llllfllllljlllllllil’llilllllillllilllll Z ”“7 7035 93 00891 0055 f LIBRARY Michigan State University \ A ‘— This is to certify that the thesis entitled SOUWA TherAeK m HfPERmealA; AN WWW; Shel») unTh Hypfifiwed. presented by PEr‘lc‘e5 V Gomeg has been accepted towards fulfillment of the requirements for MASTER 0f AQT‘Sdegree in TE/EOOMMVNWfl-neh/ l 0-7639 MS U i: an Affirmative Action/Equal Opportunity Institution Date fien/ 16/ 9o PLACE IN RETURN BOX to remove this checkout from your record. TO AVOID FINES return on or before date due. DATE DUE DATE DUE DATE DUE J ‘1 M” W ____L _ MSU to An Affirmative Action/Equel Opportunity Institution emu-gunman AUDIO FEEDBACK IN HYPERHEDIA AN APPLIED STUDY WITH HYPERCARD by Pericles Games A THESIS Submitted to Michigan State University in partial fulfillment of the requirements for the degree of MASTER OF ARTS Departnent of Teleconnunication 1990 6460- .7 ’7. ABSTRACT AUDIO FEEDBACK IN HYPERMEDIA AN APPLIED STUDY WITH HYPERCARD BY Pericles V. Games With Hypermedia is possible to create interactive applications which link video, animation, still images, text and sounds. This study compares impacts of three modes of auditory feedback which occur when users sends a command to the system. The nodes include: a beep whenever a ”button" is pressed -BEEP, vs. a wide variety of sounds -ALL DIFFERENT, vs. no feedback -SILENT. Using an experimental design which controls for individual differences by exposing all subjects to all conditions, 18 college students explored 6 different Hypercard stacks related to Mars exploration. There were three conditions for each of the six stacks (BEEP, ALL DIFFERENT and SILENT). The design assigned each student two stacks from each condition; each stack was used 6 times in each condition and the order of presentation of condition and stack was balanced. A pre- posttest after each stack and a final posttest were administered. The experimental style of ALL DIFFERENT feedback was very well received and the students strongly preferred this condition. All 18 students chose ALL DIFFERENT over BEEP. 89% chose ALL DIFFERENT over SILENT. Reactions were split between SILENT and BEEP. There was no significant difference among conditions in terms of buttons clicked on, time spent exploring or facts recalled. However, in each case, the raw data showed nore favorable trends for ALL DIFFERENT feedback. Explorers in this condition tended to spend more time on segments, click on more different buttons and remember more facts. ACKNOWLEDGMENTS This project was completed through the Communication Technology Laboratory, College of Communication Arts and Sciences at Michigan state University. I would like to thank Dr. Carrie Jill Heeter for the knowledge, time and friendship she has shared with me during the development of this research and the Mars CD ROM project, and without whose support I may have never completed this thesis. TABLE OF CONTENTS Introduction Research Issues A. Time Variables Recall Variables Exploration Variables Credibility Variables Enjoyment Methods Description of the Experimental Stimuli Experimental Design The Sample Measurement Results Experimental study results Exploratory results Discussion Appendix Bibliography 12 13 13 14 14 15 16 16 25 26 28 31 31 35 40 43 48 Table Table Table Table Table LIST OF TABLES Diagram of the experimental design ANOVA statistics by Segments ANOVA statistics by feedback condition and Hypothesis testing ANOVA statistics by segment length Correlation Matrix within learning preferences, Type of sound buttons and cognitive orientation 27 32 33 36 37 Figure Figure Figure Figure Figure Figure Figure Figure LIST OF FIGURES The Merging of three diverse industries, by Nicholas Negroponte The instructions control panel of the CR-ROM "Expedition to Mars” The main screen of the segment A — "Problems for Humans in Space" The main screen of the segment 8 - ”Search for Life in Mars" The main screen of the segment C - "Comparing Mars to Earth” The main screen of the segment D - "Weather on Mars" The main screen of the segment E - "The Surface of Mars ” The main screen of the segment F - "Space Industries" 1 18 19 20 21 22 23 24 Broadcast & Motion Picture Industry Print & Publishing Industry r.‘ \ Computer Industry . Year 2000 Broadcast & M01101] J, , , Picture Industry 1 nnt & l ubhshmg Industry Computer Industry Figure l - The merging of three industres. from the book THE MEDIA LAB. Inventing the Future at MIT. by Stewart Brand 1 INTRODUCTION' ”Educational activities can evoke sensory curiosity by including audio and visual effects, such as music and graphics. They can evoke cognitive curiosity by leading learners into situations in which they are surprised." l Hypermedia's role in education will some day be a formidable one, the potential of which needs to be explored, studied and enhanced, particularly now, when traditional methods of education are at a crisis point and television has become such a commercially oriented and a highly competitive medium. In order to compete for attention and time with such a powerful medium, Hypermedia applications will need to be, more and more, kynesthesic, polished and compelling. What is possible with Hypermedia today is something potentially much more powerful than traditional audiovisual media. Hypermedia signifies the merging of three industries - broadcasting, publishing, and computing - which until now have remained surprisingly distinct. As these technologies and industries converge, new ways to build extraordinary learning environments are emerging. lith Hypermedia it is possible to create interactive, integrated, non-linear databases that link, annotate, and cross-reference video, animation, still images, text, saunas, and voice. In order to store such a large amount of information, there are, among others, two laser-based technologies: Videodisc and CD-ROM. Videodisc is an analog medium which can store-film, video, animation, still photos, text, and sound - about 54,000 frames on each side of a 12-inch disc. A CD-ROM (Compact Disc, Read-Only Memory) stores digital images, animation, sound, text and data - at least 550 Megabytes of information squeezed onto one 4.72-inch disc. A CD-ROM, in fact, is very similar to the digital audio compact disc. The only difference between them is that the CD-ROMs are formatted for computer data. Because of its digital format, it is an extremely cost- effective and durable medium for storing huge quantities of information. Stewart Brand, author of The Media Lab, describes it, "a cheap little parking place for 250,000 pages of text, the equivalent of 500 books - truckload - instantly computer-searchable and publishable at one-fiftieth the cost of printing on paper".2 By connecting a CD-ROM drive or Videodisc player to a microcomputer and using an authoring tool to catalog and link information, it is possible to design highly textured hypermedia learning experiences that stretch far beyond what traditional media can do. One of the most significant authoring tools available today is Hypercard, for the Macintosh 3 computer. With Hypercard, it is possible to design friendly user interfaces and organize the information on microcomputers similar to the way people do in their minds - nonsequentially, by association and context as well as hierarchy. Hypercard can be imagined as a "software construction kit” that gives someone the capabilities to merge different media - text, pictures, graphics, animation, voice, and sound - and to author a wide range of applications. It is also possible to create interactive courseware that allows students to pause and delve deeper into a subject they are curious about, stop and define a word, or jump from one related concept to another, either to follow their own interests or to respond to questions. In Hypercard, information is presented on "cards", which are kept in "stacks". It is possible to link one card to another by creating "BUTTONS". One of the most important features of these buttons - the one that will be studied here - is that they can be programmed to play sounds. There is increasing availability to education of mass storage to provide instant access to and readly availability of sound bytes and screens. In that context, this thesis provides an important chance to explore new forms of learning motivation. What can be done to encourage people in general and students in particular to care more about what they are learning? 4 Do people want to be informed, challenged, amused, guided, intrigued, or some combination of the above? What is their frame of reference, and how can they be tapped into it? What questions could be posed to pique curiosity? How can they be challenged to make discoveries on their own? The capabilities of Hypermedia are so new that it is not known yet which techniques work more effectively, which ones do not, what are the interactions between those capabilities. Hypermedia is an "infant” medium that is still being invented and needs to be researched. The Communication Technology Laboratory at Michigan State University is exploring part of the potential of Hypermedia and experimenting with how to present information and design effective user interfaces. Research is needed in order to find principles that can be generalized across different projects. One of many user interface decisions in Hypermedia design is how the application program responds and reacts to the user input. User inputs today typically take the form of keyboard keystrokes, mouse or trackball movement and clicks, or touches on a touch screen. The main concern of this study is system response to user choices. Responses of a Hypermedia information package may take several forms: 1. The application can simply execute the command; 5 2. A visual one can indicate that a particular command was made. It is common, when a user makes a choice, for that choice to become visually highlighted for a moment before the command is executed to go to the next screen, display additional information or perform some other activity. This is thought to give the user a sense of communicating with the system, and reassures the user that the input has been received. 3. An auditory cue can indicate that a selection has been made. This auditory cue serves a similar function of letting users know that their input has been received. With touch screens where a certain firmness of touch is necessary to send a command, it is important for the user to have feedback - about when they have pressed hard enough. Mainframes and videotex systems also often use auditory feedback just to let the user know that commands -which may take some time to execute- have been received. Applications also use auditory cues (usually beeps) to indicate that something wrong or not executable has been requested. Initially, ”beeps" were the primary form of audio feedback, mainly because computers had limited sound capabilities. But recently the sound capacity of microcomputers expanded dramatically, to the point where digitized sound of all forms can be stored on computer and played back at will. Speech synthesis is 6 also possible. The new capacities far exceed our experience and understanding of what are desirable and effective uses of sound in Hypermedia. This thesis will focus on the broadly applicable issue of utilizing sound capabilities of Hypermedia. Specifically, it will examine the effectiveness and desirability of using a variety of different sounds (rather than beeps) as auditory feedback to users. The study is based on an audio feedback style developed at the Comm Tech Lab. This style is the antithesis of a single, monotonous beep which is heard every time a command is issued. Instead, a Hypermedia application was developed (a CD ROM about Missions to Mars) which uses a different brief sound for almost every different command issued. In this Hypermedia project, developed with Hypercard, commands are issued by clicking on ”buttons" (visual and verbal representations of menu choice and other navigation commands). Thus, different buttons have different sounds associated with them. The impact of this new variable sound feedback style is not known. Users may find it intriguing, never knowing what sound they will hear the next choice they make. They may find it annoying or distracting. It may encourage greater exploration. It may enhance or even actually detract from learning. This thesis will empirically test the impacts of 7 different immediate audio-feedback styles on learning, enjoyment, persistence and other outcomes. Prior research addresses immediate audio feedback only tangentially. A review of pertinent studies found that most of the research done has focused on feedback that has to do with giving correct answers. Gilman (1969) suggested that providing the student with a statement of the correct response after errors, or reasons for the "correctness" of correct response was added to the effectiveness of computer-based instruction. Cohen (1985) states that informational feedback has its greatest effect after incorrect responses, assuming that the student is given enough information to take corrective action. She concludes that feedback after correct responses in not as important. In fact, feedback after correct responses may interfere with learning in students who have high confidence in their knowledge of the material (Cohen, 1985; Kulhavy,1977).3 Some studies discussed a second factor: the need (or lack of need) of immediate feedback about answers to questions. Although feedback was once deemed to be an essential element of all programmed instruction, some questions have been raised regarding the need for immediate feedback. Rankin and Trapper (1978) found that delayed feedback facilitated retention of information learned during a computer-based instruction task. Kulhavy and Anderson (1972) concluded that delayed feedback was effective only when material learned involves meaningful matter similar to that usually encountered in instruction. Immediate feedback appears to be most effective when the task involves discrimination learning (Carter, 1984), acquisition of knowledge (Cohen, 1985), or when the student is having difficulty achieving mastery of the material (Cohen, 1985). While the conclusion would seem to be that feedback can have a positive influence on learning, under, nertain_ggnditigns, the practical significance of the effectiveness of various types of feedback in Computer Aided Instruction is yet to be resolved. The purpose of this study is to analyse audio feedback related to immediate sound response every time a button is clicked in Hypercard/HyperMedia, and in this sense has not much to do with correctness of responses. The immediate audio feedback is closer related to the ”Curiosity factor" of Thomas Malone's Theory of Intrinsically Motivating Instruction (1981). In his theory, Malone distinguishes between two kinds of curiosity - sensory and cognitive curiosity- depending on the level of processing involved. It is sensory curiosity that is going to be studied here. Sensory curiosity involves the attention-attracting value of changes in the light, sound, or other sensory 9 stimuli of an environment. There is no reason why educational environments have to be impoverished sensory environments. Colorfully illustrated textbooks, television programs like Sesame Street, and tactile teaching devices (Montessori, 1912) demonstrate this point. Microcomputers provide even more possibilities for graphics, animation, music, and other captivating AUDIO and VISUAL EFFECTS. These effects can be used: (1) as decoration, (2) to enhance fantasy, (3) as a reward, and perhaps the most importantly, (4) as a representation system that may be more effective than words or numbers.4 The effectiveness of immediate audio feedback may depend on the type of sound response. In order to analyze this important factor the immediate audio feedback in this study will be subdivided in three different categories: (1) varied sound feedback, that is, when every button in a program has a distinct sound; (2) Beeping sound feedback, where the sound will be always one same beep and (3)silent sound feedback, or actually, no sound feedback at all. The varied sound feedback (1) will contain music and sound effects, and will never be longer 4 seconds. In this study, exploratory research and a formal experimental design are conducted, to advance the understanding of whether and how sensory curiosity motivates people and affects learning. 10 W 1 Thomas W. Malone and James Levin, ” Microcomputers in Education: Cognitive and Social Design Principles," in - fgn_ne§ign_and_n§gL Walker and Hess (Wadsworth, 1984). 2 The Apple Guide to Courseware Authoring, (Apple Computer Inc. 1988) 23. 3 Elizabeth Perez and Mary Alice White, ”Student Evaluation of Motivational and Learning Attributes of Microcomputer Software," — Instruction, (Spring 1985, Vol. 12, No.2) 39. 4 Thomas Malone, "Towards a Theory of Intrinsically Notivatinq Instruction." in W W lalke and Hess (Wadsworth, 1984) 89. 11 RESEARCH ISSUES The presence and the type of immediate audio feedback in Hypermedia may have a major impact on the amount of time spent and information retained by users. Some aspects of this proposition have been studied. According to Berlyne (1968) and Hunt (1966) motivational attributes include notions of curiosity or exploration of novelty. However, these factors are derived primarily from ' print learning and do not include aspects of the microcomputer, such as animation, visual effects, and sound. Banet (1979) and Malone (1980, 1981) have each formulated broader explanations of motivational qualities based on research with arcade games. Malone incorporates these game aspects into a framework for a theory of motivation. In that sense, the premise in this study is that variable sounds will encourage users to explore more the segments, based on the assumption that every time they click buttons they will be rewarded by getting to hear a new sound. Therefore, the general hypothesis is that: * Variable sound buttons enhances click motivation; * Silent buttons are neutral for click motivation; * Beeping buttons inhibits click motivation. The last assumption is based in the premise that beeping sounds are bad because it is somewhat, annoying, repetitive and monotone. 12 VA. Time Variables: Tatal_time_snant_nn_fiesment H1: There is a relationship between the total amount of time spent by users of Hypermedia applications and the type of Audio feedback present on the buttons. Specifically more time will be spent on applications that have buttons with varied sound feedback. W H2: There is a relationship between the amount of average time spent on screens by users of Hypermedia applications and the type of Audio feedback present on the buttons. Specifically, more time will be spent on screens that have buttons with varied sound feedback. B. Recall Variables: WWW H3: There is a relationship between the number of topics choices recalled from segments by users of Hypermedia applications and the type of Audio feedback present on the buttons. Specifically, more topics will be recalled from segments that have buttons with varied sound feedback. MW H4:There is a relationship between the number of facts recalled from segments by users of Hypermedia applications and the type of Audio feedback present on the buttons. Specifically, more facts will be recalled from segments that have buttons with varied sound feedback. 13 C. Exploration Variables Number_9f_buttans_clicked H5: There is a relationship between the number of buttons clicked in segments by users of Hypermedia applications and the type of Audio feedback present on the buttons. Specifically, more buttons will be clicked on segments with varied sound than on beeping ones. MW H6: There is a relationship between the number of screens accessed in segments by users of Hypermedia applications and the type of Audio feedback present on the buttons. Specifically, more screens will be accessed on segments with varied sounds than on beeping ones. D. Credibility Variables Qualitx_ef_nesian H7: There is a relationship between the quality of design perceived by users of Hypermedia applications and the type of Audio feedback present on the buttons. Specifically, Sound segments will appear better designed than beeping ones. ACCBLIQX H8: There is a relationship between the accuracy perceived by users of Hypermedia applications and the type of Audio feedback present on the buttons. Specifically, Sound segments will appear more accurate than beeping ones. 14 E. Enjoyment‘ W H9: There is a relationship between the level of interest by users of Hypermedia applications and the type of Audio .Feedback present on the buttons. Specifically, Sound segments will appear more interesting than beeping ones. Lerel.of.£n19xment. H10: There is a relationship between the level of enjoyment demonstrated by users of Hypermedia applications and the type Audio Feedback present on the buttons. Specifically, Sound segments will appear more enjoyable than beeping ones. In addition to these hypotheses, several research questions will be explored: User preference for different sound, beep and silent; relationship between preferred learning mode and type of sound feedback; preference for musical versus sound effect variable sound. 15 unruoos MW Six stand-alone segments were drawn from the Hypermedia software package, Mission to Mars, developed by the Michigan State University Chmmunication Technology Laboratory. The "Mission to Mars" CD ROM project itself comprises nearly 45 megabytes of information, with images, text, animations, sound, and narration, and is intended to offer an informal, exploratory hypermedia discovery learning experience. There is no fixed path through the information. At every step, the user is offered choices of what to see and hear, in what order. There is more information available than any single user is likely to encounter. Because users are free to select content in depth and at length or to ignore it, motivation to explore is a key factor. In this context, the immediate audio feedback present in the buttons is intended to elicit a curiosity factor (compensating people for pressing buttons). This context is very different from a tutorial that uses sound to confirm right of wrong answers. All six segments share a common user interface and presentation style. Content appears within a space ship control panel and windows. Common to all screens are a ”forward" and ”reverse" lever which move the user to 16 different screens within the segment, although not in an order the user can anticipate (see figure 2). Also, the space between forward and reverse contains different navigational "buttons”, varying depending on the screen. Within the spaceship window, graphics and text display menus and information. Buttons to click on for more information within the window are identified by a special symbol: (*) - A circle with a star inside. The six segments contain different content and varying numbers of screens and buttons: OVERVIEW OF SEGMENTS 12.329HEH1_NAHE;_ I_SSBEENS____1_EQIIQNE A- Surface of Mars 19 68 B- Problems for Humans in Space 6 20 C- Comparing Mars and Earth 8 31 D- Space Industry .19 80 E- Weather on Mars 7 24 F- Search for Life in Mars 19 82 The subject matter for all segments used in the study relates to Space Science. Surface of Mars referred to as segment A, contained images and text about the characteristics of the surface of Mars; Problems for humans in space, segment B, discussed the physical and psychological dangers and implications of long permanence of humans in space; Comparing Mars and 17 Instructions; \\‘ mdtvtdual buttons to he.” then function. The reverse lever moves you backward thru the Mars Mission cards. These buttons will very. Figure 2 - The instructions control panel of the Cit-ROM I'lxpeciition to Mars" 18 PRO-BITE HISWFVORI HUMAN 8; IN SPACE UUREDUH 3‘ ISULAYIUN J k SOLAR FLARESQ) RADIATION (g) j\ navigating buttons Figure 3 - The main screen of the segment A - "Problems for Humans in Space" 19 Mount 0 . Martian G) i Invasion ' Mariner 9 1971 0 Martian Channel 1 877 Viking 6) Results 69 1976 Figure 4 - The main screen of the segment B - "Search for Life in Mars" 20 @Slzc (9M00n5 CEGruVIty ISSeasons (“Atmosphere WUIstance from ear Length [J E) )r If (2 II t; t it Figure 5 - The main screen of the segment C - "Comparing Mars to Earth" 21 (21- ('2‘ nautgatingnbutton Figure 6 - The main screen of the segment D - "Weather on Mars" 22 nemat—u—mng Figure 7 - The main screen of the segment I - 'The Surface of Mar 23 S u Iplplorrit I n :9 Earth I‘ - . Supporhng (E) Industnes Space III the (10" low docsdns companion. ‘ Iiol IIOUHIIIL‘b may play thr- ltnrit-‘Vl Ioln ,_ _ III ',[).Ir-‘ ”\' ——_“- _l , ‘ A Factory In Space (é) ' lav» ‘ A's role .Figure 8 - The main screen of the segment F - "Space Industries" 24 Earth, segment C, was about differences and similarities between the planets; Space Industry , segment D, related the future occupation of space by private companies and the development of research and development of space industry; Weather on Mars, segment E, was related to weather and ambient conditions at Mars; The last segment, F, was Search for Life which explored the past, present and future expectations and theories about life on Mars. Segments B, C and E contained similar small amounts of information and few cards and buttons. Segments A, D and F represented more extensive collections of information. Three versions of each segment were created: one which was silent when buttons were clicked on; one that beeped when buttons were clicked, and one that made all sorts of different sounds, both musical and sound effects, when buttons where clicked. W In the formal experimental design used in this thesis, each subject (participant) is exposed to all 3 audio feedback conditions. Differences between individuals are thus controlled for by experimental design (Campbell and Stanley, 1963).1 Specifically, 18 subjects used all 6 segments. Thus, there were 108 observations for analysis. For each 25 subject, 2 segments were silent, 2 beeped and 2 contained all different sounds. Conditions were balanced across segments so that each segment was used in each audio feedback condition by 6 subjects. The order in which segments were used was balanced, so that each segment was used first by 3 subjects, to control for order effects. The order in which conditions were used was balanced, so that each condition was used first by 6 subjects. Table I summarizes the experimental design. Technical problems caused a minor deviation in the execution of the experimental design. The beep audio feedback condition was used 1 too many times (37), and the silent condition 1 too few times (35). Subjects were surveyed, one at a time, during the summer of 1989. All phases of the research were developed entirely at the laboratory. One computer Macintosh SE with 20 megabyte hard- drive was prepared and designated for the experiment, with all the segments, plus one special instruction for the beginning. All the movements done by the students were recorded by the computer for later compilation. We As in most experimental designs, sample size was small. The goal was examination of differences within persons, rather than generalization of differences 26 TABLE I DIAGRAM OF THE EXPERIMENTAL DESIGN Subject ID I Segment and Conditions, I in order of presentation N1 I As Bb Cd I Ds Eb Cd N2 l Ab 85 Cd I Db Es Fd N3 I Ad Bb CS I Dd Eb FS N4 I Bb Cd As I Eb Fd Ds N5 I Bd Cs Ab I Ed Fs Db N6 I 85 Cd Ab I Es Fd Db N7 I Cs Ab Ed I Fs Db Ed N8 I Ch As Ed I Fb Ds Ed N9 | Cd Ab Bs I Fd Db Es N10 I Db Ed Fs I Ab Bd Cs N11 | Dd ES Fb I Ad BS Cb N12 I DS Ed Fb I AS Bd Cb N13 I Es Fb Dd l Bs Cb Ad N14 l Eb Fs Dd I Bb Cs Ad N15 | Ed Pb 05 l Bd Cb As N16 I Fb Dd Es I Cb Ad Bs N17 I Fd Ds Eb I Cd As Bb N18 I Fs Dd Eh I Cs Ad Bb Capital letters A,B,C,D,E and F are the segments and the small letters are the types of sound environment at each segment ("5" is silent, "b" is beeping and "d" is all different sounds). 27 between persons. The sample consisted of eighteen students ( five females and 13 males) volunteering for extra credit from an introductory telecommunication undergraduate class of a Midwest university. The mean age was 22.1 years, which breaks down as follows: 19 years 20 years 21 years 22 years 36 years old old old old old 5.6‘ 5.6% 38.9‘ 44.4‘ 5.6‘ Sixteen of the 18 participants reported previous experience with microcomputer software, which breaks down as follows: daily use weekly use monthly use less than monthly never use ligament 27.83 11.1‘ 27.8% 22.23 11.1% When subjects entered the laboratory, they filled out a pretest questionnaire which asked about - experience with computers - interest in space science - opinions about desirability of missions to Mars. 28 They were given instructions at the computer explaining the basic user interface. They were then shown their first segment, and asked to notify the researcher when they were done using it. While subjects used segments, the computer was programmed to keep track of number of buttons clicked on and time spent on each screen. After using every segment, subjects filled out a questionnaire about that segment. Seven point semantic differential scales were used to measure - How easy they thought the segment was to use - How much they enjoyed the segment - How interesting they found the segment - How well designed the segment was - How professional the design was - How accurate the information was. For sound feedback conditions, subjects rated the segment sounds as irritating, creative, exciting, interesting, and long. Subjects were also asked to list, on a page with space for answers numbered from 1 to 10, what topic choices they remembered from the segment and, in a separate question, what information they remembered from the segment. Number of topics and number of facts remembered were tabulated. This process was repeated for all 6 segments. Afterwards, subjects filled out a post test 29 questionnaire which asked about: - Opinions about desirability of missions to Mars - Preference for different sounds versus silence - Preference for beep sound versus silence - Preference for different sounds versus beep feedback - Ranking of preference for learning from TV/VCR, Lecture, Magazine, Hypermedia and Book - Ranking of personal orientation to text, sound or visual - Liking of musical sound buttons - Liking of sound effect buttons. A complete copy of the questionnaire appears in the appendix. Endnorss. 1 Donald Campbell and Julian Stanley, (Rand McNally Publishing, 1979). 30 RESULTS W Varied sounds Feedback was strongly preferred by the experimental subjects. One hundred percent preferred Varied Sound Feedback to Beeping Feedback. Eighty-nine percent chose Varied Sound Feedback over Silence. Less than half (44%) preferred beeping to silence, even though varied sounds would have been better. ANOVA statistics were used to test the hypotheses. The first ANOVA table (Table 2) examines differences among segments, collapsed across feedback conditions, as a validation of the design Cross-segment comparisons show predictable differences in terms of time spent and buttons clicked on the basis of segment size, and acceptable similarities in terms of quality and enjoyment. Time spent using the segments ranged from 204 to 472 seconds. More time was spent per card in the 2 segments with more text on them. The range was 10.7 to 17.3 seconds, on the average, per card accessed. The number of buttons clicked closely paralleled the number of cards viewed, both ranging from 12-13 to 42. In terms of quantification of the number of topics and number of facts the students recalled after using each segment, variation in number of topics was small (3.4 to 4.7) and number of facts only slightly more 31 ANOVA TABLE TABLE 2 : ANOVA BY SEGMENT 'calculaied' A B-Probiems for Mars ~Weather on Mars F tor Life A -Surtace B-Problems tor - Mars E-Weather on Mars -Search tor Lite 41 1 ‘Time In seconds 'selt reported' .9 3 For easy to Use. t-very easy. 7-very difticult For Interest. t-very Interesting. 7- not Interesting For Enloyment. t-very enjoyable. 7-not enjoyable For Design Quality. t-pooriy designed. 7-weil designed For Professionalism. t-very protessionat. 7- unprofesslonai For Accuracy. 1- very accurate. 7-not very accurate 32 ANOVA TABLE TABLE 3: ANOVA BY FEEDBACK CONDITION AND HYPOTHESIS TESTING 'calculated" 'Time In seconds 'seIf-reported' Silent Bee .63 .62 For easy to Use. t-very easy. 7-very difficult For Interest. t-very interesting. 7- not Interesting For Enjoyment. I-very enjoyable. 7-not enjoyable For Design Quality. t-poorly designed. 7-well designed For Professionalism. I-very professional. 7- unprofessional For Accuracy. t- very accurate. 7-not very accurate 33 diverse (3.6 to 5.4). The segments were generally rated easy to use, with no segment rate lower than 2 out of 7 (where 1 is "very easy”). Interest level was in the middle (3.9 or 3.8) for all but one segment which was rated more interesting (2.8). Enjoyment ranged from 2.4 to 3.6. Design quality was rated no lower than 5.3 (where 7 is well designed). Ratings of professionalism ranged from 2.5 to 3.1, and accuracy from 2.8 to 3.3. The second ANOVA (Table 3) tests the hypotheses. None of the Table 3 hypothesis tests achieved statistical significance. However, consistent trends were evident. The mean ratings for time spent on segment (Hypothesis 1) were 316.2 for Silent, 334.3 for Beeping and 341.4 for Different, with a p=.82. The mean ratings for time spent on cards (Hypothesis 2) were 12.7 for Silent, 14.4 for Beeping and 14.1 for Different, with a p=.44. The trend for time variables was for silent feedback to be associated with less time spent. Hypotheses 3 and 4 looked at recall of topics and facts.The number of topics recalled showed the following means: Silent got 4.5, Beeping 3.9 and Different 4, with a p=.27. The mean ratings for the number of Facts remembered were 4.1 for Silent, 4.4 for Beeping and 4.5 for Different, with a p=.73. Thus, SILENT feedback recalled more topics and fewer facts. 34 Motivation to explore (Hypotheses 5 and 6) was assessed by counting number of buttons pressed and numbers of screens accessed. For the number of buttons pressed, the mean ratings were 27.4 for Silent, 28.3 for Beeping and 29.4 for Different, with a p=.87. For the number of cards accessed, the means were 25.6 for Silent, 25.4 for Beeping and 26.4 for Different, with a p=.95. ALL DIFFERENT feedback tended to result in more exploration. For the self—reported, credibility and enjoyment hypotheses, the results were also not significant. The mean ratings for Interest (where 1 is very interesting and 7 is not interesting) were 3.7 for Silent, 3.9 for Beeping and 3.5 for Different, with a p=.63. The mean ratings for Enjoyment were 3.3 for Silent, 3.1 for Beeping and 3 for Different, with a p=.62. The Design Quality assessment rated 5.5 for Silent, 5.6 for Beeping and 5.9 for Different ( where 7 = well designed ), with a p=.32. Accuracy had mean ratings of 3.1 for Silent, 3.1 for Beeping and 2.9 for Different, with a p=.84. Thus, DIFFERENT segments were rated slightly more interesting and well designed. W The exploratory results are divided into two tables. Table 4 is an ANOVA table that analyzes impacts of length of the segment, comparing long and short. 35 ANOVA TABLE TABLE 4: ANOVA BY SEGMENT LENGHT 'calculated' 'Time In seconds 'self report' For easy to Use. 1-very easy. 7-very difficult For Interest. t-very interesting. 7- not Interesting For Enjoyment. l-very enjoyable. 7-not enjoyable For Design Quality. t-poorly designed. 7-well designed For Professionalism. t-very professional. 7- unprofessional For Accuracy. 1- very accurate. 7-not very accurate 36 CORRELATION TABLE Table 5 - Correlation Matrix within learning preferences, Type of sound buttons and cognitive orientation Significantly more facts were recalled from the short segments. The mean for short segments was 5.1 compared to 3.7 facts recalled for the long ones. Although not significantly different, there was a trend for the short segments to be rated more easy to use, more interesting, more enjoyable, more professional and more accurate. The other results were extracted from the Correlation Matrix (Table 5), which shows relationships among learning style preferences, cognitive orientation and reactions to Music and Sound Effect buttons. * There is a strong negative correlation between text orientation and sound orientation (r=-.9). In other words, subjects who like learn from text do not like to learn by listening. * There is a strong positive correlation between sound and visual orientation (r=.8). Subjects who like images like sounds. Within Learning Media Preferences: '* There is a strong negative correlation between Hypermedia and Magazines (r=-.6). * There is a negative correlation between TV/VCR and Books (r=-.5). * There is a moderate negative correlation between Lecture and Magazine (r=.-4). Liking of sound effect buttons was only weakly correlated with liking of musical buttons (r=.2). 38 Apparently these tap two different sets of preferences. Correlations between Learning Preferences and Cognitive Orientation shows: * Moderate negative correlation between Book preference and Visual Orientation (r=-.5). In other words, subjects who like books do not like images. * Positive correlation between Book preference and Text Orientation (r=.5). In other words, subjects who like books do like text. * Negative correlation between Lecture and Texture Orientation (re—.5). In other words, subjects who like lectures do no like text. * Positive correlation between Lecture and sound orientation (r=.5). Finally, there were two interesting significant correlations involving liking of wound effect and music buttons * There is a strong positive correlation between liking to learn from Hypermedia and liking sound effect buttons (r=-.6). In other words, subjects who like Hypermedia like sound effect buttons. * There is a negative correlation between liking to learn from TV/VCR and liking music buttons (r=.5). 39 DISCUSSION Overall, the experimental style of ALL DIFFERENT sound feedback was very well received. The students strongly preferred the ALL DIFFERENT feedback condition. All 18 students chose the ALL DIFFERENT condition over BEEP condition. Eighty-nine percent chose the ALL DIFFERENT condition over the SILENT condition. Reactions were divided between SILENT condition and BEEP condition. Easy of Use and Design Quality were rated uniformly high across conditions and Enjoyment and Interest levels were uniformly moderate. There was no significant difference among conditions in terms of number of buttons clicked on, time spent exploring the stack or number of facts remembered. However, in each case, the raw data showed more favorable trends for ALL DIFFERENT than for other feedback conditions. Explorers in the ALL DIFFERENT feedback condition tended to spend more time on segments, click on more different buttons and remember more facts. One point that need to be criticized from the design of this study is that explorers appeared to feel an obligation to look at almost every screen in every segment, despite being instructed to explore only as much as they wanted to. Thus, enhanced motivation effects of ALL DIFFERENT sound feedback may have been masked. 40 From the exploratory study, the number of facts recalled from the short segments was significantly higher than the number of facts recalled from the long segments. That indicates that people tend to retain more information if it is presented in smaller quantities, somewhat like if it were small chapters. Another point that should be further investigated is the possible relation between the highly visual orientation of this study and the sound feedback. Because of the great emphasis on graphics that the "Mission to Mars" package was given, this study need to be interpreted taking this important factor in consideration. A new study that compares a more text- oriented hypermedia with the three conditions for sound feedback probably would bring new answers to design questions. Also, future studies should try to analyze in more depth the relationships of the three conditions for sound feedback for different user ages. This study was focused only on college students. It could be that teenagers and children would react even more favorable and it might be that older people would behave much more traditionally towards sound feedback. From the correlation matrix it became clear that there are two main groups of users: one that is visual/sound- oriented and other that is more text-oriented. This is an important clue for future development in Hypermedia, in a 41 sense that peeple may react distinctly to sound feedback, according to their cognitive orientation. This need to be taken into consideration when doing research or developing applications that use immediate sound feedback. 42 APPENDIX The purpose of this questionnaire Is help us to develop new Interactive computer applications. Your cooperation is critical for our research. Therefore. we would like to ask you to fill out the items below very sincerely. Your results will NOT be connected with your name or grade. Please begin by filling out this page. I. 2. 3. I0. ll. I2. What Is your age? Are you: FEMALE MALE What Is your major? How often do you use a computer? DAILY WEEKLY NONTHLY LESS NEVER How often do you use a Macintosh? DAILY WEEKLY MONTHLY LESS NEVER Have you ever heard of the Macintosh program called Hypercard? YES NO If yes, have you have ever used Hypercard ? YES NO Do you consider computers difficult or easy to use? Easy-l - 2 - 3 -4-5- 6 -7 -leflcult Do you ownacomputer? YES NO If yes, what kind? Do you subscribe Cable TV? YES NO 00 you own a VCR? YES NO 00 you have computer or video games? YES NO )3. On a scale of l to 7, where I is no Interest in Outer Space and 7 Is l5. )6. I7. l8. )9. 20. very strong Interest, how would you rate your interest In Space? VERYWEAK- l -2-3-4—5-6-7 -VERYSTRONG . What Is your Interest In Science Fiction? VERYWEAK- l -2-3-4-5-6-7 -VERYSIROIIG How would you rate your knowledge about the US Space Program? VERYWEAK- I -2-3-4-5-6- 7- VERYSTRONG What about your knowledge about Mars? VERY WEAK- I ~2-3-4-5-6-7- VERYSTRONG How strongly are you In favor of exploring outerspace? VERYWEAK- l -2-3-4-S-6-7- VERYSTRONG Ilow strongly are you In favor of humans going to Hm s? VERYWEAK- l -2-3-4-5-6-7 -VERYSTRONG How strongly are you in favor of a US - USSR joint mission to Mars? VERYWEAK- l -2-3-4-S-6-7 -VERYSTRONG How llkeky is It that life exists elsewhere In the solar system? VERY UNLIKELY - l - 2 - 3 - 4 - 5 - 6 - 7 - VERY LIKEKY 43 SEGMENT ____... l-low interested were you In the information In this segment? very Interested - I - 2 - 3 - 4 - S - 6 - 7 - not interested How easy to use was this segment? veryeasy - I - 2- 3- 4 -S- 6- 7- verydlfflcult How enjoyable was it to use? very enjoyable - I - 2- 3 - 4 -S- 6 - 7 - not enjoyabie .How Interesting was the information of this segment? very Interesting - I - 2 - 3 - 4 - S - 6 - 7 - not interesting How did you like the way the information was presented poorly designed - l - 2 - 3 - 4 - S - 6 - 7 - well designed professional - I - 2 - 3 - 4 - S - 6 - 7 - unprofessional very accurate - l - 2 - 3 - 4 - S - 6 - 7 - not very accurate How did you like the sounds in the buttons of this section? Irritating - l - 2 - 3- 4- S- 6 - 7 - pleasant creative - l - 2- 3- 4- S- 6- 7-'- repetitive exciting - I - 2 - 3- 4- S- 6 - 7 - boring Interesting - I - 2 - 3- 4- S- 6 - 7 - not Interesting toolong-I-2-3- 4- S- 6-7- tooshort tooloud-l-2-3-4- S- 6-7-toolow Sometimes the sounds were musical. Sometimes they were sound effects; In general, how much did you like the musical buttons? NOT VERYMUCH- l -2-3-4-S-6- 7 - VERYMUCH How much did you like the sound effect buttons? NOT VERYMUCH- I -2-3-4-S-6-7 -VERYMUCH if there were a knob that let you turn of f the sound, would you have used It to turn off the sound? YES NO 44 What topics choices do you remember from the menu in this segment? 2 l0 What Information or facts do you remember from this segment? _2_ IO 45 What Images do you remember from this segment? 2 End of this section 46 FINAL QUEST IONS You used segments that were silent, segments that beeped when you clicked on buttons and segments that made all different sounds when you clicked on buttons. If you could choose between silent or beeping segments, which would you prefer? SILENT BEEPING if you could choose between silent or all different sound segments, which would you prefer? ALL DIFFERENT SILENT if you could choose between beeping or all different sound segments, which would you prefer? ALL DIFFERENT BEEPING If you were to choose among different ways to access the same Information that we just presented to you, which one would you prefer? Please rank your preference, from i for your favorite to S for your least favorite: _ TV/VCR __ BOOK _ LECTURE __ MAGAZINE __ lNTERACTIVE COMPUTER (like Hypercard) How would you consider yourself: (Link your preference) Visual/Picture Oriented. Word/Text Oriented. Voice/Sound Oriented. After using this computer aplication, . How strongly are you in favor of exploring outerspace? VERYWEAK- l -2-3-4-5-6-7 -VERYSTRONG How strongly are you In favor of humans going to Mars? VERYWEAK- l -2-3-4-5-6-7 -VERYSTRONG How strongly are you In favor of a US - USSR joint mission to Mars? VERYWEAK- l -2-3-4-S-6-7 -VERYSTRONG Sometimes the sounds were musical. Sometimes they were sound elfects; in general, how much did you like the musical buttons? NOT VERYMUCH- I -2-3-4-S-6- 7 -VERYMUCH How much did you like the sound effect buttonS? NOT VERYMUCH- l -2-3-4-S-6- 7 -VERYMUCH if there were a knob that let you turn of f the sound, would you have used it to turn of f the sound? YES NO Thank you very much for your participation! 47 BI BLI OGRAPHY arsnroonnpar‘ Barton, E. A. and Dvyer, P. M., "The Effect of Audio Redundancy on Students' Ability to Profit from Printed-VerbaI/Visualized Instruction," 1onrna1_of_1nstrnctional_nsdia. Vol 14(2). 1987- Berlyne, D.E., ”Curiosity and Exploration.", science, 153, 25-33. Brand, 8., ° Penguin Books, 1987. Caupbell, '1'. C. and Stanley, J. C., W Quaei_srncr1nental.2crisne_f2r_nesearcn_ - ", Rand chally Publishing Company: Chicago, 1966. Chapanis, A., ”Studies of Interactive Communication: The Effects of Four Communication Nodes on Behavior of Teams During Cooperative Problem Solving," Hanan tasters, 14, 1972, 48?. Cohen, V. B., ”A Reexamination of Feedback in Computer- Based Instruction: Implications for Instructional Design." Educational_1echnologx. January 1985. 25-1- Lessold. A- N-. ” Earadisns_for_sonnntsr:based. ," Unpublished Article. University of Pittsburgh, 1982. Jonassen, 0.8. and Hannum, w.H., "Research-Based Principles for Designing Computer Software,” Educationa1_xecnnelogx. December. 1987- Malone. T- w.._1bat_uakes_tb1ngs_£un_to_nearnz Technical RepOrt No. C187, Palo Alto, Xerox Palo Alto Research Center, 1980. Perez, E. C. and White, H.A., ”Student Evaluation of Motivational and Learning Attributes of Microcouputers Software.”learnal_of_£onnnterznased. 1n:trngt13n,,12(1985), No.2, 39-43. Roblyer, H., ”The Greening of Educational Computing: A Proposal for a More Research-Based Approach to Computers in Instruction." Edncationa1_mechnolosxl. 'January 1985, 25-1, 40-44. Shneidernan. B-. Qss1snins_th:_user_1nterfaccv Addison Wesley Publishes, 1987. 48 Waldrop, P. B., Justen III, J. E. and Adams II, T. N., ”A Comparison of Three Types of Feedback in a Computer-Assisted Instruction Task," Educational zgghnglggy, November, 1986. walker, D.F. and Hess, R.D., "instructional_fioftxare WWW Wadsworth, 1984. Unauthored Bibliography , Apple Computer f Inc., Cupertino, California, 1988. 49 "Illiiiiiilillillilliflllifl