Evaluating and Managing Cognitive Load in Games

Evaluating and Managing Cognitive Load in Games

Slava Kalyuga (University of New South Wales, Australia) and Jan L. Plass (New York University, USA)
Copyright: © 2009 |Pages: 19
DOI: 10.4018/978-1-59904-808-6.ch041
OnDemand PDF Download:
No Current Special Offers


This chapter provides an overview of our cognitive architecture and its implications for the design of game-based learning environments. Design of educational technologies should take into account how the human mind works and what its cognitive limitations are. Processing limitations of working memory, which becomes overloaded if more than a few chunks of information are processed simultaneously, represent a major factor influencing the effectiveness of learning in educational games. The chapter describes different types and sources of cognitive load and the specific demands of games on cognitive resources. It outlines information presentation design methods for dealing with potential cognitive overload, and presents some techniques (subjective rating scales, dual-task techniques, and concurrent verbal protocols) that could be used for evaluating cognitive load in electronic gaming in education.
Chapter Preview


The field of gaming and play-based virtual environments as a new educational technology and research area is rapidly expanding (e.g., Gee, 2003; Nelson, Ketelhut, Clarke, Bowman, & Dede, 2005; Shaffer, 2006). If we expect this technology to be efficient in helping students to acquire new, complex knowledge and skills, its design should be based on knowledge of our cognitive architecture and its role in learning and problem solving. Processing limitations of working memory represent a major factor influencing the effectiveness of learning and performance, especially for novice learners. For example, committing limited cognitive resources to processing irrelevant, non-essential, distracting information; on searching for inadequately located references; or on trying to make essential connection between sources of information that are artificially separated in space or time due to poor interface design could substantially slow down learning and performance.

Considering these limitations is particularly important for educational gaming technologies because games usually require simultaneous performances of several cognitive and motor activities. For example, in the game Peeps, designed as part of the RAPUNSEL project to teach middle-school girls how to program, players have to navigate the 3D virtual environment, search for objects of value, communicate with other players, avoid gobblers who try to steal from them, and collect peaches to maintain their energy level (Plass, 2007a). The educational portion of the game, aimed at learning a Java-like programming language in order to design outfits and dances for their avatar, makes additional requirements on the players’ cognitive resources. Efficient information designs therefore must focus on substantially reducing cognitive stress in order to enhance learning outcomes.

Levels of learner prior knowledge and experience in a domain represent another important related factor that may significantly influence learning from educational games. Performance and learning characteristics of experienced learners differ considerably from those of novices. Well-organized and often fully or partially automated schematic knowledge structures allow more experienced learners to rapidly recognize and categorize familiar patterns of information without overloading working memory, thus avoiding cognitive stress (Sweller, van Merriënboer, & Paas, 1998; van Merriënboer & Sweller, 2005). The information design in educational games should support the rapid acquisition and use of such knowledge structures by reducing or eliminating unnecessary cognitive overload that may otherwise prevent the allocation of sufficient cognitive resources required for efficient learning and performance.

It should be noted that cognitive load—that is, the demand on cognitive resources during problem solving and reasoning—is always associated with conscious cognitive processes that take place in the learner working memory while performing a current cognitive task. Therefore, the issue of cognitive overload is different from (although it may be related to) problems of general information content overload over longer periods of time or perceptual overload that is traditionally considered in interface design and usability evaluation procedures (e.g., Nielsen, 1995). Cognitive load theory is dealing with factors that influence conscious information processing as we perform a specific task in real time on a scale of seconds or minutes rather than hours or days (in other words, we are dealing with micro- rather than macro-level analysis). Many games have procedures in place that have the potential to overcome high cognitive load for critical tasks, for example, by explicitly providing critical information to solve a task on demand and just in time (Gee, 2003), though the effectiveness of these strategies in reducing cognitive load has not yet been tested empirically.

Key Terms in this Chapter

Sources of Cognitive Load: Features of external information structures or cognitive characteristics of individual users that determine required working memory resources. Intrinsic cognitive load is caused by levels of interactivity between elements of information that need to be processed simultaneously. Extraneous cognitive load is imposed by the design of information presentations (e.g., separated in space- and/or time-related elements; an excessive step-size or rate of introducing new elements of information; limited user knowledge that is not compensated by provided support; user knowledge base that overlaps with provided external guidance).

Long-Term Memory (LTM): A major part of our cognitive architecture, an organized knowledge base that stores a massive amount of hierarchical knowledge structures.

Cognitive Load: Working memory resources required for processing specific information by an individual user. Cognitive load theory distinguishes between the essential (intrinsic) and wasteful (extraneous) forms of cognitive load, and suggests a variety of techniques and procedures (cognitive load effects) for managing essential and reducing extraneous load in learning.

Working Memory (WM): A major part of our cognitive architecture, a functional mechanism that limits the scope of immediate changes to long-term memory. Depending on a specific model, WM is considered either as a separate component of our cognitive system or as an activated part of LTM. The essential attribute of WM is its severe limitation in capacity and duration when dealing with novel information.

Expertise Reversal Effect: Reversal in the relative effectiveness of information presentation formats and procedures as levels of user knowledge in a domain change. For example, extensive external support could be beneficial for novices when compared with the performance of novices who receive a low-support format, but is disadvantageous for more expert users when compared with the performance of experts who receive a low-support format.

Cognitive Load Theory: An instructional theory describing instructional implications of processing limitations of human cognitive architecture (capacity and duration of working memory) and evolved mechanisms for dealing with these limitations (long-term memory knowledge base and its role in cognition).

Cognitive Architecture: A general cognitive system that underlies human performance and learning. The understanding of human cognition within a cognitive architecture requires knowledge of corresponding models of memory organization, forms of knowledge representation, mechanisms of problem solving, and the nature of human expertise.

Complete Chapter List

Search this Book: