IIMS 96 contents
[ IIMS 96 contents ]

Interface design considerations for educational multimedia

Geoff Ring
Edith Cowan University, Perth
While the advent of educational multimedia environments has provided richer forms of human-computer communication and interaction for learning, it has also created greater challenges for interface designers. Users of all types struggle with many of the existing interfaces for interactive multimedia (IMM) products but the problem is of particular concern with IMM products that are intended to help people learn. Any process attempting to establish guidelines for interface designers of educational multimedia must consider sociological, psychological and educational issues, particularly those relating to human cognition. This paper discusses those issues and attempts to place interface design for educational multimedia in context by discussing in sequence: (1) the broad field of human-computer interaction (HCI); (2) interface design as a sub-field of HCI; (3) interface design for interactive multimedia products and systems; and (4) interface design for educational multimedia products.

Introduction

Laurel (1990) offers a very broad definition of the user interface, seeing it as a combination of "the physical properties of the interactors, the functions to be performed, and the balance of power and control" (p. xii), with the interactors being the human user and the interactive system. The fact that digital data has no single physical form led to the desire for the software component of Laurel's user interface to be an illusion of substance and form, allowing the user to treat elements on the computer screen as if they were real objects. It is this software component (often referred to as the graphical user interface), together with supporting hardware input/output devices, which is the focus of this paper and which will be referred to variously as "the human-computer interface", "the user interface" or just "the interface".

While the advent of educational multimedia environments has provided richer forms of human-computer communication and interaction for learning, it has also created greater challenges for interface designers. Users of all types struggle with many of the existing interfaces for interactive multimedia (IMM) products but the problem is of particular concern with IMM products that are intended to help people learn. Any process attempting to establish guidelines for interface designers of educational multimedia must consider sociological, psychological and educational issues, particularly those relating to human cognition. This paper discusses those issues and attempts to place interface design for educational multimedia in context by discussing in sequence: (1) the broad field of human-computer interaction (HCI); (2) interface design as a sub-field of HCI (3) interface design for interactive Multimedia products and systems; and (4) interface design for educational multimedia products.

Human-computer interaction

The term "human-computer interaction" began to be used routinely in the 1980s. HCI is concerned with understanding, designing, developing, evaluating and implementing interactive computing products and systems for human use, and its goals relate to the useability, effectiveness and efficiency of such products and systems. It is a multi-disciplinary subject, drawing its theory and practice from more established disciplines such as computer science, psychology, cognitive science, graphic design, sociology, learning and ergonomics. Since there is no unified theory underlying HCI, the field tends to rely on the practical application of theory drawn from all those disciplines. Like architecture, HCI requires a marriage of art and science. According to Preece et al (1995), the challenge for HCI designers is "knowing how to make the transition from what can be done (that is, functionality) to how it should be done in order to match users' needs (that is, useability) in the natural work environment" (p. 43). Educational multimedia interface designers must take HCI seriously if the requirements for greater complexity in educational IMM products are to be matched by interface improvements in the key areas of visual communication, clarity and useability.

The ideal designer of an interface would have expertise in a range of areas: computer science and engineering to be able to build the interface; psychology and cognitive science to understand the capabilities and limitations of the users; sociology to understand the wider context of the interaction; ergonomics and human factors to understand the physiological demands made of the user by the hardware and software which make up the interface; graphic design to create an effective "look and feel"; and the list goes on! Clearly, no one person can have all these skills and the best approach is to use interface design teams consisting of specialists in the various areas.

HCI approaches

Four approaches to HCI have been identified by Eberts (1994): empirical, cognitive, predictive and anthropomorphic. The empirical approach is practical rather than theoretical and is based on the analysis and implementation of the results of empirical studies. It is usually need driven rather than theory driven and tends to be employed for testing existing ideas rather than for generating novel designs. The cognitive approach requires the application to practice of theories from the disciplines of cognitive science and psychology, such as analogical reasoning and attentional models. It considers the interface as an entity which presents problems which the user must solve and it focuses on the goal of ensuring that the user develops a consistent, complete and functional "mental model" of the interface.

The predictive approach aims to predict the performance of humans interacting with computers and employs various predictive modelling techniques (eg GOMS) to determine the viability and useability of an interface before prototyping commences. This approach is limited by the fact that the interpretation of such modelling techniques is somewhat subjective. The anthropomorphic approach is based on the notion that HCI is more effective if it is more human like and if it uses normal human-human communication. The full application of this approach requires further advances in technology and artificial intelligence. This approach is characterised by a strong research base and is the approach most likely to yield novel improvements in the interface. Critics of this approach perceive it to be overly dependent on technology advances and question the assumption that "natural and friendly" communication between computers and people is the best approach for most situations.

Social and organisational aspects

In recent times, more attention has been paid to social and organisational aspects which impinge on the effectiveness of HCI. According to Preece et al (1995), social aspects usually involve communication and coordination issues, while organisational aspects usually involve more global issues such as the nature of work and the role of the technology which supports work. The assumption is that if social and organisational knowledge is taken into account when designing computer systems then users will cope more successfully in work environments which use such computer systems. For example, some computer systems may need to be designed to support group work. Implementation strategies for introducing new computer based products into the workplace are more likely to be successful if the culture of an organisation (including work practices and organisational structure) and impact issues (such as the re-distribution of work and the differential effects on various workers), are taken into account.

Basic elements of HCI

The three foundation elements of HCI are the human, the computer and the interaction between them. The human can be seen as an information processor, receiving information through the senses, particularly sight, hearing and touch when working with a computer. Information needed is either stored temporarily in working memory or permanently in long term memory and may then be used in reasoning and problem solving. The computer is a very fast and reliable data processor with a variety of methods for input and output via devices such as keyboards, mice, joysticks, speech, data gloves, screens, printers and speakers. Computer memory can be likened to human working memory (RAM) and human long term memory (magnetic and optical disk media). Much of the remainder of this paper focuses on the third basic element: the interaction between humans and computers and the effectiveness of that interaction process. Interface designers should understand the strengths and limitations of both the human and the computer and how these influence the nature and style of the interaction between them.

A primary objective for interface designers, within the wider goal of developing more humane technologies, is to develop interfaces that take into account the needs and capabilities of users. Any process attempting to establish commonsense guidelines for interface designers must consider sociological, psychological and educational issues, particularly those relating to human cognition. Knowledge of human characteristics related to memory, attention, learning, perception and actions is fundamental to the interface design task.

Human-computer interface design

Human-computer interface design may be considered to be a subset of the wider field of HCI. Designers of human-computer interfaces must be able to think in terms of the eventual user's tasks and be able to translate that knowledge into a finished product. It is a widely held belief among those outside the HCI field that the task of designing an interface for users is straightforward and that no specific knowledge and skills need be learned to design an effective human-computer interface. As systems become more complex, and users more diverse, it is becoming apparent that it is not at all intuitive to design easy to use robust interfaces. Human-computer interfaces should not be added on to a product or system as an afterthought, a practice which is still all too prevalent. The development of an interface should occur in parallel with the development of the rest of the product or system, if not before.

Graphical user interfaces

The conceptual basis for most current user interfaces, which are essentially graphical in nature, was developed at Xerox's Palo Alto Centre (PARC). The major concepts which came out of the work at PARC, and which underpin the design of current interfaces, were: (1) the direct manipulation of graphics objects on the screen (with a mouse); and (2) the use of graphical metaphors of familiar real world objects. PARC researchers attempted to build interfaces that addressed the three fundamental skills Bruner (1966) assumed humans use in understanding the world around them, viz: (1) enactive skills (manipulating objects, knowing where you are in space); (2) iconic skills (visually recognising, comparing, contrasting); and (3) symbolic skills (the ability to understand long sequences of abstract reasoning).

Objects such as windows, icons, menus and buttons have become standard in today's user interfaces and they convey messages to the user about their functionality via their visual characteristics and their familiarity. The aspects that create the "look and feel" of graphical interfaces are the visual characteristics of interface elements and the images and interaction metaphors and themes used to convey meaning and function.

User understanding of interfaces

The goal of interface design is to create a user environment which facilitates the creation of a coherent mental model in the mind of the user of the functions and structure of a computer product or system. The concept of users' mental models of interfaces has attracted considerable interest from researchers. Users and designers both develop conceptual models of human-computer interfaces and systems. A primary goal in interface design is to create an interface that facilitates the mapping of the interface designers model onto the users model. A basic premise of mental model theory is that the quality of human-computer interaction with a system depends upon the functionality of the mental model the user has constructed of the system. The construction of workable mental models of human-computer interfaces assists in the tasks of recalling and interpreting information related to the interface so that they represent a promising approach for understanding HCI in general and interface design in particular.

It is useful for interface designers to know something about the cognitive styles and learning styles employed by users interacting with a human-computer interface. Determining what users know about interfaces and the nature of their understanding of an interface's functionality can assist interface designers in predicting behaviour related to likely errors, task difficulty, learning time and other variables. Further, if the processes by which users acquire knowledge and misconceptions about the interface can be ascertained, it may be possible to design interfaces which support the acquisition of more effective mental models by users. This includes the various ways both novices and more experienced users learn about interfaces, particularly complex ones.

Users who have different levels of knowledge will require different interfaces for optimal functionality. Learning about a human-computer interface can occur individually or collaboratively and generally occurs best in context. It is an active process which includes making errors and using analogies to try and interpret interface behaviour. One such approach to encourage and facilitate interface learning by novices is to restrict access to advanced features in order to reduce the chances of error.

Interface design for interactive multimedia

The emerging area of IMM has raised new issues in interface design. IMM interfaces are often aesthetically pleasing and frequently offer the user sophisticated interactions along with easy to use cross referencing and indexing. Another characteristic of many IMM products is the large amount of information accessible to the user. However, the advent of IMM has resulted in greater complexity of presentation and interaction methods, thereby increasing the number of interface decisions required of the user and the potential for user confusion. Also, IMM software typically uses a variety of media to engage the user with information via several sensory channels. The human brain has limited capabilities and can be easily overloaded with information. Hence it is important for interface designers to understand the memory and attention limitations of users and to know how to create meaningful and memorable interfaces that require minimal effort to learn and remember. Interface metaphors play an important role in achieving these goals as they combine a familiar domain with elements of the interface to make a meaningful image. The use of graphical interfaces in general substantially reduces the amount of knowledge the user has to remember.

The emergence of interactive multimedia software has made greater demands on interface designers. Gone are the days when most interfaces were based on a single screen with text and still graphics the only elements. Interactive multimedia products often demand a complex structure involving several key screens and other interactive modes such as sound and touch. The goal of an interface for an IMM product is to focus and orchestrate the interaction with the user and to make visibly apparent the organisational structure of the IMM product.

Interface media elements

Common interface media elements are: text; static graphics, such as illustrations, photographs diagrams, charts, maps, and icons; dynamic graphics (video and animations), such as simulations, cartoons, "walk throughs" and virtual reality environments; and sounds, such as music, voice and concrete sounds. Each element can have one or more roles to play; for example, screen graphics are often used to give the visual appearance of the interface a degree of coherence but may also be used to support text and other media elements as part of the information presented; sounds can complement a visual interface where attention is focused away from the screen or there is too much visual information.

Visual design skills are highly relevant to the process of interface design. They are required in routine tasks such as determining the appearance of generic objects like windows as well as more complex tasks such as an animated illustration in a simulated three-dimensional environment. Visual cues are particularly important in assisting users to understand the organisation of the interface, to show how information is related, to indicate the functionality of screen objects, to indicate what is possible in specific program modes and to clarify the navigation options available. Since reading is more demanding of human memory than perceiving, well designed interfaces incorporate elements that give clues as to their use, usually by making them appear like objects in the real world. Graphic designers are key people in this task and they use a range of devices to aid user perception such as spatial depth, perspective, tonality, etc.

Instructional message design

Instructional message design, in the context of interface design, refers to the use of patterns of media elements to modify the psychomotor, cognitive or affective behaviour of the user for the purpose of facilitating learning about, and efficient usage of, the user interface. Designers of instructional messages are primarily concerned with how message form and structure influence the users' information processing and perceptual organisation. Typical issues include the relative placement of objects on the screen and the way the eye travels over the screen in response to various techniques of visual composition.

Interface designers cannot be sure what a user will be looking at on a given screen at a given time, nor can they know how an instructional message is being perceived. Fleming and Levie (1993) point out that this is why many messages are designed for "preattentive processing", an automatic neurophysiological process which operates on the retinal image and begins to organise it prior to the user consciously focusing attention on it. Designers should be aware of what these early perceptual processes are and of the factors which influence them. Messages are also designed for conscious "attentive processing" and an understanding of the principles that apply to this process, together with message design variables related to the perception of pictures, diagrams, charts, graphs, text, sound, animations and video, can be helpful in creating effective interfaces.

Mental models

Mental models of IMM interfaces evolve in the mind of learners as they engage with an IMM product. Users' mental models of IMM products are likely to differ in themselves and also differ from the mental model possessed by the designer(s) of that interface. To be functional it is not essential that a user's mental model match that of the designer; nor do mental models need to have a one to one mapping with the actual interface design structure of an IMM product in order to be functional for a particular user.

The relationship between what the user does in an environment and the result of their actions is referred to as "mapping" by Norman (1988). Ideally, mapping occurs naturally, is simple and easy to understand. Norman makes the point that highly intuitive mappings often involve a spatial relationship - a mapping between a user's mental model and some physical object or objects in the real world. Unfortunately, mappings with abstract IMM software environments are more complex and this is where the real challenge lies for interface designers. The presentation to the learner of a good conceptual model is at the heart of the design process. It allows the operations required of them to make sense, thus increasing the likelihood of the learner developing a functional mental model of the interface.

User disorientation is a significant problem needing to be addressed with IMM interfaces. This disorientation is felt by many users when interacting with IMM software products that employ a hypermedia style design in which data is stored in networks of nodes connected by semantic and structural links. User disorientation is often the price paid for the more complex navigational methods used by IMM products which employ hypermedia approaches to the organisation of data. The role of visual cues in the interface to reveal complex organisational structure is very important. Good interface designs make use of a variety of visual cues to enhance their interactivity as well as their appearance. Program structure and information organisation can be revealed with judicious use of design principles and rules.

Interface design for educational multimedia

As indicated earlier, a primary goal of an IMM interface is that it facilitates the interaction between the user and the IMM product and that it makes visibly apparent the product's organisational structure. This is particularly true of educational multimedia, especially those designed as encyclopaedic databases, where interactive maps, overviews and "histories" provide a means for picturing (and navigating) the entire IMM information space and enable the user to quickly and easily determine where they are and where they have been. Navigational issues are currently providing interface designers of educational multimedia with one of their greatest challenges.

The learner, when interacting with an educational IMM product, is simultaneously faced with cognitive demands related to the content (what they are supposed to be learning), the structure of the abstract information space and the behaviour of various interface elements. The structure and the actions available to the learner essentially define the interface. The smaller the cognitive demands made by the interface, the greater the cognitive effort that can be directed by the learner to the content and the shorter the learning curve before the interface becomes essentially transparent to the learner.

Interface design guidelines

Most interface guidelines have been written for applications software and have little to say about the incorporation, and integration, of all the media elements used in IMM products; nor do they make reference to the special requirements of educational IMM products. Educational multimedia interface designers looking for assistance must glean and adapt from the few general interface guidelines which exist, and/or adapt principles used in traditional related fields. Most guidelines are either a set of principles or a list of detailed rules (or both). Examples of principles include ensuring ease of use, being consistent, accommodating different learning styles, reducing cognitive load, etc. Rules, on the other hand, include items such as "place page turning arrows in the bottom right hand corner of the screen", "seek confirmation before permanently saving changes to a file", "require the user to press the return key to signal the end of any text input", "use a bold sans serif font for menu entries", etc.

Guidelines usually have their origins in either psychological theory or practical experience. The application of guidelines is rarely straightforward and is more of an art than a science. Good interface design for educational multimedia results from a combination of the knowledge of the designers (instructional designers, interaction designers, graphic designers, etc) and their application of that knowledge. Common sense, experience and insight underlie many good design decisions especially when guidelines have to be traded off against each other or against other inherent constraints. Contradictions at the level of principles will often disappear when converted into design rules because of constraints imposed on the product design by the nature of the user and the way the product is to be implemented. The application of any of the published or unpublished guidelines for the design of interfaces for educational multimedia is a subjective process; none provide a recipe guaranteed to produce an effective, useable interface.

Interface design guidelines for computer systems

The dominant graphical user interfaces in use today (the Macintosh operating system and Microsoft Windows) have well documented user interface standards that should be carefully considered during the initial design phase of an educational multimedia product. One of the first design considerations in any IMM project should be to analyse the relationship between the visual and functional requirements of the intended product and the standard interface guidelines of the operating system environment (eg Macintosh or Windows) in which it will run. Compromises are sometimes required when developing IMM products for "cross platform" delivery. However, if designers of interfaces for educational multimedia take into account these system interface guidelines they will have gone a long way towards ensuring the effectiveness and transparency of their IMM product interface.

The graphical interface standards set by Apple and Microsoft represent two of the few sets of guidelines available to IMM interface designers. Apple (1992) published interface design guidelines containing eleven principles for the graphical user interface of the Apple Macintosh computer. They do not differ significantly from Microsoft's (1993) guidelines for Windows and both contain relevant information for designers of user interfaces for IMM products. The first of Apple's "principles", metaphors, takes advantage of the learner's existing knowledge of the world. In the early days of educational software it was necessary for the learner to rely on either their memory or paper documentation to determine what options were available. Nowadays it is common, indeed expected, that a lot of information is available on the screen or accessible via the interface. The memory requirements on the user are consequently less; users need merely look at the screen and access needed information via interface elements such as menus and icons. These visual aids make it much easier to recognise (rather than recall) what needs to be done. The interface designer's goal is to come up with a conceptual model based on a familiar metaphor so that the learner can easily understand what is occurring or what is required.

As for Apple's other ten interface design principles, direct manipulation refers to the immediate visibility of the effects of operations on objects; see and point refers to the sequence of selecting an object then choosing an action, consistency refers to the look and behaviour of screen elements; WYSIWYG (What You See Is What You Get) refers to printed images being identical to screen images as well as the idea of actions being immediately obvious on the screen, user control requires a balance between learner's initiating actions and being guided and protected; feedback and dialogue refers to keeping the users informed as to what is happening; forgiveness refers to actions being reversible and warnings for actions that could result in data loss; perceived stability refers to the consistent placement and actions of interface elements such as icons and menus; aesthetic integrity refers to visual appeal and modelessness refers to the general principle of avoiding a multiplicity of user modes and where these are used, giving a clear visual indicator of the current mode.

Rating interfaces for educational multimedia products

A set of ten "dimensions" for rating the quality of user interfaces for educational multimedia products was developed by Reeves and Harmon (1994). They are: ease of use which is both an aggregate and an individual dimension; navigation which refers to user orientation; cognitive load which refers to the mental demands made on the user by the interface; mapping which indicates a user's path through the software content as well as providing a boundary for the "information space"; screen design which refers to the application of generally accepted principles to do with the use of text, colour, graphics, etc; knowledge space compatibility which refers to a relationship between the user interface and a user's mental model of it; information presentation which refers to the clarity of information which is part of the interface; media integration which refers to the bringing together of the various media elements into a cohesive whole; aesthetics which refers to artistic aspects such as beauty and elegance; and overall functionality which refers to the user's perception of the utility of the software. These dimensions can be used to evaluate the user interface of an educational multimedia product. Reeves and Harmon point out that the set of dimensions is neither comprehensive nor mutually exclusive and that substantial overlap exists among them. Considerable expertise (or specific training) in educational multimedia evaluation techniques would be needed to recognise the meaning of, and understand the interpretation of, all ten specified user interface dimensions, some of which are quite complex.

According to Reeves & Harmon (1994),

... reliable and valid data concerning effectiveness, efficiency, and impact would be the ideal basis for validating interactive multimedia (IMM) but such data is rarely available. An additional approach to evaluating IMM involves the application of a set of user interface dimensions, ie, those aspects of IMM that assure that users have a meaningful and purposeful experience with IMM. User interface dimensions are necessary, but insufficient, criteria for comprehensive evaluation of IMM (p.487)

Conclusion

During the design phase there will be a number of different types of people working on the interface design of a product (programmers, interaction designers, subject matter experts, graphic designers, etc), each of whom will bring their own experiences and conceptions to the process of designing the user interface. It is important to have a plan for managing these various design inputs - and someone responsible for implementing that plan.

A useful insight into the fundamental principles underlying interface design is provided by Marcus (1992) who lists the following as "key components" of well designed interfaces:

While the above remain valid for educational multimedia products, the following additional "key components" could be added for IMM products intended to facilitate learning: This list, while not exhaustive, is intended to provide a basis for the discussion of interface design issues unique to educational multimedia. While guidelines for interactive multimedia interface design are hard to find, guidelines for educational interactive multimedia interface design are even more rare. However, it did take more than a Century after the first printed book in 1456 (Gutenburg's Bible) before features such as title pages, table of contents, indexes and page numbering became the norm in books. Presumably, the development of interfaces for IMM software, including educational multimedia, will evolve over time also. In the meantime, HCI guidelines in journals, house style guides and general handbooks usually make reference to interface design. Interface design guidelines, such as those found in Laurel (1990), Schneiderman (1992), Marcus (1992) and Tognazzini (1992), also exist in their own right. Some are based on theory, some on practical experience, and some on both. However, none of them should be rigidly adhered to and conflicts between guidelines are best resolved by taking into account the context of the interface. The implementation of guidelines is likely to be different for different products. At best, guidelines provide interface designers with a conceptual framework that can help organise their challenges in designing effective user interfaces.

This paper was not intended to offer a set of specific instructions of "what to do and when to do it" when designing a user interface for educational multimedia. Interface design for educational IMM products should aim, as far as possible, to maximise the advantages offered by IMM to education such as the encouragement of metacognitive skills, taking each learner's prior knowledge into account and the provision of a balance between exploration and guidance. With so many questions unanswered, there are virtually unlimited opportunities for researchers in this field, as indicated by Schneiderman (1992):

There are so many interesting and doable (research) projects that it may be hard to choose a direction. Each experiment has two parents: the practical problems facing designers, and the fundamental theories based on psychological principles of human behaviour... Each experiment also has three children: specific recommendations for the practical problem, refinements of your theory of human performance, and guidance to future experimenters. (p. 36)

References

Apple Computer, Inc. (1992). Macintosh human interface guidelines. Reading, MA: Addison-Wesley.

Bruner, J. (1966). Towards a theory of instruction. New York: W. W. Norton.

Eberts, R. E. (1994). User interface design. Englewood Cliffs, NJ: Prentice-Hall.

Fleming, M. & Levie, W. H. (1993). Instructional message design: Principles from the behavioural and cognitive sciences. 2nd Ed. Englewood Cliffs, New Jersey: Educational Technology Publications.

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

Microsoft Corp. (1993). Microsoft Windows 3.1. Redmond, WA: Microsoft Corporation.

Marcus, A. (1992). Graphic design for electronic documents. New York, NY: ACM Press.

Norman, D. (1988). The psychology of everyday things. New York, NY: Basic Books.

Preece, J. (1995). Human-computer interaction. Wokingham, England: Addison-Wesley.

Reeves, T. C. & Harmon, S. W. (1994). Systematic evaluation procedures for interactive multimedia for education and training. In S. Reisman (Ed), Multimedia computing: Preparing for the 21st century (p.472 -505). Harrisburg, PA: Idea Group Publishing.

Schneiderman, B. (1992). Designing the user interface: Strategies for effective human-computer interaction. 2nd ed. New York: Addison-Wesley.

Tognazzini, B. (1992). TOG on interface. Reading, MA: Addison-Wesley.

Author: Dr Geoff Ring, Associate Professor
Chairperson, Department of Multimedia Learning Technologies
Edith Cowan University, Mt Lawley Campus
2 Bradford St, Mt Lawley 6050
Telephone: 370 6369 Facsimile: 370 6780
Email address: g.ring@cowan.edu.au

Please cite as: Ring, G. (1996). Interface design considerations for educational multimedia. In C. McBeath and R. Atkinson (Eds), Proceedings of the Third International Interactive Multimedia Symposium, 355-361. Perth, Western Australia, 21-25 January. Promaco Conventions. http://www.aset.org.au/confs/iims/1996/ry/ring.html

[ IIMS 96 contents ] [ IIMS Main ] [ ASET home ]
This URL: http://www.aset.org.au/confs/iims/1996/ry/ring.html
© 1996 Promaco Conventions. Reproduced by permission. Last revision: 15 Jan 2004. Editor: Roger Atkinson
Previous URL 4 Jan 2001 to 30 Sep 2002: http://cleo.murdoch.edu.au/gen/aset/confs/iims/96/ry/ring.html