A

LETTER '10 PROFESSOR BREHM

NOvember l, 1973

Dr. Shirley Brehm

Elementary Special Education
351 Erickson Hall

Campus

Dear Shirley:

I truly appreciate your willingness to assist with my research
study. I have delayed the inevitable much too long. Now, however,
my degrees of freedom are somewhat restricted and so away we go!

The nature of my study seems somewhat simplistic on the surface,
but as one delves into the problem it becomes quite apparent that it
is not simple at all. The study is tied into a whole range of related
kind of things including information processing, channel capacity and
communication noise to mention but a few.

The title of the study is "The Effect of Production Errors in the
Visual Track of a Multi-Sensory Program Format on Cognitive Learning,"
and is motivated by the fact that many of us in the "media business"
have traditionally tended to follow a position which is often
associated with Eastman Kodak. The essence of their advice is that
visual errors of any kind, low quality visuals of any kind, are to
be avoided like the plague for they associate this with reduction in
cognitive learning and a negative attitude shift. For years, many
people have followed this position literally without question. In
my own experience I have Often seen multi-sensory materials--slide
tapes, motion pictures, television and the like--used in classrooms
with apparent disregard for this advice. I often wondered what might
be happening to viewers. As time went on evidence began to make me
suspect that some of these production errors--things like bad
exposure, poor focus, misspelled words, sloppy copy techniques
and the like--might not be as bad as we were lead to believe. By
contrast it would be dysfunctional. My own personal practice over
the years has been to make my multi-sensory programs as perfect as
possible. I know what it means in terms of time and money to perfect
programs and.thus this also became a concern.

Eventually I decided to look into research and to my amazement
discovered that there was little or no research to support this
particular position. Upon talking to people at Eastman Kodak I
discovered that the basis for their contention was probably a result

82

83
Dr. Shirley Brehm November 1, 1973

Page 2

of experience rather than any kind of empirical evidence. I
then talked to a number of visual communication experts around the
country and they were supportive of this suspicion.

Procedurally the study will go somewhat as fellows. I selected
a professionally produced cartoon-style slide tape presentation which
was originally designed to teach groups of educators the essense of
an instructional development systems model--a very simple model
consisting of three stages and nine functions. I went through fairly
rigorous procedures in analyzing the information content of both the
audio and visual track--something which is only infrequently done in
research studies of the type I am concerned with. Following this
I rephotographed all of the art work and at previously and randomly
selected points induced production type errors into the otherwise
perfect visual channel. Called "A Fable Revisited," the three pigs
are the vehicle for presenting the information and the story is very
similar to the original with the exception Of the systems terminology.

There will be a control and three treatment groups. The control
group will see a perfect presentation, Tl will see a version with four
errors, T2 will have eight errors and T3 will have twelve. Following
the nine minute presentation, all populations will take a paper and
pencil test of information presented in any of the channel options.
Following this, they will complete a short questionnaire designed to
determine if errors are actually perceived.

I am hypothesizing that the errors will cause a lowering in
cognitive learning and that this will be directly related to the
number of errors involved. However, the fun thing about this piece of
research is that I don't really care how it turns out for either way
we will get information which is immediately applicable within the
restrictions imposed by the study. Part of the application value
resides in its association with the whole visual literacy movement
for there is sure a lot of material of questionable quality being
produced in our schools. we also have several spin-Off studies in
mind depending on which way the thing goes.

Again Shirley, I wish to extend my sincere thanks and I will do
everything possible to make this valuable to your students--including
any follow up information which some of them might like. I would
like to get a class list as soon as possible so that I might make
assignments to groups.

Sincerely,

@113" .

Bruce L. Miles

Please answer the following questions to the beSt of your ability.
Work quickly and complete each question in order. Do not return to a
question or change your answer once you have left the question. First
impressions and answers will be most representative and accurate.
In each case there is only one best answer. Do not guess unless you .
are reasonably sure your answer could be correct. Use response (5)
when you really don't know the answer. Please mark the number which
corresponds to your choice on the response form.

1. Upon entering the first stage in the systems process, the team
attempted to:

(1) identify performance objectives

(2) specify methods to solve their problems
(3) organize themselves

(4) determine their problem

(5) I really don't know

2. According to A FABLE REVISITED, the Systems Approach may best be
described as a process for:

(l) solving problems

(2) measuring solutions

(3) specifying problems through team work
(4) data-based problem specification

(5) I really don't know

3. The evidence indicates that Straw Pig selected his source of
building material:

(1) on the basis of need

(2) completely by chance

(3) with little thought

(4) after very brief discussions with other pigs
(5) I really don't know

4. In one scene System Pig made task assignments. There is evidence
to support the following statement:

(1) he had written assignments for all team members

(2) there is no evidence to suggest that assignments were in writing

(3) assignments may have been.written for all team members, but only
one pig actually had a sheet

(4) he had written assignments for Straw and Twig, but not for
himself

(5) I really don't know

5. Twig Pig showed some evidence of having:

(1) thoroughly researched wolves
(2) organized management

(3) analyzed the setting

(4) defined his problem correctly
(5) I really don't know

84

APPENDIX B

RECALL INSTRUMENT

10.

11.

85

At one point the wolf was shown carrying a bag. On the bag was
written:

(1)
(2)
(3)
(4)
(5)

bag "o" tricks

nothing was written on the bag
pig "skinning" tricks

bag of tricks

I really don't know

The second step in the final stage was to:

(l)
(2)
(3)
(4)
(5)

construct the actual house

recycle on the basis of data from the test
test the actual house

analyze results of model test

I really don't know

"What must be done," "what should be” and "what is" are most.
closely associated with:

(1)
(2)
(3)
(4)
(5)

identifying performance objectives
a needs assessment

organizing management

testing the real house

I really don't know

Straw Pig selected his building material:

(1)
(2)
(3)
(4)
(5)

after very brief discussions with other pigs
on the basis of need

completely by chance

almost by chance

I really don't know

The third stage is:

(l)
(2)

(3)
(4)

(5)
From

(1)
(2)
(3)
(4)
(5)

the evaluation stage and that is where the simulated wolf
attack came in

the evaluation stage and it was here that the proposed
solutiOn was built

not the last stage in the system model

the development stage and this comes before the evaluation
stage

I really don't know

the evidence it is safe to say that:

pig resources were brought in to build the model

human resources were brought in to build the model

only human resources were consulted

pig resources were consulted but only in the planning stage
I really don't know

12.

13.

14.

15.

16.

17.

86
According to A FABLE REVISITED, the Development Stage:

(1) is the second stage and the first activity in that stage is
to identify the problem

(2) is the second stage and the first activity in that stage is
to identify performance Objectives

(3) is not the second stage

(4) is the first stage and the first activity in that stage is
to identify the problem

(5) I really don't know

At one point we saw the wolf holding a sign. The sign said:

(1) do not disturb, watching pigs
(2) pig watching, do not disturb
(3) do not disturb, pig studying
(4) do not disturb, pig watching
(5) I really don't know

In the process of identifying their problem, the pigs collected
information on wolves. At this point they were:

(1) analyzing the setting

(2) in the development stage

(3) identifying performance objectives
(4) specifying methods and means

(5) I really don't know

When the Three Pigs collected information on the wolf they:

(1) had completed the first activity in the first stage
(2) were in the second stage

(3) were doing a needs assessment in the first stage
(4) had started the third activity in the second stage
(5) I really don't know

At one point the team built a model of their prOposed solution.
It was also referred to as:

(l)
(2)
(3)
(4)
(5)

prototype or simulation
simulation

prototype
protosolution

really don't know

Human»

In one scene Straw Pig carried an umbrella. The color of the
umbrella was:

(1) red
(2) black
(3) orange
(4) blue

(5) I really don't know

18.

19.

20.

21.

22.

23.

87

The performance objective as identified by the pigs is best
stated as follows:

(1) design and construct a defense system to ward off all wolf
attacks as soon as possible

(2) plan and test before the wolf comes a defense system for
preventing wolves from successfully huffing and puffing

(3) plan, test and build a defense system against all possible
wolf attacks before he comes '

(4) plan, test and build a defense system to ward off all
possible wolfian attacks

(5) I really don't know

At one point the wolf was shown carrying a bag. From the evidence
shown, the bag contained:

(1) some explosives, a saw and a mallet
(2) some explosives, a club and a saw
(3) some explosives and a saw

(4) some explosives and a mallet

(5) I really don't know

The straw house was built:

(1) with little or no thought

(w) with no thought at all

(3) on the basis of little thought

(4) thoughtfully, but of the wrong material
(5) I really don't know

Straw Pig's approach to the world of troubles may be characterized
as:

(1) no approach at all

(2) an incomplete systems approach
(3) a very weak systems approach
(4) almost no approach

(5) I really don't know

In the scene where the pigs were building the house:

(1) one pig had a hammer, a second a trowel

(2) one pig had a trowel and no other tools were shown
(3) one pig had a hammer and no other tools were shown
(4) all three pigs had tools

(5) I really don't know

In one scene System Pig held a list of simulated wolf attack
results. Listed:

(1) were two results

(2) was one result

(3) was no results

(4) were more than two results
(5) I really don't know

24.

25.

26.

27.

28.

88
As far as problems are concerned, Straw Pig's concern was:

(1) he wasn't concerned

(2) protection from the weather and the wolf
(3) protection from the wolf

(4) protection from the weather

(5) I really don't know

After constructing the model, the first thing the system team
did in the final stage was to:

(1) test the model

(2) evaluate the results of the model test carried out in the
second stage

(3) check the results of the model test carried out in the second
stage against the objectives

(4) construct the actual house

(5) I really don't know

According to System Pig, which of the following best represents
the concept of needs assessment?

 

 

(1) What should be (4) What should be
-What is -What is
What will be What must be done
(2) What must be done (5) I really don't know
-What is

 

What should be

(3) What is
-What should be
What must be done

 

The pig team determined the availability of resources--human, pig
and material. This was a part of:

(l) organizing their management
(2) analyzing the setting

(3) the develOpment stage

(4) specifying methods

(5) I really don't know

According to A FABLE REVISITED, assigning tasks is a part of:

(l) the first stage and third activity
(2) the first stage and second activity
(3) the second stage and third activity
(4) the second stage and second activity
(5) I really don't know

29.

30.

31.

32.

33.

89

According to the story the wolf:

(1)
(2)
(3)
(4)

(5)

spent a good deal of time studying pigs, but little time
observing them

spent a good deal of time studying and watching pigs
spent a good deal of time watching pigs

Spent a good deal of time watching pigs, but little time
studying them

I really don't know

The narration indicated that "there is more than one way to skin
a pig and . . .

(1)
(2)
(3)
(4)
(5)

the wolf has alternate methods

the adversary has alternate strategies
the wolf has optional strategies

the adversary has optional plans

I really don't know

The final house which was built:

(1)
(2)
(3)
(4)
(5)

had a door bell

didn't have a door bell

may have had a door bell

may have had a door bell, but there is no evidence to tell
I really don't know

According to A FABLE REVISITED, the Design Stage:

(1)
(2)

(3)
(4)
(5)

is not the first stage

is the first stage and the first activity in that stage is

to analyze the setting

is the first stage and the first activity in that stage is to
identify the problem

is the second stage and the first activity in that stage is
to identify performance objectives

I really don't know

After entering the second stage, our pig heroes first:

(1)
(2)
(3)
(4)
(5)

When

(1)
(2)
(3)
(4)
(5)

developed a model of the proposed solution
identified the proposed solution

recycled performance data

identified performance objectives

I really don't know

the team was testing the model, the "huff and puff" control was:

set on medium speed

set on high speed

not set on high speed
not shown in the picture
I really don't know

90

35. After checking the results of the simulated wolf attack, the pigs

discovered that:

(1) they needed more defense devices
(2) there was a weakness in the chimney
(3) they needed a trap door

(4) the fireplace Showed a weakness

(5) I really don't know

APPENDIX C

ERROR PERCEPTION INSTRUMENT

AND RESPONSE FORM

ERROR-PERCEPTION INSTRUMENT
(Tape Recorded)

I hope you have enjoyed viewing and listening to "A Fable
ReVisited" and participating in this validation. FOr your information
the slides were prepared by taking pictures of original art wOrk
of professional quality.

Our final task is to respond to 10 short-answer questions
regarding the visual materials. You'll have approximately eight
secondstx>respond to each question on the answer sheet provided.
Remember, this concerns only the visual material.

QUESTION 1

Did you observe any deviation from the generally high quality
of the visual material? Circle yes or no.

QUESTION 2

Did you observe any instances of what might be called pro-
duction mistakes or errors in photography? Circle yes or no.

QUESTION 3*

Production mistakes include such matters as misspelled words,
poorly focused or "fuzzy--unclear" slides, improperly exposed
slides which are excessively light or dark and poor copy tech-
niques which allow things to Show in the slide which shouldn't
be there. Did you notice any instances of these kinds of
mistakes? Circle yes or no.

QUESTION 4

If your answer to the last question was yes, how many mistakes
do you think you observed? write the estimated number.

QUESTION 5
With the exception of the word tunnel which was intentionally

misspelled did you notice any misspelled words? If yes, write
the estimated number.

91

92

QUESTION 6

Four different versions of "A Fable Revisited" were shown today.
The one you saw contained either pg mistakes, four mistakes,
eight mistakes or twelve mistakes. Which did you see--none,
four, eight or twelve? Write the number.

QUESTION 7

Assuming you saw a version with mistakes, did these bother you
in any way? Circle yes or no.

QUESTION 8

If your answer to the last question was yes, did any one type
of mistake bother you more than the others? If yes, indicate
which--copy, exposure, focus or spelling by circling c, e, f
or s.

QUESTION 9

Assuming you saw a version with mistakes, did you find your-
self counting or keeping track of them? Circle yes or no.

QUESTION 10
Upon reflection, how many of each of the possible mistakes do

you think you saw. write the estimated number opposite the
letter standing for copy: exposure, focus and spelling.

APPENDIX D

CREDITS FOR "A FABLE REVISITED"

93

RESPONSE FORM - PART II

QUESTION 1

 

YES ‘ No
QUESTION 2
YES NO

QUESTION 3

 

YES NO

QUESTION 4

 

 

QUESTION 5

 

 

QUESTION 6

 

 

QUESTION 7

 

YES NO

QUESTION 8

 

QUESTION 9

 

YES NO

QUESTION 10

 

 

 

 

 

C REDI TS

The author wishes to acknowledge the use of "A Fable Revisited."
This material was designed and produCed for the National Special
Media Institutes, Instructional Development Institute by Doug
Armstrong, Peggy Laird, Arlene Magnus, Russ McGregor, and Sue Meador,
all from the University of Southern California. The excellent
art work was by Ms. Laird and the entire production was adapted from
an original idea by Dr. Roger A. Kauffman. "A Fable Revisited"
is used by permission of the National Special Media Institutes,

Dr. Charles F. Schuller, Director.

94

APPENDIX E
INSTRUCTIONAL DEVELOPMENT MODEL

(IDI)

_eo.\.zmz

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

mm:o_z:ome - eeee
34 222345 _ 292343.
moons 80:52 53.50
3.5.... $>E83o 535 .6828 m._.<3._<>m
muo>omm mhmzmmm mma>h0homa :
.\ezmzu3e2_ u~>e<z< emu»
34.52:
292356 Sou:
3552: 29.82.52. 82 3.335
34298352. 3.25.5 3: .2233: m04m>mo
amatbhoma maozhmz mm>_._.om_.mo
52:sz ; . tamem 1::sz
«338$ 238$. “:er
$2.32: 543:: 3:50;...
meceaazoemme 39:38 5.333
85: 3353 832 3.32 szma
hzwzmo<z<a oz_._.._.mw amomomd
, wN_z<omo wN>._<z< >n._._.zmo.

 

 

Juno: hzmzaoqm>mo ._<zo_._.o=m._.mz_

95

APPENDIX F
AUDIO CHANNEL CONTENT DETERMINATION FORM

(mDIEIED IDI MODEL)

    

 

   

uqu>ou¢
\hzuauaaa.

 

«warhobbca
poachmzoo

 

hzulue<z<a
m~.z<@¢o

 

 

 

   

”hi—mu:
wN>4<z<

  

wastage:
hmuh

 

 

 

weoxeua
tame...

 

 

  

awaken .60

>..=._.zmo.

 

 

    

car—Fum
u~>4<z¢

 

   

Sumac»...
€3.23.

 

mks: <>u

eo3u>ue

22$:

 

Juno: hzmaaoam>uo ._<zo_._.o=¢._.wz..

96

APPENDIX c
SCRIPT FOR "A FABLE REVISITED"
WITH ERROR LOCATION AND TYPE

AND CHANNEL RELATIONSHIPS

 

VISUAL CHANNEL

AUDIO CHANNEL

 

Slide Time Time

 

 

 

 

1 7.5 6.5 G:

ERROR LOCATION

 

no ne

CHANNEL RELATIONSHIP

 

Now, let's consider how this space con-
quering strategy might be used in a
problem of less cosmic dimensions.

 

2 8.5 5.0 G:

ERROR LOCATION

 

none

CHANNEL RELATIONSHIP

 

unrelated

Take for instance a problem which faced
three of nature's creatures not too long
ago . . .

 

3 10.0 9.5 G:

ERROR LOCATION

 

none

CHANNEL RELATIONSHIP

 

unrelated

They lived--if you could call it that--
out in the country, where lurked a
number of problems--not the least of
which was simple survival.

 

4 9.5 9.0 G:

ERROR LOCATION

 

none

CHANNEL RELATIONSHIP

 

related

You remember these guys. Each one
looked for solutions to a pressing
problem in a different way--and with
varying degrees of success.

 

 

s 6.5 , 5.75 G:

ERROR LOCATION

 

none

CHANNEL RELATIONSHIP

 

redundant

It was this fellow--a very clever characn
ter indeed--who presented the most
immediate hazard.

 

 

97

98

 

VISUAL CHANNEL

AUDIO CHANNEL

 

Slide Time Time

 

 

 

 

6 6.5 6.25 G:

ERROR LOCATION

 

none

 

Now don't forget that we are looking at
this story with an eye to the Systems
Approach.

 

 

 

 

 

 

 

 

 

 

 

M: I gotcha.
CHANNEL RELATIONSHIP
unrelated
7 6.0 6.0 G Okay, the first pig, you will recall,
was a very improvident soul.
ERROR LOCATION
M: My hero.
none
CHANNEL RELATIONSHIP
unrelated
8 14.0 14.25 G: Right. It is pretty obvious that he
didn't use the systems approach.
ERROR LOCATION In fact, he had no approach at all--
let alone a systems approach. He
none
was happy and carefree.
CHANNEL RELATIONSHIP
M: Yeah, my hero.
related G: Just like you, George.
9 9.5 9.75 G: This pig--we'll call his "Straw Pig"
had never really seen the wolf. He
ERROR LOCATION concerned himself with the problem of
COpy, T1, T2, T3 protecting himself from the elements
of the weather.
CHANNEL RELATIONSHIP
unrelated
10 7.5 7.25 G: So our pig went on his way, easy and

ERROR LOCATION
none
CHANNEL RELATIONSHIP

redundant

carefree, until he met, by chance, a
man carrying straw.

 

 

99

 

VISUAL CHANNEL

AUDIO CHANNEL

 

Slide Time Time

 

 

 

 

11 7.0 6.75

ERROR.LOCATION

 

none

CHANNEL RELATIONSHIP

 

Ah, this would do nicely for a house!
Besides it was free or inexpensive
building material

 

 

 

 

 

 

 

 

 

 

 

related

12 6.5 5.25 G: "Straw Pig" had no plans. Pig lore had
it that some protection was helpful

ERROR LOCATION for pig survival.

none

CHANNEL RELATIONSHIP

unrelated

13 6.0 5.25 G: His solution strategy was simple: just
build a house for protection from the '

ERROR LOCATION wind and rain.

none

CHANNEL RELATIONSHIP

related

14 8.5 8.25 G: As we have seen, the straw house was
built with little or no thought . . .

ERROR LOCATION and then guess what happened?

Exposure, T3 M: The wolf came.

CHANNEL RELATIONSHIP

unrelated G: Right'

15 7.5 8.0 G: With minimum effort and a little

ERROR LOCATION

 

no ne

CHANNEL RELATIONSHIP

 

related

huffing and puffing on the part of the
wolf the straw house was an easy
victory.

 

 

100

 

VISUAL CHANNEL

AUDIO CHANNEL

 

Slide Time Time

 

 

 

 

_ 16 8.0 7.75 G:

ERROR LOCATION

 

none

CHANNEL RELATIONSHIP

 

related

Perhaps "Straw Pig" had been too hasty.
It's entirely possible that he got the
cart before the horse, so to speak.

 

17 i 3.5 3.25 M:

ERROR LOCATION

 

non e

CHANNEL RELATIONSHIP

 

unrelated

He was just busy playing and gamboling
around, that was all.

 

18 12.0 12.0 G:

ERROR LOCATION

 

none

CHANNEL RELATIONSHIP

 

related

Yes, but meanwhile, our second pig
named "Twig" has been casually inquiring
about the prospects of a wolf attack.

He discovers that wolves huff and puff.

 

19 6.0 6.25 G:

ERROR LOCATION

 

no ne

CHANNEL RELATIONSHIP

 

related

His neighbors and "Straw Pig" warn
him about the especially clever wolf
who skulks nearby.

 

 

20 6.0 ’ 6.0 G:

ERROR LOCATION

 

no ne

CHANNEL RELATIONSHIP

 

unrelated

"Twig Pig" is convinced not only should
he protect himself from the weather,
but also from the wolf.

 

 

101

 

VISUAL CHANNEL

AUDIO CHANNEL

 

Slide Time Time

 

 

 

 

21 5.0 4.5 G:

ERROR LOCATION

 

no ne

CHANNEL RELATIONSHIP

 

related

"Twig Pig" quickly looks into the
methods and materials available to him.

 

22 11.0 10.75 G:

ERROR LOCATION

 

no me

CHANNEL RELATIONSHIP

 

related

Now, based on the experience of your
"Straw Pig"--he decides to use something
more creative and effective! Genuine
lath and plaster twigs would do the
trick!

 

23 14.0 13.75 G:

ERROR LOCATION

 

none

CHANNEL RELATIONSHIP G:

 

unrelated

"Twig Pig" has a better insight into the
situation, right?

Yeah. But he hasn't gone far enough.
His information is casually selected--

it's incomplete. After all, what does
he actually know about the wolf?

 

24 10.0 10.25 M:

ERROR LOCATION

 

no me

CHANNEL RELATIONSHIP M:

 

unrelated

Well, he knows he huffs and puffs and
stuff like that.

Right. He thought his plan might work.

Well that's about all you can hope for
in this world.

 

 

25 6.0 6.75 G:

ERROR LOCATION

 

no ne

CHANNEL RELATIONSHIP

 

unrelated

Now, wait a minute, back at the wolf
lair, our antagonist--in his own crafty
and intuitive way . . .

 

 

102

 

VISUAL C

AUDIO CHANNEL

 

Slide Time Time

 

 

 

 

26 6.5 7.0 G:

ERROR LOCATION

 

no ne

CHANNEL RELATIONSHIP

 

related

looks into his problem. Being a very
hungry fellow, he is highly motivated
to succeed,

 

27 5.0 4.0 G:

ERROR LOCATION
Spelling, T1, T2, T3
CHANNEL RELATIONSHIP

 

 

he has spent a good deal of time
studying and observing pigs.

 

 

 

 

 

 

 

 

related

28 11.5 10.5 G: Would you say our wolf is well prepared?

ERROR LOCATION M: Ah hum, ah hum . . .

none G: Yes, he uses some very creative

CHANNEL RELATIONSHIP approaches before resorting to the

related huff and puff buSIness.

29 5.0 5.0 G: And . . . well you have to admit George,
the planning wasnt' too thorough here--

ERROR LOCATION was 11:?

none

CHANNEL RELATIONSHIP

related

30 7.0 6.5 G: The problem sure overcomes our loveable

ERROR LOCATION
none
CHANNEL RELATIONSHIP

related

but less than successful firend. Guess
he fOrgot something too.

 

 

103

 

VISUAL CHANNEd

AUDIO CHANNEL

 

Slide Time Time

 

 

 

 

31 6.5 5.75 G:

ERROR LOCATION

 

none

 

But you'll be happy to know that "Twig”
and "Straw" escape by the hairs of
their you know whats.

 

 

 

 

 

 

 

 

 

 

 

 

M: Oh yeah.
CHANNEL RELATIONSHIP
related
32 5.5 5.5 G: Now for the moment you have been waiting
for. Is pig #3 a systems pig?
ERROR LOCATION
none
CHANNEL RELATIONSHIP
related
33 6.0 5.5 G: How will he react to the wolf threat?
Flippantly lay his plans?
ERROR LOCATION ‘
M: I bet he will.
none
CHANNEL RELATIONSHIP
related
34 3.5 3.5 G: Casually gather information?
ERROR LOCATION M: That's the way he'll do it.
none
CHANNEL RELATIONSHIP
related
35 4.5 4.25 G: Or will he approach his problem
differently?
ERROR LOCATION
M: Go on, go on . .

none
CHANNEL RELATIONSHIP

related

 

 

104

 

VISUAL CHANNEL

AUDIO CHANNEL

 

Slide Time Time

 

 

 

 

36 2.5 2.5 G:

ERROR LOCATION

 

none

CHANNEL RELATIONSHIP

 

He is our system man--I mean pig.

 

 

redundant

37 13.5 12.75 G: First, the DEFINE stage. TOgether with
"Twig" and "Straw" pigs they proceed to

ERROR LOCATION thoroughly identify the problem.

Focus, T2, T3 G. Want to know how?

CHANNEL RELATIONSHIP.

 

 

 

 

 

 

 

 

 

o I
related M. Go ahead, you ll tell my anyway.
38 14.5 14.75 G: For instance they analyze the setting by
collecting information on their adver-
ERROR LOCATION sary: the wolf--his characteristics,
none size, likes, dislikes, you know--abi1i-
ties, habits and that type of thing.
CHANNEL RELATIONSHIP
related
39 8.0 8.5 G: You might say, they assessed their needs,
Well you might not say that.
ERROR LOCATION
M: No, I wouldn't have said that.
none
CHANNEL RELATIONSHIP
related
40 9.0 8.5 G: They decided WHAT MUST BE DONE to

ERROR LOCATION

 

Spelling, T3
CHANNEL RELATIONSHIP

 

redundant

protect themselves from the wolf by
finding out WHAT IS and comparing it to
WHAT SHOULD BE.

 

 

105

 

VISUAL CHANNEL

AUDIO CHANNEL

 

Slide Time Time

 

 

 

 

41 6.0 5.0 G:

ERROR LOCAT ION

 

none

 

. . . THEN IT'S ONLY A MATTER OF
assigning priorities and starting to
work, right?

 

 

 

 

 

 

 

 

 

 

 

M: Right.
CHANNEL RELATIONSHIP
redundant
42 7.0 6.75 G: Wrong: The team must still determine
what resources are available, like
ERROR LOCATION materials, people or pigs.
none
CHANNEL RELATIONSHIP
redundant
43 9.0 8.5 G: Recognizing the value of team work,

‘ System Pig had earlier organized the
ERROR LOCATION management by assigning tasks to himself
Exposure, T2, T3 and the others.

CHANNEL RELATIONSHIP

Related

44 12.5 12.0 G: The second stage in our team's survival
plan was the DEVELOPMENT stage. They

ERROR LOCATION first identified their performance

none objective. You knew they had to come
in here somewhere didn't you?

CHANNEL RELATIONSHIP

related M: I figured.

45 8.5 8.0 G: . . . to design, test and complete a

ERROR LOCATION

 

Focus, T3

CHANNEL RELATIONSHIP

 

redundant

a defense system against all possible
wolfian attacks, as soon as possible.

 

 

106

 

VISUAL CHANNEL

AUDIO CHANNEL

 

Slide Time Time

 

 

 

 

46 5.0 4.75 G:

ERROR LOCATION

 

none

CHANNEL RELATIONSHIP

 

related

Now they attempted to specify methods
which could be used to help solve the
problem.

 

47 6.0 5.75 G:

ERROR LOCATION

 

CHANNEL RELATIONSHIP

 

related

They called in experts and they con-
structed a model or a prototype of the
prOposed solution.

 

48 7.0 6.5 G:

ERROR LOCATION

 

none

CHANNEL RELATIONSHIP

 

unrelated

But . . . before building the actual
house our team found it necessary to
take care of some other details.

 

49 10.5 10.5 G:

ERROR LOCATION

 

Exposure, T1, T2, T3

 

They now entered the final or evaluation
stage. First they tested the model with
a simulated wolf attack . . . if you
can imagine a simulated wolf!

 

 

CHANNEL RELATIONSHIP M: Speaking
related
50 9.0 8.5 G: They analyzed the results of the test

ERROR LOCATION
Spelling, T2, T3
CHANNEL RELATIONSHIP

 

 

related

and concluded that in order to outfox
the wolf, something had to be done
about the chimney.

 

 

107

 

VISUAL CHANNEL

AUDIO CHANNEL

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Slide Time Time

51 6.0 6.0 M: To outfox the wolf--that's like being
Shanghaied to TOkoyo.

ERROR LOCATION

G: Right!

none

CHANNEL RELATIONSHIP

unrelated

52 7.5 7.75 G: The data were recycled and modifications
made. Now our friends are ready to

ERROR LOCATION implement their plan on a full scale.

none

CHANNEL RELATIONSHIP

related

53 6.5 5.75 G: The final house was also tested.
Although the brick house took longer

ERROR LOCATION to build . . .

none

CHANNEL RELATIONSHIP

related

54 3.0 3.25 G: When the last brick was in place, they

ERROR LOCATION

 

none
CHANNEL RELATIONSHIP

related

were ready:

 

 

55 6.0 3.5

ERROR LOCATION

 

none
CHANNEL RELATIONSHIP

related

So the stage is set for the final scenes
in our drama.

 

 

108

 

 

 

 

 

 

VISUAL CHANNEL AUDIO CHANNEL
Slide Time Time
56 2.5 Change after §_seconds of silence.

ERROR LOCATION

 

none
CHANNEL RELATIONSHIP
related - 55

 

57 2.5 Change after 3 seconds of silence.

ERROR LOCATION

 

none

CHANNEL RELATIONSHIP

 

related - 55

 

58 2.5 Change after §_seconds of silence.

ERROR LOCATION
none

CHANNEL RELATIONSHIP

 

lrelated - 55

 

 

 

59 7.5 5.75 G: This part is the one in which our
System Pig's "System Approach" is a real

ERROR LOCATION life-saver.

none

CHANNEL RELATIONSHIP

 

 

 

 

unrelated

60 7.0 6.75 G: The wolf came and he huffed and puffed
. . and he puffed and he huffed to

ERROR LOCATION no avail.

none

CHANNEL RELATIONSHIP

unrelated

 

 

109

 

VISUAL CHM

AUDIO CHANNEL

 

Slide Time Than

 

 

 

 

A 61 3.0 3.0 .G:

ERROR LOCATION
unrelated
CHANNEL RELATIONSHIP

 

related

A pig more clever than he? But as you
know '

 

62 6.5 5.25 G:

ERROR LOCATION
COPY! T1, T2, T3
CHANNEL RELATIONSHIP

related

there is more than one way to skin a pig
and the adversary has alternate
strategies.

 

63 4.0 5.0

ERROR LOCATION

 

none
>CHANNEL RELATIONSHIP
related - 62

 

64 4.5 5.0

ERROR LOCATION

none

CHANNEL RELATIONSHIP
unrelated - 62

 

 

65 4.5 5.0

ERROR LOCATION

 

none
CHANNEL RELATIONSHIP
related - 62

 

 

110

 

VISUAL CHANNEL

AUDIO CHANNEL

 

Sl ide Time Time

 

 

 

 

66 4.5 21.25

ERROR LOCATION

 

none

CHANNEL RELATIONSHIP

 

After--seconds of music.

 

 

 

 

 

 

 

 

 

 

 

 

related

67 8.0 8.5 G: NOw, it's important to note that our
team survived not necessarily because

ERROR LOCATION they were smarter than other pigs, just
more knowledgeable.

none

CHANNEL RELATIONSHIP

unrelated

68 6.5 5.0 G: System pig knew about the System Approach
to problem solving and he used it.

ERROR LOCATION

Focus, T3

CHANNEL RELATIONSHIP

unrelated

69 6.0 6.0 G: And don't forget, a casual intuitive
approach to problems may be better than

ERROR LOCATION none at all . . .

none

CHANNEL RELATIONSHIP

unrelated

70 3.0 2.25 G: But a system approach is better still.

ERROR LOCATION

 

none

CHANNEL RELATIONSHIP .

unrelated

 

 

111

 

VISUAL CHANNEL

AUDIO CHANNEL

 

Slide Time Time

 

 

 

 

71 7.0 5.5 G:

ERROR LOCATION

 

no ne

CHANNEL RELATIONSHIP

 

unrelated

The moral of this fable is--the world
belongs to him who does his homework.

 

ERROR LOCATION

 

CHANNEL RELATIONSHIP

 

 

ERROR LOCATION

 

CHANNEL RELATIONSHIP

 

 

ERROR LOCATION

 

CHANNEL RELATIONSHIP

 

 

 

ERROR LOCATIONS

 

CHANNEL RELATIONSHIP

 

 

 

APPENDIX H

INSTRUCTION-GROUP ASSIGNMENT FORM

VALIDATION PROCEDURES

Please read the following directions carefully.

Make sure you have a pencil.

l.

2. Locate your name under group 0, l, 2 or 3 below.

3. Note that group numbers are displayed on the screens. Proceed
immediately to the seating area in front of the screen displaying
your group number.

4. On the seats are packets of materials. Your name in on one. Please
locate it and sit in that seat.

5. DO NOT OPEN this packet until you are instructed to do so.

6. Shortly, an image will be projected on the screen. If you have

difficulty reading the material exchange places with someone
nearer the screen.

0 l 2 3
Budziszewski, P. Billig, M. Berger, B. Adcock, A.
Cavanaugh, S. Coomes, C. Bowden, C. Bascom, K.
Champine, E. Croucher, L. Chance, L. Borean, S.
Garner, P. Frank, M. Dul, P. Cardinal, P.
Gedert, R. Gendein, C. Florea, K. Caruso, J.
Hyde, P. Gillman, S. Hodge, D. Czarnecki, L.
Leavens, G. Kaniarz, P. Jacoby, P. Evans, S.
Marks, L. Mabley, A. Janer, N. Gibler, M.
Mattice, B. Madden, P. Kronberg, L. Gifford, M.
Maurer, M. Mihaly, D. Kuczynski, D. Grymonprez, D.
McGrath, M. Morrison, C. Medley, D. Hunt, C.
Meloche, R. Nobliski, K. Montei, D. McQueen, J.
Mohr, M. Peggs, C. Olton, L. Mieczkowski, L.
Prevost, A. Reinertson, S. Oneil, K. Monks, K.
Reeves, R. Revak, D. Packer, S. Payne, L.
Samra, M. Ryde, L. Roller,.M. Ross, D.
Wander, M. Sturley, J. Rowe, L. Stein, C.
Webb, L. Towslee, L. Thorndill, R. Smith, L.
Wonnacott, S. Yancey, P. Tilben, B. Swartzendruber, R.

112

Vansyckle, J.

APPENDIX I

SCRIPT OF TAPE RECORDED INTRODUCTION

PRE-PRESENTATION INTRODUCTION
(Tape Recorded)

Good afternoon. In a few moments you will see and hear a slide
tape presentation entitled A FABLE REVISITED. This short program con-
sists of a series of slides and an audio-tape narration and was produced
at the University of Southern California for a nation wide training pro-
ject carried out by the National Special Media Institute which has its
national office at Michigan State University. In its entirety the
presentation is designed to introduce the concept of systems to educators
and the strategy goes something like this.

Mary, who is an elementary teacher, is attempting to explain the
systems process to one of her colleagues and convince him that this
approach is valuable for certain educational purposes. George has
difficulty accepting and comprehending either the value or the procedure.
In a final almost desperate effortMary turns to an example which she]
feels even George can understand.

Before joining some old friends I would like to offer a few words
of explanation. You have probably surmised that we are going to
compare something--in this case four versions of our story. This is a
validation and should tell us which version is most effective. In the
process you will acquire some new and useful information. There is
still another reward in that we will discover some facts about this
type of presentation which will be of immediate importance and applica-
tion to virtually every teacher in this country. The extent to which
the results are valuable depend entirely upon your attention and
cooperation during the next few minutes. Following the Slide tape you

will be asked to respond to a series of questions to see what you

113

114
have learned and then we will ask your opinions on certain aspects of

the program. Now, let's join Mary.

BIBLIOGRAPHY

BIBLIOGRAPHY .

Allen, William. "Instructional Media Research: Past, Present, and
Future." Audiovisual Communication Review, 19 (1971), 5-18.

 

Anderson, Charles. "In Search of a Visual Rhetoric for Instructional
Television." Audiovisual Communication Review, 20 (1972), 43—63.

 

Andrews, Charles R. "The Effect of Short-term Interruptions Upon
Factual Acquisition from Motion Picture Films Among Intermediate
Elementary School Pupils." Audiovisual Communication Review,

20 (1972), 348 (Abstract).

Arnheim, Rudolf. "What Do the Eyes Contribute?" Audiovisual Communi-
cation Review, 10 (1962), 10-21.

 

 

Ball, JOhn and Byrnes, Francis C. (Ed.) Research Principles and
Practices in Visual Communication. Washington D. C.: Department
of Audiovisual Instruction, National Education Association, 1960.

 

 

Berlo, David K. The Process of Communication: An Introduction to
Theory and Practice. New York: Holt, Rinehart and Winston, 1960.

 

Blackwell, H. R. "Vision Engineering." Lighting, (July, 1963 - March,
1967).'

Bollman, Charles G. "An Index to Doctoral Dissertations in Educational
Media." Unpublished manuscript, Michigan State University, 1967
(mimeographed).

Bourisseau, Whitfield; Davis, 0. L. Jr.; and Yamamoto, Kaoru. "Sense-
impression Responses of Negro and White Children to Verbal and
Pictorial Stimuli." Audiovisual Communication Review, 15 (1967),
259-268.

 

Broadbent, D. E. Perception and Communication. London: Pergamon
Press, 1958.

 

Campeau, P. L. Selective Review of the Results of Research on the
Use of Audiovisual Media to Teach Adults. Strasbourg: Council
of Europe, 1972.

 

Cherry, Colin. On Human Communications: A Review, A Survey, and a
Criticism. New York: JOhn Wiley, 1957 (Science Ed. 1961).

 

115

116

Chu, Godwin C., and Schramm, Wilbur. Learning From Television: What
the Research Says. Stanford University, Contract 2 EFC 708 94,
0.8. Office of Education, 1967. '

 

 

Conway, Jerome K. "Multiple-sensory Modality Communication and the
Problem of Sign Types." Audiovisual Communication Review, 15
(1967), 371-383.

Cornog, D. Y., and Rose, F. C. Legibility of Alphanumeric Characters
and Other Symbols: A Reference Handbook. National Bureau of
Standards Miscellaneous 262-2. U. S. Government Printing
Office, Washington, 1967.

 

Debes, John L., III. Personal Conversation. Eastman Kodak Company,
1973.

 

Deutschmann, Paul J.; Barrow, Lionel C. Jr.; and McMillan, Nita.
"The Efficiency of Different Modes of Communication."
Audiovisual Communication Review, 9 (November - December, 1961),
263-270. '

 

Dwyer, Francis M. Jr. "Exploratory Studies in the Effectiveness of
Visual Illustrations." Audiovisual Communication Review,
18 (Fall, 1970), 235-249.

Dwyer, Francis M. Jr. A Guide for Improving Visualized Instruction.
State College, Pennsylvania: Learning Services, 1972.

Eastman Kodak Company. Eye for Learning. Rochester, New York: Eastman
Kodak Company, 1967.

Ely, Donald P. (Ed.) "The Changing Role of the Audiovisual Process
in Education: A Definition and a Glossary of Related Terms."
Audiovisual Communication Review, 2 (January - February, 1963).
3-148.

Fleming, Malcolm. "Classification and Analysis of Instructional Illus-
trations. Audiovisual Communication Review, 15 (1967) 246-58.

 

Fonseca, L., and Kearl, B. "Comprehension of Pictorial Symbols: An
Experience in Rural Brazil." Wisconsin Department of Agriculture
Journal, 30 (1960).

Fullerton, B. J. "The Comparative Effect of Color and Black and
White Guidance Films Employed With and Without 'Anticipatory'
Remarks Upon Acquisition and Retention of Factual Information."
Dissertation Abstracts, 16; 1413 (1956).

 

Gibson, James J. "A Theory of Pictorial Perception." Audiovisual
Communication Review, 2 (Winter, 1954), 3-23.

117

Harrison, Randall P. "Pictic Analysis: Toward a Vocabulary and Syntax
for the Pictorial Code: With Research on Facial Communication.”
Unpublished Ph.D. dissertation, Michigan State University, 1964.

Hartman, Frank R. "Single and Multiple Channel Communication: A
Review of Research and a Proposed Model." Audiovisual Communica-
tion Review, 9 (November - December, 1961), 235-62.

 

Hoban, Charles F., and van Ormer, E. B. Instructional Film Research
1918-1950 (Rapid Mass Learning). SCD Human Engineering Project
20-E-4, The Pennsylvania State College, Department of the Army
and Department of the Navy, 1951.

Hsia, Hower J. "On Channel Effectiveness." Audiovisual Communication
Review, 16 (1968), 245-67.

Hsia, Hower J. "Intelligence in Auditory, Visual, and Audiovisual
Information Processing." Audiovisual Communication Review 17
(1969), 272-82.

Hsia, Hower J. "The Information Processing Capacity of Modality and
Channel Performance." Audiovisual Communication Review, 19
(Spring, 1971), 51-75.

Ketcham, Carl H. and Heath, Robert W. ”Teaching Effectiveness of Sound
With Pictures That DoNot Embody the Material Being Taught."
Audiovisual Communication Review, 10 (March-April, 1962), 89-93.

 

Levie, W. Howard. Technical Quality and Color in Training_§ilms, In
Acquiring Teaching Competencies, National Center for the Develop-
ment of Training Materials in Teacher Education. School of
Education, Indiana University. Report #4, 1973 (mimeographed).

 

Long, A. L. "The Influence of Color on Acquisition and Retention as
Evidenced by the Use of Sound Films." Unpublished Doctoral
Dissertation, University of Colorado, (date unknown).

Lumsdaine, Arthur H. and Gladstone, A. "Overt Practice and Audio-
Visual Embellishments.” In May, M. A. and Lumsdaine, A. H.
(Ed.), LearningyFrom Film. New Haven, Conn.: Yale University
Press, 1958, 58-71.

 

Lumsdaine, Arthur H. "Instruments and Media of Instruction." In Gage,
N. L. (Ed.), Handbook of Research on Teaching. Chicago: Rand
McNally, 1963, 583-682.

McCoy, Edward P. "Influence of Color on Audiences' Rated Perception
of Reality in Film." Audiovisual Communication Review, 10
(January-February, 1962), 70-72.

McLean, W. P. "A Comparison of the Effectiveness of Colored and
Uncolored Pictures." Unpublished Doctoral Dissertation,
University of Chicago. (Date unknown).

118

Meierhenry, Wesley C. (Ed.) "Needed Research in the Introduction and
Use of Audiovisual Materials: A Special Report." Audiovisual
Communication Review, 10 (November-December, 1962), 307-16.

 

 

Miller, Neal E. "Graphic Communication and the Crisis in Education."
Audiovisual Communication Review, 5 (1957), 3-120. (Special
Supplement).

 

Moldstad, John. ”Doctoral Dissertations in Audiovisual Education."
Audiovisual Communication Review, 4 (Fall, 1956), 291-333.

 

Moldstad, John, "Doctoral Dissertations in Audiovisual Education."
Audiovisual Communication Review, 6 (Winter, 1958), 33-48.

 

Moore, David M. and Sasse, Edward B. "Effect of Size and Type of
Still Projected Pictures on Immediate Recall of Content.”
Audiovisual Communication Review, 19 (Winter, 1971), 437-50.

 

Neu, D. M. Effects of Attention-gaining Devices on Film-mediated
Learning. Report SDC 269-7-9. The Pennsylvania State Univer-
sity, United States Naval Training Devices Center, 1950.

Neu, D. M. "The Effect of Attention-gaining Devices on Film-mediated
Learning." Journal of Educational Psychology, 42 (1951),
479-90.

Perrin, Donald G. "A Theory of Multi-image Communication." Audiovisual

 

Communication Review, 17 (Winter, 1969), 368-82.

 

Popham, W. James. "Pictorial Embellishments in a Tape-Slide Instruc-
tional Program." Audiovisual Communication Review, 17 (Spring,
1969), 28-35.

 

Razinsky, E. L. "The Effects of Message Quality Deterioration on
Message and Source Evaluation, Attitudes, and InfOrmation Gained."
Unpublished masters thesis, Michigan State University, 1967.

Saul, Ezra V. and others. A Review of the Literature Pertinent to
the Design and Use of Effective Training Aids. Tech report SDC
494-08-1, Tuffs College for Office of Naval Research, February,
24, 1956.

Schlanger, B. "Criteria for Motion Picture and Viewing a New 70 mm
System: Its Process and Viewing Arrangements." Journal of the
Society_of Motion Picture and Television Engineers, 75 (1966),
161-67.

Schlater, Robert W. "Effect of Speed of Presentation and Irrelevant
Cues on Recall of Television Messages." Unpublished Ph.D.
dissertation, Michigan State University, 1966.

119

Seibert, W. F.; Kasten, D. F.; and Patter, J. R. "A Study of Factors
Influencing the Legibility of Televised Characters." Journal
of the Society_of Motion Picture and Television Engineers,

LXV III (1959). 461-72.

Sencer, RObert A. "A Investigation of the Effects of Incorrect
Grammar on Attitudes and Comprehension in Written English
Messages." Unpublished Ph.D. dissertation, Michigan State
University, 1965. '

Severin, Werner. "Another Look at Cue Summation." Audiovisual
Communication Review, 15 (Fall, 1967), 233-45.

 

Severin, Werner. "The Effectiveness of Relevant Pictures in Multiple-
channel Communication.” Audiovisual Communication Review, 15
(Winter, 1967), 386-401.

Shamo, G. W. and Meador, L. M. "The Effect of Visual Distraction Upon
Recall and Attitude Change." The Journal of Communication, 19
(1969). 157-62.

Shannon, Claude E. and weaver, Warren. The Mathematical Theory of
Communication. Urbana: University of Illinois Press, 1949.

 

Smith, Karl U. "Audiovisual Teaching: A New Dimension in Education
and Research." Audiovisual Communication Review, 8 (May-June,
1960), 85-103.

Spaulding, Seth. "Communication Potential of Pictorial Illustrations."
Audiovisual Communication Review, 4 (Winter, 1956), 31-46.

Spaulding, Seth. "Scientific Principles of Textbook Design and
Illustration:' A Comment.” Audiovisual Communication Review,
9 (January-February, 1961), 60-62.

Spooner, H. James. "An Exploratory Study to Determine Guidelines for
the Selection and Production of Visuals for Use on Controlled-
Scan Television in a Classroom Setting," Unpublished Ph.D.
dissertation, Michigan State University, 1973.

Travers, Rdbert M. W. "The Transmission of Information to Human
Receivers." Audiovisual Communication Review, 12 (Winter, 1964)
373-85.

Travers, Robert M. W. Research and Theory Related to Audio-visual
Information Transmission. (Rev. Ed.) Western Michigan
University, OE Contract No. 3-20-003, U. S. Department of Health,
Education and Welfare, 1967.

Travers, RObert M. W. and Alvarado, Victor. "The Design of Pictures
for Teaching Children in Elementary School." Audiovisual
Communication Review, 18 (Spring, 1970) 47-64.

 

 

120

Tukey, J. w. "The Problem of Multiple Comparisons." Unpublished
Manuscript, Princeton University, 1953.

VanderMeer, A. w. "Relative Contributions to Factual Learning of the
Pictorial and Verbal Elements of a Filmstrip." School Review,
58 (1950). 84-89.

VanderMeer, A. w. "Color Vs. Black and White in Instructional Films.
Audiovisual Communication Review, 2 (1954), 121-34.

VanderMeer, A. w. Relative Effectiveness of Color and Black and White
in Instructional Films. Progress Report No. 9, State College,
Pennsylvania, Pennsylvania State College, Instructional Film
Research Program.

Williams, Catherine M. Learning From Pictures. Washington D. C.:
Department of Audiovisual Instruction, National Education
Association, 1963.

Williams, Catharine M. and Debes, John L., III. Proceedings of the
First National Conference on Visual Literacy. New York:
Pitman Publishing Corporation, 1970.

MICHIGAN STATE NI V LI RRRI ES

32||| 3|3|||||3 ||| ||||||3|||3|||3 |3||