[ CogSci Summaries home | UP | email ]

Ericsson, K. A. & Charness, N. (1997). Cognitive and developmental factors in expert performance. In P. J. Feltovich, K. M. Ford, & R. R. Hoffman (Eds.), Expertise in context: Human and machine (pp 3-41). Cambridge, MA: MIT Press.

Author of the summary: David Zach Hambrick, 1998, gt8781a@prism.gatech.edu

What accounts for the highest levels of human performance? Ericsson and Charness claim that this question has not been well integrated with general psychological theories. This lack of integration is explained in the context of two major traditions in psychology: information processing (experimental) and individual differences. The first perspective attempts to explain performance in terms of knowledge and skills acquired through experience. The problem is that it requires many years to reach expert levels of performance. The second perspective focuses on individual difference characteristics, that is, "on the individual differences of exceptional performers that would allow them to succeed in a specific domain" (p. 5).


Ericsson and Charness propose an alternative approach—the study of expert performance. The goal of this approach is to identify mechanisms responsible for consistently reproducible expert performance. (A mechanism can be defined as the process through which a given result is achieved, that is, a cause effect explanation.) Extraordinary performances do not fall within the purview because they are not reproducible. It follows that "expert performance [is] consistently superior performance on a specified set of representative tasks for the domain that can be administered to any subject" (p. 6). Social recognition is not a valid indication of expert performance. Nor is domain-relevant experience. In short, expert performance is defined in terms of performance. (See Wagner and Stanovich, 1996, for a discussion of this definition; see also Stein, 1997, for a discussion of expertise in social contexts.) The focus then becomes: What distinguishes outstanding performers from less outstanding performers and from the general population. This definition makes sense (Wagner & Stanovich) if one assumes that expertise is acquired and that access to environmental support/opportunities is limited.


A critical step in the study of expert performance is to develop tasks that capture critical task-dimensions. The task should be complex enough to provide a valid measure of skill, but simple enough to be amenable to laboratory research. As Ericsson and Charness explain: "In general, researchers should strive to define the simplest situation in which experts’ superior performance can still be reliably reproduced" (p. 8). An example of a valid indicator of skill is de Groot’s choose the best move task for chess. Also, a distinction must be made between tasks that capture the essence of expert performance, and those on which experts show a performance advantage. Ericsson and Lehmann (1996) state: "It is important to distinguish tasks that capture the essence of expertise in a domain from other tasks in which the experts may also excel" (p. 282). An example of the former is the choose-a-move task; an example of the latter is an incidental memory task for meaningful and random chess configurations.


Once a laboratory administerable task is developed, the next step is to analyze performance and to identify mechanisms responsible for expert performance. Verbal protocol analysis is one approach. Performers are instructed to verbalize, but not to explain, their thoughts. Experimental techniques can also be used. For example, Chase and Simon showed novice and expert chess players both meaningful and randomly arranged configurations of chess positions. Their results indicated that knowledge is critical mediating mechanism. How meaningful configurations are recognized was then investigated.


As the preceding implies, a major tenet of the expert performance view is that the mechanisms that govern expert performance are acquired. Ericsson and Charness propose that one of the strongest pillars of support for this view is the evolution of domains. For example, "The knowledge in natural science and calculus that represented the cutting edge of mathematics a few centuries ago, and that only experts of that time were able to master, is today taught in high school and college" (p. 19). Historical improvements are evident in nearly every field (e.g., music, athletics, etc.).


How is expertise acquired? Ericsson and Charness argue that deliberate practice is the primary mechanism responsible for the attainment of expert levels of performance. Deliberate practice affords optimal opportunities for improvement through feedback. They distinguish this activity from other types of domain-relevant experience, including work and play. The basic assumption of the deliberate practice framework is that performance improves monotonically with amount of deliberate practice. Individual differences in performance among individuals who engage in comparable amounts of deliberate practice are explained by differences in the age at which deliberate practice was started.


Age and Expert Performance


Individuals often maintain high levels of performance throughout adulthood, while less proficient performers show declines in performance earlier. This performance advantage is, however, restricted to domain-specific activities. There is also evidence that older and younger performers can achieve a given level of proficiency in different ways. For example, in chess, Charness found that older chess players rely more on their extensive knowledge base than on planning (an interesting question, however, is whether this strategy shift is age-related). Salthouse has shown that in typing older adults rely on a large eye-hand span.


What is the role of deliberate practice in preserving proficiency with advancing age? Performance declines with age. Reductions in deliberate practice involvement might explain. For example, physiological adaptations revert to normal levels after cessation of deliberate practice. But what accounts for decreased involvement in deliberate practice activities? Social factors may contribute. For example, older individuals are expected to devote more time to family and/or career responsibilities than younger individuals. It seems likely, however, that there might be a reciprocal relationship between age-related performance and deliberate practice changes such that declines lead to less deliberate practice involvement, which lead to further declines.


Implications of Study of Expert Performance


Claims about deliberate practice are relevant to a number of more general psychological issues. For example, according to a popular view of aging, age-related declines in functioning are attributable to disuse—use it or lose it. Deliberate practice predictions are directly relevant to this claim. The deliberate practice framework is also relevant to the question of the modifiability, or invariance, of cognitive capacities. That is, "Extended deliberate practice gives near maximal values on the possible effects of environmental variables (in interaction with development variables) relevant to theoretical claims for invariant cognitive capacities and general laws of performance" (p. 34). Finally, the deliberate practice framework attempts to show how factors operating at other levels of analysis (e.g., social) affect psychological functioning.


Back to the Cognitive Science Summaries homepage
Cognitive Science Summaries Webmaster:
JimDavies (jim@jimdavies.org)
Last modified: Fri Mar 3 16:22:49 EST 2000