Cognitive load theory has been designed to provide guidelines intended to assist in the presentation of information in a manner that encourages learner activities that optimize intellectual performance. The theory assumes a limited capacity working memory that includes partially independent subcomponents to deal with auditory/verbal material and visual/2- or 3-dimensional information as well as an effectively unlimited long-term memory, holding schemas that vary in their degree of automation. These structures and functions of human cognitive architecture have been used to design a variety of novel instructional procedures based on the assumption that working memory load should be reduced and schema construction encouraged. This paper reviews the theory and the instructional designs generated by it. KEY WORDS: cognition; instructional design; learning; problem solving.

Traditionally, Cognitive Load Theory (CLT) has focused on instructional methods to decrease extraneous cognitive load so that available cognitive re-sources can be fully devoted to learning. This article strengthens the cognitive base of CLT by linking cognitive processes to the processes used by biologi-cal evolution. The article discusses recent developments in CLT related to the current view in instructional design that real-life tasks should be the driving force for complex learning. First, the complexity, or intrinsic cognitive load, of such tasks is often high so that new methods are needed to manage cogni-tive load. Second, complex learning is a lengthy process requiring learners’ motivational states and levels of expertise development to be taken into ac-count. Third, this perspective requires more advanced methods to measure expertise and cognitive load so that instruction can be flexibly adapted to in-dividual learners ’ needs. Experimental studies are reviewed to illustrate these recent developments. Guidelines for future research are provided.

Students viewed a computer animation depicting the process of lightning. In Experiment 1, they concurrently viewed on-screen text presented near the animation or far from the animation, or concurrently listened to a narration. In Experiment 2, they concurrently viewed on-screen text or listened to a narration, viewed on-screen text following or preceding the animation, or listened to a narration following or preceding the animation. Learning was measured by retention, transfer, and matching tests. Experiment 1 revealed a spatial-contiguity effect in which students learned better when visual and verbal materials were physically close. Both experiments revealed a modality effect in which students learned better when verbal input was presented auditorily as speech rather than visually as text. The results support 2 cognitive principles of multimedia learning. Technological advances have made possible the combina-tion and coordination of verbal presentation modes (such as narration and on-screen text) with nonverbal presentation modes (such as graphics, video, animations, and environmen-tal sounds) in just one device (the computer). These ad-vances include multimedia environments, where students can be introduced to causal models of complex systems by the use of computer-generated animations (Park & Hopkins, 1993). However, despite its power to facilitate learning, multimedia has been develo~ed on the basis of its technologi-.d cal capacity, and rarely is it used according to research-based principles (Kozma, 1991; Mayer, in press; Moore, Burton, & Myers, 1996). Instructional design of multimedia is still mostly based on the intuitive beliefs of designers rather than on empirical evidence (Park & Hannafin, 1994). The purpose of the present study is to contribute to multi-media learning theory by clarifying and testing two cogni-tive principles: the contiguity principle and the modality principle.

In this article, we discuss cognitive load measurement techniques with regard to their contribu-tion to cognitive load theory (CLT). CLT is concerned with the design of instructional methods that efficiently use people’s limited cognitive processing capacity to apply acquired knowledge and skills to new situations (i.e., transfer). CLT is based on a cognitive architecture that consists of a limited working memory with partly independent processing units for visual and auditory information, which interacts with an unlimited long-term memory. These structures and func-tions of human cognitive architecture have been used to design a variety of novel efficient in-structional methods. The associated research has shown that measures of cognitive load can re-veal important information for CLT that is not necessarily reflected by traditional performance-based measures. Particularly, the combination of performance and cognitive load measures has been identified to constitute a reliable estimate of the mental efficiency of instruc-tional methods. The discussion of previously used cognitive load measurement techniques and their role in the advancement of CLT is followed by a discussion of aspects of CLT that may benefit by measurement of cognitive load. Within the cognitive load framework, we also discuss some promising new techniques.

This article reports findings on the use of a partly auditory and partly visual mode of presentation for geometry worked examples. The logic was based on the split-attention effect and the effect of presentation modality on working memory. The split-attention effect occurs when students must split their attention between multiple sources of information, which results in a heavy cognitive load. Presentation-modality effects suggest that working memory has partially independent processors for handling visual and auditory material. Effective working memory may be increased by presenting material in a mixed rather than a unitary mode. If so, the negative consequences of split attention in geometry might be ameliorated by presenting geometry statements in auditory, rather than visual, form. The results of 6 experiments supported this hypothesis. In recent years, working memory limitations have been identified as a major factor that needs to be considered when instruction is designed. Researchers have used cognitive load theory (e.g., Sweller, 1988, 1989, 1993, 1994) to sug-gest that many commonly used instructional procedures are

The goal of this study was to investigate individual differences in learning from worked-out examples with respect to the quality of self-explanations. Restric-tions of former studies (e.g., lacking control of time-on-task) were avoided and additional research questions (e.g., reliability and dimensionality of self-explanation characteristics) were addressed. An investigation with 36 university freshmen students of education working in individual sessions was conducted. The domain was probability calculation. Prior knowledge and the quality of self-explanations (protocols of the individuals ’ thinking aloud) were assessed as predictors of learning. A post-test was employed to measure the learning gains as the dependent variable. The following main results were obtained. Most self-explanation characteristics could be regarded as relatively stable person characteristics. The individual differences in the quality of self-explan-ations were, however, found to be multidimensional. Most important, even when controlling for time-on-task (quantitative aspect), learning gains could be substantially predicted by qualitative differences of self-explanation char-acteristics. In particular, swccessfwl learners tended to employ more principle-based explanations, more explication of operator-goal combinations, and more anticipative reasoning. In addition, there were two types of effective learners, labeled anticipative reasoners and principle-based explohers. Worked-out examples consist of the givens of a problem, solution steps, and the final solution itself. Learning from worked-out examples is an important source of learning (VanLehn, 1986, 1996), and it is a learning mode preferred by novices

Two experiments investigated alternatives to split-attention instructional designs. It was assumed that because a learner has a limited working memory capacity, any increase in cognitive resources required to process split-attention materials decreases resources available for learning. Using computer-based instructional material consisting of diagrams and text, Experiment 1 attempted to ameliorate split-attention eects by increasing eective working memory size by presenting the text in auditory form. Auditory presentation of text proved superior to visual-only presentation but not when the text was presented in both auditory and visual forms. In that case, the visual form was redundant and imposed a cognitive load that interfered with learning. Experiment 2 ameliorated split-attention eects by using colour coding to reduce cognitive load inducing search for diagrammatic referents in the text. Mental load rating scales provided evidence in both experiments that alternatives to split-attention instructional designs were eective due to reductions in cognitive load.

Worked examples are instructional devices that provide an expert's problem solution for a learner to study. Worked-examples research is a cognitive-experimental program that has relevance to classroom in-struction and the broader educational research community. A frame-work for organizing the findings of this research is proposed, leading to instructional design principles. For instance, one instructional de-sign principle suggests that effective examples have highly integrated components. They employ multiple modalities in presentation and em-phasize conceptual structure by labeling or segmenting. At the lesson level, effective instruction employs multiple examples for each concep-tual problem type, varies example formats within problem type, and employs surface features to signal deep structure. Also, examples should be presented in close proximity to matched practice problems. More-over, learners can be encouraged through direct training or by the structure of the worked example to actively self:explain examples. Worked examples are associated with early stages of skill develop-ment, but the design principles are relevant to constructivist research and teaching. The Historical Context In recent years, learning from "worked examples " has received a consider-able amount of attention from researchers (e.g., Chi, Bassok, Lewis, Reimann, & Glaser, 1989; Ward & Sweller, 1990), particularly in such fields as mathematics, physics, and computer programming. Although there is no precise definition, worked examples share certain family resemblance (Wittgenstein, 1953). As instructional devices, they typically include a problem statement and a proce-dure for solving the problem; together, these are meant to show how other similar problems might be solved. In a sense, they provide an expert's problem-

Intelligent tutoring systems are highly interactive learning environments that have been shown to improve upon typical classroom instruction. Cognitive Tutors are a type of intelligent tutor based on cognitive psychology theory of problem solving and learning. Cognitive Tutors provide a rich problem-solving environment with tutorial guidance in the form of step-by-step feedback, specific messages in response to common errors, and on-demand instructional hints. They also select problems based on individual student performance. The learning benefits of these forms of interactivity are supported, to varying extents, by a growing number of results from experimental studies. As Cognitive Tutors have matured and are being applied in new subject-matter areas, they have been used as a research platform and, particularly, to explore interactive methods to support metacognition. We review experiments with Cognitive Tutors that have compared different forms of interactivity and we reinterpret their results as partial answers to the general question: How should learning environments balance information or assistance giving and withholding to achieve optimal student learning? How best to achieve this balance remains a fundamental open problem in instructional science. We call this problem the “assistance dilemma ” and emphasize the need for further science to yield specific conditions and parameters that indicate when and to what extent to use information giving versus information withholding forms of interaction.

Abstract not found