Use [BACK] to return to source hyperlink.
Recent research in the area of cognitive science indicates that traditional computer instruction, which requires learners to simultaneously attend to a manual, computer screen and keyboard, overloads working memory and interferes with the learning process. This paper reports on a series of studies which used a variety of computer applications in both educational and industrial settings and found that trainees following conventional computer instructional techniques could successfully work through a manual and complete the required computer based tasks yet learn very little (Chandler & Sweller, 1995; Sweller & Chandler, 1994). Alternative, cognitively guided manuals which physically integrated manual, screen and keyboard information and could be studied without the use of the computer were shown to be far superior learning tools than conventional computer manual instruction. This paper also reports on findings from ongoing research into other aspects of computer instruction including computer based training, multimedia presentations and the use of computer based animation. It will be asserted that conventional computer based training techniques may also incorporate many of the poor instructional design qualities that are a feature of conventional manual based instruction. Novel computer based training packages generated by cognitive load theory (Sweller & Chandler, 1991; 1994, Sweller, 1988; 1993), which appropriately focus attention and reduce mental load, may be superior training techniques. The same theoretical framework may also place multimedia instruction into a theoretical context and clarify how multimedia presentations should be structured. Finally, ongoing studies investigating the advantages and disadvantages of computer based animation will be reported and discussed.
How do we learn to use a new computer package?
Suppose you have just purchased a new software application and have successfully loaded it onto your computer. How will you now proceed to learn the new package? A number of options are usually available. The most traditional would be to consult the relevant manual and work through the instructions directly on the computer. This activity usually involves reading segments of text from the manual, memorising these details, while searching for the related information on the computer screen and keyboard before performing the required task. Another option would be to utilise a computer based training program that now accompany many new applications. In this case, the manual is not needed and learners only need to follow the required computer assisted instruction perhaps in the form of a tour guide, tutorial or a set of examples. Many of these packages convey instructional information in "text" boxes located at fixed positions on the computer screen. This textual information usually refers to information on the screen. The learner must then assimilate this information by matching the text with the related screen items.
Recent research in the area of cognitive science suggests that neither conventional computer manual based instruction or conventional computer based instruction, discussed above, are ideal training formats. Alternative instructional techniques based on our growing knowledge of human cognitive are now available and are demonstrating their superiority over conventional computer instruction. The same cognitive theory may also assist in the structuring of multimedia presentations and computer based animation.
Specifically, this paper has four objectives:
* To report on a group of studies which demonstrated that cognitive guided integrated training manuals were far better learning tools than conventional computer instruction (Chandler & Sweller, 1995; Sweller & Chandler, 1994);
* To discuss ongoing research comparing cognitive computer based training packages with conventional computer manual and conventional computer based instruction;
* To show how auditory/visual research (Mousavi, Low & Sweller, in press; Penney, 1989) can be applied to multimedia instruction and provide guidance into how computer based multimedia presentations should be structured.
* To reveal some of the cognitive implications of introducing animation into computer based training and to present some very recent findings demonstrating when it is useful and areas where it may be counterproductive.
Before dealing with each of these four sections, I will very briefly discuss the cognitive framework that directs this research.
Aspects of human cognition
We have known for some time that humans have a very limited processing capacity. For example, perhaps you are briefly shown a shopping list and ask to remember as many items on the list as you can. Chances are you will remember only a few items. Miller (1956) believed that our short term memory was limited to about seven chunks of information, while Simon (1974) claimed that the number of items people can process is closer to five. Cognitive scientists now favour the term working memory, a concept that emphasises an active centre where current mental activity takes place. While working memory is widely thought to be strictly limited, long term memory is generally regarded as a huge information store. Expertise in a range of areas from chess (De Groot, 1965) to algebra (Sweller & Cooper, 1985) can be explained by a superior level of domain specific knowledge in long term memory.
Information stored in long term memory is thought to be highly organised into cognitive constructs called schemas. In general terms, a schema can be viewed as a cognitive construct that categorises information in the manner with which it can be dealt. An enormous body of research in a diverse range of areas indicates that most organised knowledge is encapsulated in schemas. Besides being the fundamental building blocks of knowledge, schemas also have the function of reducing the burden on working memory. For example, we can rapidly read the text on this page as the words correspond to previously acquired schemas. We do not have to look at the detailed shape of each letter and use working memory to combine the huge variety of shapes into meaningful prose. We only need to look at some of the shapes and use our highly sophisticated schemas acquired over many years to fill in the rest. Thus, we essentially bypass working memory and making use of our long term memory. However, a child learning to read, without complex schemas, will have great difficulty processing this information through limited working memory.
A second learning mechanism that also dramatically reduces the mental load on working memory is automatic processing. Automatic processing (Shiffrin & Schneider, 1977; Schneider & Shiffrin, 1977) allows information to be processed automatically, with little or no conscious effort. Most learning tasks initially demand conscious effort with the transition from controlled to automatic processing requiring considerable time and practice. While schema acquisition is the essential component in solving similar problems and exercises to those learned, automation seems to be the key ingredient in transfer and dealing with novel problems (Cooper & Sweller, 1987; Kotovsky, Hayes & Simon, 1985). Thus, schema acquisition and automation are major factors involved in skilled performance and learning.
Cognitive load theory
While many of the aspects of the human mental processing system discussed above are largely accepted, instructions are rarely constructed with these points in mind. Training programmes (including those in computer instruction) frequently overload the working memory of learners and in doing so hinder schema acquisition and automation, the two primary ingredients of learning. Cognitive load theory (Sweller & Chandler, 1991; 1993; Sweller, 1988; 1989; 1993; 1994) have utilised the above cognitive model as well as its own notions of information complexity to develop a range of alternative cognitively guided training techniques. Cognitively guided instruction has shown to result in far more rapid learning than traditional instruction in a range of educational and industrial areas including mathematics, numerical control programming, electrical engineering, biology, physics and CAD/CAM (see Chandler & Sweller, 1991; 1992; 1995; Sweller, Chandler, Tierney & Cooper, 1990; Sweller & Chandler, 1994).
Conventional computer manual instruction
The opening to this paper described the behaviour one usually engages when learning a new computer software application. It was noted that conventional computer manual instruction, involves the learner simultaneously attending to the manual and computer screen and keyboard, a phenomena labelled as the split-attention effect (see Chandler & Sweller, 1991; 1992; Sweller et al., 1990; Sweller & Chandler, 1991 for full discussion of the effect). Sweller and Chandler (1994) claimed that the split-attention effect, the continuous process of reading segments of text from a manual, searching for corresponding screen or keyboard entities and assimilating this related information would impose a heavy mental load on working memory and consequently hinder learning. The authors developed a self-contained cognitively guided manual which physically integrated manual information with diagrammatic representations of the computer screen and keyboard. The authors then compared an integrated instructional manual group which studied their material in isolation, a conventional manual group which worked through their material on the computer and a third group which also worked on the computer but with a self-contained integrated manual. The latter two groups worked on the computer until they successfully completed all the required tasks. Results from a number of studies using a variety of software packages indicated that a cognitively guided integrated manual group, which had no contact with the computer during instruction, demonstrated superior learning, as measured by written knowledge and practical computer skills, over the other two formats.
Applying cognitive load theory
Over the last few years there has been enormous emphasis placed on computer based training. For instance, most if not all, new software applications are now accompanied with some form of computer assisted instruction (e.g., tour guides, tutorials & examples), designed to teach learners how to use the application. Furthermore, computers are increasingly becoming the chosen instructional medium for many educators to present teaching and training materials. The studies reported in Sweller and Chandler (1994) showed that a cognitively guided computer manual which physically integrated related information was a highly efficient learning tool. Could these same cognitive principles be applied to computer based training? Conventional computer based training does away with a manual, but may still involve split-attention and unnecessary searching and matching of disparate information. For example, consider the user presented with a standard computer based training program to learn a new software application. Many packages convey instructional information in "text" boxes located at fixed positions on the computer screen. This textual information usually refers to various entities on the screen. For instance, a textual commentary for a spreadsheet training package may refer the learner to the diagrammatic layout of the spreadsheet. To understand this material, learners must hold relevant text in working memory while searching for the related diagrammatic entity on the screen. The goal of understanding the application can only be achieved by performing a series of mental integrations, an activity that is likely to impose a heavy extraneous cognitive load. Thus, we are faced with a typical split-attention situation, an instructional format known to be ineffective for learning. How can we reduce extraneous load yet maintain a computer based instructional format? One alternative would be to redesign the software application so textual commentaries are physically relocated in close proximity to related screen entities, resulting in an integrated computer based training format. Cognitive load theory predicts that an integrated computer based training package would provide the same advantages as integrated hardcopy computer manuals (Sweller & Chandler, 1994) and integrated instructions in non-computing areas (Chandler & Sweller, 1991; 1992; Sweller et al., 1990). The need to mentally integrate disparate sources of information may be eliminated, mental load reduced and learning enhanced.
The hypothesised advantages of cognitively structured computer based training packages is currently under investigation. Specifically, an integrated computer based training program for a commonly used windows package has been developed and is currently being compared with conventional methods of computer based instruction.
Structuring multimedia presentations
At present, there is much debate and discussion concerning multimedia technology. Yet a comprehensive explanation of when it is useful and how it should be structured and presented is not available. Our current research is involved in testing the conditions under which multimedia presentations are beneficial and putting findings in a theoretical context.
Considerable research into a cognitive phenomena coined the modality effect suggests that if instructions are communicated in dual information modes (e.g., auditory and visual) then working memory will be expanded and performance will be enhanced (see Penney for detailed review). This view is consistent with modern conceptions of working memory (e.g., Baddeley, 1992) and is supported by a considerable body of research. For example, Allport, Antonis and Reynolds (1972) found that people were better able to carry out two tasks if the two tasks involved different modalities rather than the same modality. Frick (1984) found that more items were recalled in a memory test if some of the items were presented in a visual modality and some in an auditory modality rather than all in a single modality. Findings such as these lend credence to the multiple modality, working memory hypothesis and indeed, suggest that the effective size of working memory could be increased by using multiple rather than a single modality.
More recently, Mousavi, Low and Sweller (in press) utilised geometry instructional materials and found that a dual mode instructional format which used both auditory and visual sources of information was superior to a visual only instructional format. The authors noted that while diagrammatic information should be presented in its natural visual mode, associated text could be presented just as easily in an auditory mode as a visual mode, with a distinct advantage found for an auditory presentation of text.
This finding has obvious, direct applications for multimedia presentations. multimedia computer based presentations, where learners hear explanatory text while viewing related diagrammatic screen information, may be preferable to equivalent visual only instruction. Both instructional formats involve split-attention, but a multimedia format uses dual modalities and consequently supplies an effectively larger working memory for learners to assimilate instructional material. Conventional computer based multimedia packages are also usually equipped with a facility where textual instructions are presented simultaneously in both auditory and visual modes. With this format, the learner can hear textual information, and see identical information on the computer screen while viewing associated diagrammatic screen entities. Cognitive load theory predicts that this format would be inferior to a standard multimedia format where auditory text is not repeated visually on the screen. Repetitive visual text is not necessary for understanding and processing this information when the information is already provided in an auditory mode, imposes an extraneous cognitive load on working memory. The visual text is functionally redundant and if attended to may interfere with learning. In fact, one experiment in Mousavi et al., demonstrated that an auditory text/visual diagram format was superior to auditory text/visual diagram plus visual text.
We are currently investigating the implications of multimedia instruction on a wide scale using detailed training materials in industry. Once developed, a series of studies will compare three types of computer instruction. A conventional visual diagrams and text format will be compared with two multimedia formats, namely, a visual diagrams/auditory text format and a visual diagrams/auditory and visual text format. It is predicted, that a visual diagrams/auditory text format will be superior to both a conventional visual diagrams and text format and a visual diagrams/auditory and visual text format.
Cognitive consequences of computer animation
As animation increasingly becomes a highly prominent feature of computer based training packages, there has been much debate in the literature over the role of animation in instruction (Mayer & Anderson, 1991; 1992; Mayer & Sims, 1994; Reiber, 1990; 1991). What is needed, however, is further research examining the important factors involved in animation and the conditions under which it is useful and the situations where it may be counterproductive.
Computer based animations may be useful for a variety of reasons. For example, animations can efficiently demonstrate how a system works, such as blood flow through the human body. Basic animation such as highlighting, flashing or simple movement may also be useful as it has the potential to reduce the search process for learners. For example, if a multimedia presentation consists of a very complex or unfamiliar visual component, then an auditory commentary may be more readily assimilated if the related visual entity is moving, highlighted or under some form of animation. For example, if an auditory narration for a spreadsheet training package states that the "the sum of cell A1 and A3 is inserted in cell C2" then the appropriate highlighting of the related visual entities may considerably reduce the search, reduce the working memory burden and aid learning. If no animation is available, the auditory commentary must be held in working memory until the learner searches and matches auditory items with related visual entities.
Cognitive load theory predicts that in instructional areas where there is a high level of screen search, incorporating basic animation into a multimedia presentations will be superior to equivalent multimedia presentations without animation. Will such animations always be beneficial? It is suggested that animation will only be useful if there is a high level of screen search. In instructional areas, where there is no search or trivial search, then basic animation such as highlighting or flashing may distract or misdirect attention and interfere with learning. Specifically, if there is no screen search or trivial search, it is predicted that a multimedia presentation without animation will be superior to the equivalent presentation with animation. In fact, we have very recent preliminary evidence to support these hypotheses. Using primary school mathematical materials in a multimedia computer task, it was found that very simple animation in high search areas (i.e., corresponding angles) was notably beneficial but in areas of low search (i.e., perimeter of a rectangle), animation had negative learning consequences. This is a very significant finding that requires confirmation and closer examination with a range of instructional materials. This is currently under investigation.
Conclusion
This paper has attempted to apply the most recent findings in cognitive science directly to computer instruction. It has been suggested that most traditional computer instruction may have been designed without taking into full account the cognitive capabilities of its users and consequently may restrict learning by overloading learner's working memory. A range of cognitively generated alternatives have been developed and have demonstrated their effectiveness over conventional computer instructions. In addition, the ongoing research discussed in this paper may lead to a range of alternative computer based instructional techniques based on sound cognitive theory.
References
Allport, D., Antonis, B., & Reynolds, P. (1972). On the division of attention: A disproof of the single channel hypothesis, Quarterly Journal of Experimental Psychology, 24, 225-235.
Baddeley, A. (1992). Working memory. Science, 255, 556-559.
Bartlett, F. (1932). Remembering: A study in Experimental and Social Psychology. New York & London: Cambridge University Press.
Chandler, P., & Sweller, J. (1991). Cognitive load theory and the format of instruction. Cognition and Instruction, 8, 293-332.
Chandler, P., & Sweller, J. (1992). The spit-attention effect as a factor in the design of instruction. British Journal of Educational Psychology, 62, 233-246.
Chandler, P., & Sweller, J. (1995). Cognitive load while learning a computer program. Manuscript submitted for publication.
Kotovsky, K., Hayes, J. R. & Simon, H. A. (1985). Why are some problems hard? Evidence from Tower of Hanoi. Cognitive Psychology, 17, 248-294.
Mayer, R. E., & Anderson, R. (1991). Animation needs narrations: An experimental test of a dual-coding hypothesis. Journal of Educational Psychology, 83, 484-490.
Mayer, R. E., & Anderson, R. (1992). The instructive animation: Helping students build connections between words and pictures in multimedia learning. Journal of Educational Psychology, 84, 444-452.
Mayer, R.E., & Sims, V, K. (1994). For whom is a picture worth a thousand words? Extensions of a dual-coding theory of multimedia learning.Journal of Educational Psychology, 86, 389-401.
Miller, G.A. (1956). The magical number seven plus or minus two: Some limits on our capacity for processing information. Psychological Review, 63, 81-97.
Mousavi, S., Low, R., & Sweller, J. (In press). Reducing cognitive load by mixing auditory and visual presentation modes. Journal of Educational Psychology.
Penney, C.G. (1989). Modality effects and the structure of short-term verbal memory, Memory and Cognition, 17, 398-422.
Reiber, L.P. (1990). Using computer animated graphics in science instruction with children. Journal of Educational Psychology, 82, 135-140.
Reiber, L.P. (1991). Animation, incidental learning, and continuing motivation, Journal of Educational Psychology, 83, 318-328.
Schneider, W. & Shiffrin, R. (1977). Controlled and automatic human information processing: I. Detection, search and attention. Psychological Review, 84, 1-66.
Shiffrin, R. & Schneider, W. (1977). Controlled and automatic human information processing: II. Perceptual learning, automatic attending, and a general theory. Psychological Review, 84, 127-190.
Simon, H. (1974). How big is a chunk? Science, 183, 482-488.
Sweller, J. (1988). Cognitive load during problem solving: Effects on learning. Cognitive Science, 12, 257-285.
Sweller, J. (1989). Cognitive Technology: Some procedures for facilitating learning and problem solving in mathematics and science. Journal of Educational Psychology, 81, 457-466.
Sweller, J. (1993). Some cognitive processes and their consequences for the organisation and presentation of information. Australian Journal of Psychology, 45, 1-8.
Sweller, J. (1994). Cognitive load theory, learning difficulty and instructional design, Learning and Instruction, 4, 295-312.
Sweller, J., & Chandler, P. (1991). Evidence for cognitive load theory. Cognition and Instruction, 8, 351-362.
Sweller, J., & Chandler, P. (1994). Why is some material difficult to learn? Cognition and Instruction, 12, 185-233.
Sweller, J., & Cooper, G. A. (1985). The use of worked examples as a substitute for problem solving in learning algebra. Cognition and Instruction, 2, 59-89.
Sweller, J., Chandler, P., Tierney, P. & Cooper, M. (1990). Cognitive load as a factor in the structuring of technical material. Journal of Experimental Psychology: General, 119, 176-192.
Use [BACK] to return to source hyperlink.