IIMS 94 contents
[ IIMS 94 contents ]

Mental models research and human computer interface design

Geoff Ring and Rod Ellis
Edith Cowan University, Western Australia
Thomas C. Reeves
The University of Georgia, USA
This paper describes human computer interface issues as they apply to interactive multimedia environments and discusses 'mental models' as a focus for research regarding user interfaces in interactive multimedia. The design of better user interfaces remains a major challenge to developers of interactive multimedia products, a challenge that may be met by mental models research. The concept of mental models is explained, measures of mental models are presented, and an on going research study at Edith Cowan University is described.


The purpose for doing multimedia research is that knowledge of how users represent systems will lead to a better understanding of what is needed to design more useable systems. In particular it is assumed that insights into user's cognitive processes when using interactive multimedia (IMM) products will result in user interface designs that are more efficient and easier to use. Current design principles are largely based on beliefs unsupported by research knowledge of the way users interact with IMM environments.

The literature about user interface design for IMM ranges from the artistic (cf, Laurel, 1990) to the technical (cf, Blattner & Dannenberg, 1992). Laurel (1990) defines user interface as a combination of "the physical properties of the interactors, the functions to be performed, and the balance of power and control" (p.xii). In IMM, the interactors are the human learner and the IMM system itself. According to Laurel, the primary concerns in user interface design are the intentions of the user. From Laurel's perspective, everything about the user interface should serve to empower the learner to accomplish the tasks that he/she has determined (eg, learn a new language, experience time travel, or find information needed to solve a problem). The user desires to focus exclusively on the learning and information seeking aspects of the experience of using IMM. Any mental effort that the user must invest in understanding the user interface of IMM is effort that cannot be focused on the content of the program.

In their more technically oriented volume, Blattner and Dannenberg (1992) refer to interface design in terms of media (eg, text, audio, and video) and modes of interacting with media (eg, pointing, speaking, and keying). With respect to media, Blattner and Dannenberg "are most concerned with how the information content of a media is conveyed" (p.xxv). With respect to modes, they "are usually more concerned with the dynamics of interaction" (p. xxv). Their focus on interface design is first, making the modes of interaction as obvious as possible so that they even become "transparent" or "second nature" to the user, and second, making the media as appropriate as possible to the nature of the content.

The user friendliness of software has also become of increasing concern to designers of interactive multimedia (IMM). This concern is well placed. Any confusion involved in using programs designed for education and training is especially detrimental. After all, IMM programs are designed to help people learn, not bewilder them. Learners often end up lacking the clear understanding of the structure and functions of IMM intended by designers.

This problem can be especially acute in multimedia and hypertext environments that allow learners to explore the content and activities in the program according to their own needs and interests (Heller, 1990; Jonassen & Grabinger, 1990). According to Shneiderman and Kearsley (1989), hypertext "is a database that has active cross references and allows the reader to 'jump' to other parts of the database as desired" (p. 3). Hypertext, hypermedia, and multimedia are all being designed and used for instructional purposes; in each case, the materials allow learners to navigate through the contents and activities via largely unrestrained exploration. However, there is a price for the freedom these types of systems permit. Users of hypertext materials may become confused and quickly lose track of what is going on, what they can do, and/or where they are located in the program. The resulting frustration prohibits learners from taking full advantage of the learning opportunities presented by these systems.

Human computer interface issues

1. Interactivity

The major advantage of interactive learning materials over other instructional media lies in the unique feature of "interactivity". To be interactive, the exchange of information, responses, and feedback between learner and computer should be as individualised, adaptive, and personal as possible (Price, 1991). Ideal interactivity would approximate the exchanges that occur between a human tutor and a student. Maximising the effectiveness and efficiency of learner computer communication, ie, interactivity, leads to learner satisfaction and to enhanced performance and productivity. Although interactivity is perhaps the most important feature of IMM, the nature of the interactions between humans and IMM is a neglected area of study.

2. Human factors research

The study of human computer interaction is relatively new. Three forces promote this research: a heterogeneous population of computer users including novice and non-technically trained people; increasing organisational dependence on interactive systems; and high performance requirements in life critical applications in which error rate must be kept extremely low (Shneiderman, 1982). A machine oriented perspective has been replaced by a human oriented perspective as designers attempt to accommodate the unique skills, experiences, and expectations of individual users. The goal of human factors research in IMM is simple, viz, the more we understand human computer interaction, the better we can design effective IMM.

Many factors influence interactive behaviours, such as nature of computer interfaces and the individual differences learners bring to the human computer interaction process (Canter, Rivers, & Storrs, 1985). Different understandings of interfaces are also likely to influence user's navigational patterns (McAleese, 1989). Navigational patterns are most likely to vary in IMM programs designed to allow a learner to explore content and activities at his/her own volition. Although it seems to make sense that learners' understanding of the structure and functions of IMM will have an impact on their navigational behaviours as well as learning performance, there is little research to support this assertion. Insufficient or ineffective interaction with a program leads to confusion, getting lost, and/or partial use, non-use, or misuse of the information provided by the IMM, getting bored, and/or frustration. These and other problems result in less effective learning performance.

On the other hand, it may be that disorientation in IMM is not necessarily bad. Kahn and Landow (1993) present an analysis that indicates that some types of disorientation may be appropriate, even pleasurable, especially when the learner is engaged in dealing with complex, novel material. However, it is clear that stressful, unproductive disorientation results from a poorly designed interface. Instead of thinking that learning outcomes are an exclusive function of IMM or of the learners, designers should recognise that there are multiple contributors to learning outcomes: the learners themselves, instructional content and activities, and the interface between the learner and program.

3. Individual differences

All learning is unique. Each learner must perceive, respond, process, store, and retrieve his or her own information, attitude, and motor routines. The impact of differences in student characteristics on learning is one of the major themes in the history of educational research (Peterson, 1982). Learning is a function of the learner, the content to be learned, and the features of the instruction (Gardner, 1983; Sternberg, 1985). The learners brings their unique concepts, perspectives, experiences, cognitive patterns, characteristics, preferences, and expectations, namely "individual differences", to the learning environment. Learning through any instructional medium requires the activation of cognitive processes that are specific to the content, treatment, and medium of instruction. If the cognitive processes called for by the content are not present, some sort of instructional support is required.

The cognitive demands of human learning are complex and involve various variables which interact with learner characteristics such as aptitude, age, gender, ability, experiences, and cognitive styles (Ackerman, Sternberg, & Glaser, 1989). Learning is influenced by three dimensions of individual differences factors: (1) personological factors such as aptitude, ability, skill, prior knowledge, educational level, experiences, and learning styles; (2) affective factors such as, value, self efficacy, motivation, attitude, and anxiety; and (3) physiological factors such as eye hand coordination and visual acuity.

4. Cognitive load

Working with IMM requires different mental efforts than performing learning tasks via print or other formats, eg, watching linear video. In order to make any meaningful response to IMM, learners must cope with and integrate three cognitive loads or demands, ie, (a) the content of the information, (b) the structure of the program, and (c) the response strategies available. These threefold demands imply that there are physical, perceptual, and conceptual contacts happening during user IMM interactions (Jih, 1991).

Learners acquire knowledge about screen displays and figure out how to use IMM through unique ways. Learners learn how to use programs by forming mental models consisting of a few uncomplicated, low level heuristics and by making inferences based on mismatches between their expectations and actual system reactions (Lewis, 1986). A shift away from mere concern with specification of mechanical aspects of interfaces and toward developing a better picture of how users understand the workings of computer programs is a new direction in the field of human computer interaction (Ackermann & Tauber, 1990; Kerr, 1987).

The limited capacity of working memory to hold only 5 to 9 chunks of information simultaneously (Miller, 1956) makes it difficult for users of complex structured IMM to reason when numerous cognitive load factors must be handled at once. Users feel overwhelmed by abundant menus, option commands, icons, windows, functions, sequence controls, etc (Jih, 1991). This complexity increases the cognitive load. It also increases the risks of confusion, especially when users confront multimedia materials which intentionally include many interactive options. The possibility of user disorientation is a major concern in hypertext materials that feature a network like, more flexible and complex structure (Edwards & Hardman, 1989; Kulin & Streitz, 1989; Tsai, 1988/89).

Mental models

"Mental model" theory is an attempt to model and explain human understanding of objects and phenomenon (Ackermann & Tauber, 1990; Gentner & Stevens, 1983; Johnson-Laird, 1983; Rogers, Rutherford & Bibby, 1992). The construction of a plausible mental model "eases the interpretation and memorability of information" (Gardiner & Christie, 1987). Given that our understanding of human perception should play a crucial role in the design of interfaces, research on mental models is a promising approach for analysing the human computer interaction and improving interface designs.

Users seek new information in ways that depend on and are limited by their current mental model and learning goals (Tweney, 1987). The existence and value of the mental models concept (Moray, 1987; Rasmussen, 1990) can be summarised by the premise that the quality of interaction within system operation depends upon the functionality of the mental models users have of the system. Mental models are the vehicle for understanding overall system configurations, their elements, and functional interrelationships.

The risk of learner disorientation in IMM is decreased by the quality of the mental model of the interactive program possessed by the learner. Jih (1991) found that students with a more complete mental model of the graphical user interface (GUI) of the Apple Macintosh computer were more effective users of an IMM program that adopted elements of the Macintosh GUI. Research on mental models can identify salient characteristics of cognitive processes engaged in human computer interaction and help in the development of research based guidelines for the design of effective IMM. There is a need to identify the user's understanding about interfaces in term of mental models to provide principles and guidelines for design of IMM.

A mental model is a model evolving in the mind of a user as the user is learning and interacting with a computer system (Gentner & Stevens, 1983; Norman, 1983). The mental model represents the structure and internal relationships of a system, although in a simpler form. The user's mental model is the source of the user's expectations about the effects of actions, can guide navigation or planning of actions, and contributes in interpretation of feedback (Van Der Veer, 1989). It seems to make sense that an ideal mental model would be consistent with the conceptual model of the interface developed by designers, but this has not been demonstrated consistently. However, according to Norman (1983, 1986), strong or accurate mental models show a functional or spatial similarity to the system or to the image the system presents to the users.

Support for the existence and value of mental models in facilitating learning performance, retention of procedures, and the invention of new operational procedures can be found throughout the research literature (cf, Ackermann & Tauber, 1990; Jagcinski & Miller, 1978; Kieras & Bovair, 1984; Mayer, 1989; Moray, 1987). Mayer (1981) concluded that users systematically develop a cognitive structure or mental model for any task environment they engage. Individual differences relevant to the way novice users acquire a mental model are prior knowledge, information processing style, and general intelligence (Van Der Veer & Felt, 1988). Human factors research, indicates that mental models can help in users' interpretation of cues from the system, in working out appropriate actions, in detecting and repairing errors, and in learning (Green, 1990). Despite these claims, it is safe to say that the research on users' mental models is not yet well developed, especially with respect to the design and implementation of IMM. The lack of a strong theoretical basis for mental models is a continuing problem (Fetzer, 1993; Johnson-Laird & Byrne, 1991).

Applying mental models research to IMM

The following key points have been derived from the literature reviewed and ideas discussed above: (1) There is a need for systematic studies regarding human understanding (mental models) of interfaces in order to understand the processes involved in human interactions with IMM; (2) Individual differences of learners (eg, learning styles and previous experience) will influence the construct of their mental models; (3) Previous research studies of mental models have concentrated on information retrieval systems or application/tool programs rather than on IMM; and (4) There has been little research focused on the relationship among learners' individual differences, their mental models, their navigational pathways in IMM, and their learning performance. Hence, there is a need for exploratory research concerning the relationship among learners' individual differences, their mental models of interface features, their navigational paths, and their performance in IMM for education.

In the absence of strong theoretical support for design of IMM interfaces, two types of exploratory studies appear to be needed. First, qualitative analyses based on learners' descriptions of the interface and their explanations of their specific navigational choices can be used to categorise the properties of their mental models of IMM. For example, in an effort to understand what motivates exploration in IMM, Harmon (1992) asked users of a complex IMM program (ABC News Interactive Corporation, 1991) to "think aloud" during their exploration of the program. It is also appropriate to investigate the mental models of the designers of IMM programs. Jones (1993) interviewed designers of IMM programs to reveal the sources of the decisions they made in designing the user interface for their programs.

Second, if sufficient sample sizes can be reached, multiple regression methods can be used for gaining insights into the relationship among the learners' individual differences, their mental models of the structure of interface features in IMM, their actual navigation activities during use of IMM, and the results of that use.

There are at least four major questions which should be addressed in this line of research:

  1. What are the properties of the mental models that learners form of IMM interfaces and how frequently do they occur?

  2. What individual differences in selected learning styles and previous computer experience are related to differences in the mental models learners form?

  3. How are differences in learning styles, previous computer experience, and mental models related to the navigational pathways of learners in IMM?

  4. How are the individual differences, mental models, and/or navigational pathways of learners related to learning performance?

Current research

Several preliminary studies of this nature are underway at The University of Georgia in the USA and Edith Cowan University in Australia as part of a cooperative research program between the two institutions. Researchers at The University of Georgia are studying the mental models secondary students construct of two large scale IMM program created by the IBM Corporation (1991a, b), viz, Columbus: Encounter, Discovery and Beyond and Illuminated Books and Manuscripts. At Edith Cowan University, a pilot study has been initiated to investigate the mental models of pre-primary and primary children in reaction to an IMM program called Just Grandma and Me (Broderbund, 199 1).

Pilot study at Edith Cowan University

The pilot study conducted at Edith Cowan University was qualitative in nature and constituted phase 1 of ECU's contribution to a collaborative research effort between the University of Georgia and Edith Cowan University. Staff and students at both institutions are continuing to conduct research into human computer interface issues in IMM environments and exploring the potential application of mental model theory in this area.

In the pilot study one class of pre-primary students (n = 25) and one split class of grade 1 students (n = 11) and grade 2 students (n= 17) were given access to the computer based version of the children's book Just Grandma and Me written by Mercer Mayer (1983). This program was chosen because of its multimedia nature and easy navigation interface. In each of the two classrooms, one computer was permanently ad up to run the program over a 3 week trial period. In both classrooms the teacher introduced the story by reading the book version and then rostered the children to have turns with the computer based version. In the pre-primary centre each child was assisted by a parent helper and in the split grade 112 classroom the children usually worked in pairs. When needed, a peer tutor (the class 'expert') was available to assist. All children had ample opportunity to use the mouse (the keyboard is not required for this program) as they interacted with the story. All children had at least 45 minutes exposure to the program.

After the three week trial period each child was invited to demonstrate various aspects of the program and to answer a list of 18 set questions posed by a research assistant (see Appendix A). This 'demonstration and interview' session took approximately 25 minutes per student. All sessions were videotaped for subsequent analysis.

The interview questions posed were designed to explore the mental model that each child had formulated about how to use the program and what the program meant to them. The interview process is described in Appendix A. Some questions related to another electronic book ("Cinderella") in order to measure the degree to which various aspects of the electronic book experience with Just Grandma and Me transferred to a new but similar situation.

Preliminary findings

Some of the preliminary findings from the above interview are described below :
  1. The older children were more capable at demonstrating and better able to verbalise open ended questions. They indicated more lateral thinking, deeper meaning to their answers and more creativity in their responses. However, language development level clearly affected the answers given and it is difficult to see to what extent any mental model may have been independent of this overriding issue. A clear implication here is the inherent difficulty of assessing the mental models of young children who naturally have a limited command of language. Methods need to he developed for mental models research with young children that minimise this language development variable.

  2. The children were all very capable of initiating the starting procedure, selecting and clicking on objects on the screen to demonstrate what then happens and moving from one screen to the next (using the 'page turning' arrow icon). Alternatively, few children knew how to stop using the program (a 3 step process) and it became clear that in general class use the program was rarely terminated by the children.

  3. When asked what they liked about Just Grandma and Me, a wide variety of answers were given but the most preferred general response was 'clicking on things to make them move'. The pre-primary children predominantly (18 of 25) described a particular event rather than a general feature. Of the grade 2 children only 6 of the 17 described a particular event while the rest described general features. When asked what they didn't like about Just Grandma and Me, 87% of the children could find nothing to mention.

  4. The ability of the children to describe the electronic version (as if they were telling a family member or a friend) was poor. Even the grade 2 children, on average, were rated at less than two on a five point scale in terms of descriptive ability. It seems that the novel nature of electronic children's books severely tested the conceptual framework from which these young children could effectively define what they were using.

  5. When asked if 'Little Critter' (the grandson of Grandma) was real, only one child (pre-primary) answered "yes'.

  6. When asked why Grandma or Little Critter moved and talked the most common response included the word 'computer' and implied that 'the computer did it'.

  7. Over 90 % of the total sample preferred the electronic version over the traditional book, with no significant age difference evident in the preferences shown.

  8. When shown a scene (page) from an electronic book version of the Cinderella story the children showed varying ability to transfer navigational concepts from their experience with the Just Grandma and Me interface. The older children were more able to detect that the turned page corners were the metaphor for turning pages (pre-primary, 44 % : grade 1, 91 % : grade 2, 94 %). The older children were also better able to identify the loudspeaker icon next to the text block as the 'switch' necessary to have the text read aloud (pre-primary, 48 % : grade 1, 90 % and grade 2, 94 %). Maturation was again evident when the children were asked to indicated what they might like to click on in the Cinderella scene and what they thought might then happen. The diversity and creativity of the responses was more evident in the older grades.

  9. In a page sorting exercise, after the interview, each child was required to sort 12 cardboard backed pages (copied from the book which closely resembles the screen images used in the electronic version) into sequence from page 1 to page 12. The children were able to sort pages 1, 2 and 12 very successfully (close to 100% success in each grade). However, the pages in the middle of the story were sorted with only low to moderate success. It seems likely that the lack of natural story linkage between the pages of the Just Grandma & Me program and the distracting effects of the interactions which were peripheral to the main storyline had more to do with this than any conceptual understanding of story sequence by the children.

  10. Analysis of sex differences has not been concluded but cursory examination of the data shows little difference between male and female responses in general.

Suggested research methodologies

Investigations of mental models should include both observational and regression methods. These methods are recommended since no manipulation of a treatment variable is appropriate (Tuckman, 1988) and because of the exploratory nature of the research. Multiple regression methods can be used to explore relationships among specified variables and to determine the extent to which criterion variables can be predicted. The primary criterion variable should be learning performance with respect to the IMM content. The three primary predictor variables should be individual differences, mental models of learners, and their navigational pathways.

There are several methodological conditions that must be met for these studies. First, the learners should be involved in purposeful learning, driven by either intrinsic or extrinsic motivation. Volunteer subjects using IMM not directly related to their education needs are inappropriate. Second, the learners should spend many hours rather than a few minutes interacting with the IMM. Third, ideally, the population of learners should be diverse in age, education, and prior experiences with computers. Fourth, if multiple regression methods are to be used, sample sizes of learners must be large, ideally running into the hundreds, to meet the requirements of these analytical techniques.

IMM candidates for this research include interactive CD-ROM multimedia programs produced by companies such as ABC News Interactive, Apple Computer, Inc, Broderbund Software, the IBM Corporation, and Voyager Press. Many of these programs feature a windows, icon, mouse, and pull down menus (WIMP) interface. Users view a screen and respond to the programs via mouse or keyboard input. These programs feature an elaborate array of multimedia options including audio, video, graphics, animation, and text.

Specific procedures appropriate to studies of mental models and their interrelationships with individual differences, navigational paths, and performance in IMM are:

  1. Selected individual differences should be measured before learners interact with the specified IMM. Questionnaires can be used to measure demographic variables such as sex, age, and education as well as to obtain data concerning the learners' previous experience with computers. Validated tests such as the Group Embedded Figures Test can be used to assess selected learning styles such as field dependence/field independence.

  2. Learners should be allowed to interact with the specified IMM program for a significant period of time, ideally five to ten hours spread over several sessions. A tracking program should be used to collect data regarding learners paths through the program, responses to queries, and to time stamp their interactions with various meaningful sub-units of the program.

  3. At predetermined points in the Program, the learners' mental models should be assessed by a teach back approach and by "pop up" questions. The teach back techniques (Van Der Veer, 1989) requires learners to explain how a program works to a hypothetical "new" learner. Non-directive questions encourage learners to explain the structure of the program and to identify the functions of its icons, menus, etc. Pop up questions (Hassett, 1990; Koubeck, Salvendy, & Eberts, 1987; Norman, 1983) can be programmed to appear on the screen in response to certain activities initiated by the learners. These questions ask learners to provide reasons for their actions, such as why he/she did not complete certain activities.

  4. Performance can be measured using a variety of online performance tests, performance exercises, and interviews. Triangulation of learner performance is recommended over reliance on any one outcome measure.

  5. Multiple regression procedures can be used to examine the interrelationships among variables and reveal the predictive power of the individual differences, mental models, and navigational paths in respect to learner performance.

Importance of mental models research

The need for research of this kind has never been more important (Jih & Reeves, 1992; Reeves, 1993). The power and complexity of IMM programs for education and training are increasing dramatically. If multimedia systems are going to live up to the great promises that are being made for them, our knowledge of design factors regarding the user interface of these systems must be strengthened. At the present time, what we know about interface design is much more of an art than a science (Laurel, 1990; Mayhew, 1992; Shneiderman, 1987). Research studies being conducted in the field of systems engineering and applications development indicate that programmatic inquiry regarding the mental models users form of the systems with which they interact is viable and important (Ackermann & Tauber, 1990). However, few such studies have been done with IMM programs expressly designed for education. This paper calls for programmatic research in this area and provides initial guidelines for the conduct of such research. The undertaking of this research program is much too large for one or two institutions to undertake, and we strongly encourage additional partnerships and/or consortia in the conduct of this line of inquiry.

References

ABC News Interactive Corporation (1991). In the Holy Land: A history of hatred in the Middle East [Interactive multimedia program]. New York: ABC News Interactive.

Ackerman, P. L., Sternberg, R. L, & Glaser, R. (Eds). (1989). Learning and individual differences: Advances in theory and research. New York, W. H. Freeman.

Ackermann, D. & Tauber, M. J. (Ed). (1990). Mental models and human-computer interaction 1. The Netherlands: Elsevier Science.

Blattner, M. M. & Dannenberg, R. B. (Eds) (1992). Multimedia interface design. New York: ACM Press.

Broderbund Software. (1991). Just Grandma and Me [Interactive multimedia program].

Canter, D., Rivers, R. & Storrs, G. (1985). Characterising user navigation through complex data structures. Behaviour and Information Technology, 4(2), 93-102.

Edwards, D. M. & Hardman, L. (1989). Lost in hyperspace: Cognitive mapping and navigation in a hypertext environment. In R. McAleese (Ed), Hypertext: Theory into practice. Norwood, NJ: Ablex Publishing Corporation.

Fetzer, J. H. (1993). The argument for mental models is unsound. Behavioral and Brain Sciences, 16(2), 347-348.

Gardiner, M. M. & Christie, B. (Eds). (1987). Applying cognitive psychology to user interface design. New York, NY: John Wiley & Sons Ltd.

Gardner, H. (1983). Frames of mind: The theory of multiple intelligences. New York: Basic Books.

Gentner, D. & Stevens, A. L. (1983). Mental models. Hilldale, NJ: Lawrence Erlbaum Associates Inc.

Green, T. R. G. (1990). Limited theories as a framework for human computer interaction. In D. Ackermann & M. J. Tauber (Eds), Mental models and human-computer interaction 1. Human Factors in Information Technology No. 3. Amsterdam: Elsevier Science Publishers B. V.

Harmon, S. W. (1992). On the nature of exploratory behavior in hypermedia environments: Considerations of learner use patterns of hypermedia environments for the design of hypermedia instructional systems. Unpublished doctoral dissertation, The University of Georgia.

Hassett, M. R. (1990). Adding research tools to computer based video instruction programs. In D. W. Dalton (Ed), Proceedings of the 32nd Annual International Conference of the Association for the Development of Computer Based Instructional Systems. San Diego, CA, October 29 - November 1, 1990.

Heller, R. S. (1990). The role of hypermedia in education: A look at the research issues. Journal of Research on Computing in Education, 22(4), 431-441.

International Business Machines Corporation. (1991a). Columbus: Encounter, discovery and beyond [Interactive multimedia program]. Atlanta, GA: IBM Corporation.

International Business Machines Corporation. (1991b). Illuminated books and manuscripts [Interactive multimedia program]. Atlanta, GA: IBM Corporation.

Jagcinski, R. L, & Miller, R. A. (1978). Describing the human operator's internal model of a dynamic system. Human Factors, 20(4), 425-433.

Jih, H. J. (1991). The relationship among the structure of interfaces, users' mental models, and performance in computer based interactive courseware. Unpublished doctoral dissertation, The University of Georgia.

Jih, H. J. & Reeves, T. C. (1992). Mental models: A research focus for interactive learning systems. Educational Technology Research and Development, 40(3), 39-53.

Johnson-Laird, P. N. (1983). Mental models: Toward a cognitive science of language, influence, and consciousness. Cambridge: Cambridge University Press.

Johnson-Laird, P. N. & Byrne, R. M. (1991). Deduction. Hillsdale, NJ: Lawrence Erlbaum.

Jonassen, D. H. & Grabinger, S. R. (1990). Problems and issues in designing hypertext/hypermedia for learning. In D. H. Jonassen & H. Mandl (Eds), Designing hypermedia for learning, 3-26. Berlin: Springer-Verlag.

Jones, M. G. (1993). Guidelines for screen design and interface design in computer based learning environments. Unpublished doctoral dissertation, The University of Georgia.

Kahn, P. & Landow, G. P. (1993). The pleasures of possibility: What is disorientation in hypertext? Journal of Computing in Higher Education, 4(2), 57-78.

Kerr, S. T. (1987). Finding one's way in electronic space: The relative importance of navigational cues and mental models. Paper presented at the Annual Convention of the Association for Educational Communications and Technology. Atlanta, GA, February 26 - March 1, 1987.

Kieras, D. E. & Bovair, S. (1984). The role of a mental model in learning to operate a device. Cognitive Science, 8(3), 255-273.

Koubeck, R. L, Salvendy, G. & Eberts, R. (1987). Eliciting knowledge for software development. Behaviour and Information Technology, 6(4), 427-440.

Kuhn, E. & Streitz, N. A. (1989). Do users know how to act when menus are complex? A numerical model of user interface complexity. In F. Klix et al (Eds), Man-computer interaction research: MACINTER-II. North-Holland: Elsevier Science Publishers B. V.

Laurel, B. (Ed) (1990). The art of human-computer interface design. Reading, MA: Addison-Wesley.

Lewis, C. (1986b). A model of mental model construction. In M. Mantei and P. Orbeton (Eds), Human factors in computing systems: CHI '86 Proceedings. New York: Association for Computing Machinery.

Mayer, M. (1983). Just Grandma and Me. Western Publishing, New York.

Mayer, R. E. (1989). Models for understanding. Review of Educational Research, 59(1), 43-46.

Mayer, R. E. (1981). The psychology of how novices learn computer programming. ACM Computing Surveys, 13, 121-141.

Mayhew, D. J. (1992). Principles and guidelines in software user interface design. Englewood Cliffs, NJ: Prentice- Hall.

McAleese, R. (1989). Navigation and browsing in hypertext. In R. McAleese (Ed.), Hypertext: Theory into practice. Norwood, NJ: Ablex Publishing Corporation.

Miller, G. A. (1956). The magical number seven, plus or minus two: Some limits on our capability for processing information. Psychological Review, 63, 81-97.

Moray, N. (1987). Intelligent aids, mental models, and the theory of machines. International Journal of Man-Machine Studies, 27, 619-629.

Norman, D. A. (1986). Cognitive engineering. In D. A. Norman & S. W. Draper (Eds), User centered system design: New perspectives on human-computer interaction. Hillsdale, NJ: Lawrence Erlbaum Associates, Inc.

Norman, D. A. (1983). Some observations on mental models. In D. Gentner & A. L. Stevens (Eds), Mental models. Hillsdale, NJ: Lawrence Erlbaum Associates Inc.

Peterson, P. L. (1982). Individual differences. In H. E. Mitzel (Eds), Encyclopedia of educational research, 844-851. New York: The Free Press.

Price, R. V. (1991). Computer aided instruction: A guide for authors. Belmont, CA: Wadsworth, Inc.

Rasmussen, J. (1990). Mental models and the control of action in complex environments. In D. Ackermann & M. J. Tauber (Eds), Mental models and human computer interaction 1. Human Factors in Information Technology No. 3. Amsterdam: Elsevier Science Publishers B. V.

Reeves, T. C. (1993). Pseudoscience in computer based instruction: The case of learner control research. Journal of Computer Based Instruction, 20(2), 39-46.

Rogers, Y., Rutherford, A. & Bibby, P. A. (Eds) (1992). Models in the mind: Theory, perspective and application. San Diego, CA: Academic Press.

Shneiderman, B. (1987). Designing the user interface: Strategies for effective human- computer interaction. Reading: Addison-Wesley.

Shneiderman, B. (1982). The future of interactive systems and the emergence of direct manipulation. Behavior and Information Technology, 3, 237-256.

Shneiderman, B. & Kearsley, G. (1989). Hypertext hands on: An introduction to a new way of organising and accessing information. Reading, MA: Addison-Wesley Publishing Company, Inc.

Sternberg, R. J. (1985). Beyond IQ: A triarchic theory of human intelligence. Cambridge, MA: Cambridge University Press.

Tsai, C. J. (1988/1989). Hypertext: Technology, applications, and research issues. Journal of Educational Technology Systems, 17(1), 3-14.

Tuckman, B. W. (1988). Conducting educational research (3rd ed). Orlando, FL: Harcourt Brace Jovanovich, Inc.

Tweney, R. D. (1987). What is scientific thinking? Paper presented at the Annual Meeting of the American Educational Research Association. Washington DC, April 23-25, 1987.

Van der Veer, G. C. (1989). Individual differences and the user interface. Ergonomics, 32(11), 1431-1449.

Van der Veer, G. C. & Felt, A. M. (1988). Development of mental models of an office system: A field study on an introductory course. In G. C. van der Veer & G. Mulder (Eds), Human-computer interaction: Psychonomic aspects. Berlin Heidelberg, Germany: Springer-Verlag.

Appendix A

Interview Part 1: Ability to use and personal concept of the Just Grandma and Me program
  1. "Show me how you start using Just Grandma and Me? "
  2. "Show me what can you do with Just Grandma and Me?" (several minutes allowed)
  3. "Show me how you move to another part of Just Grandma and Me?"
  4. "Show me how you stop using Just Grandma and Me?"
  5. "What else have you used that is like Just Grandma and Me?"
  6. "What do you like about Just Grandma and Me?"
  7. "What don't you like about Just Grandma and Me?"
  8. "Have you told anyone at home about Just Grandma and Me? If the answer was YES : "What did you say?" OR If the answer was NO "What would you say?"
Interview Part II: Comparing the book with the electronic version
  1. "Why do Grandma and Little Critter move on the screen but not in the book?"
  2. "Why do Grandma and Little Critter have voices here (point the screen) but not in the book?"
  3. "Do you think Little Critter is real?"
  4. "What can you do with this (point to screen) that you can't do with the book?"
  5. "What can you do with the book that you can't do with this (point to screen)?"
  6. "Which do you like best, the book or this (point to screen)? Why?"
Interview Part III: Relating to a single page scene from the "Cinderella" electronic book
    "Here is something else like Just Grandma and Me." (let them absorb the scene)

  1. "What would you do to see what happened after this (point to the screen)?"
  2. "What would you do to see what happened before this (point to the screen)?" (repeat questions 1 and 2 as many times as needed)
  3. "What would you do to hear the words?"
  4. (a) "What would you like to try if I gave you the mouse?"
  5. (b) "What do you think would happen if you did that?"

Authors: Geoff Ring, Chairperson, Department of Computer Education, Edith Cowan University, Mount Lawley Campus, 2 Bradford Street, Mount Lawley WA 6050. Tel. + 61 8 9370 6369 Fax. +61 8 9370 2910. Email: g.ring@cowan.edu.au
Rod Ellis, Edith Cowan University
Dr Thomas C. Reeves, The University of Georgia

Please cite as: Ring, G., Ellis, R. and Reeves, T. C. (1994). Mental models research and human computer interface design. In C. McBeath and R. Atkinson (Eds), Proceedings of the Second International Interactive Multimedia Symposium, 485-493. Perth, Western Australia, 23-28 January. Promaco Conventions. http://www.aset.org.au/confs/iims/1994/qz/ring2.html


[ IIMS 94 contents ] [ IIMS Main ] [ ASET home ]
This URL: http://www.aset.org.au/confs/iims/1994/qz/ring2.html
© 1994 Promaco Conventions. Reproduced by permission. Last revision: 15 Feb 2004. Editor: Roger Atkinson
Previous URL 29 Sep 2000 to 30 Sep 2002: http://cleo.murdoch.edu.au/gen/aset/confs/iims/94/qz/ring2.html