Expertise

ADAPTIVE EXPERTISE
Giyoo Hatano

DOMAIN EXPERTISE
K. Anders Ericsson
Robert R. Hoffman

ADAPTIVE EXPERTISE

Investigators on knowledge transfer have almost unanimously concluded that students seldom effectively apply short-term training at school to problem-solving situations outside school. Experts, who have had years of problem-solving experience in a given domain, may not be different–they can solve familiar types of problems quickly and accurately but fail to go beyond procedural efficiency. Even in knowledge-rich domains, experts may be characterized by their possession of problem schemas by which they classify problems and apply one of the solution routines accordingly.

The notion of adaptive experts as against routine experts was proposed by Giyoo Hatano as an ideal of educational researchers looking to find ways to teach students so they can apply learned procedures flexibly or adaptively. Keith Holyoak aptly describes the distinction: "Whereas routine experts are able to solve familiar types of problems quickly and accurately, they have only modest capabilities in dealing with novel types of problems. Adaptive experts, on the other hand, may be able to invent new procedures derived from their expert knowledge"(p. 310). Giyoo Hatano and Kayoko Inagaki, in their 1986 paper, slightly expand the above characterization of adaptive experts: They are able to (1) comprehend why those procedures they know work; (2) modify those procedures flexibly when needed; and (3) invent new procedures when none of the known procedures are effective.

Sources of Adaptiveness

Where does the adaptiveness of adaptive experts come from? Adaptive experts are assumed to possess as the source of their flexibility and inventiveness, conceptual knowledge of the objects of the procedures (that is, what each of these objects is like). "Flexibility and adaptability seem to be possible only when there is some corresponding conceptual knowledge to give meaning to each step of the skill and provide criteria for selection among possible alternatives for each step within the procedure" (Hatano 1982, p. 15). Such conceptual knowledge enables experts to construct mental models of the major entities of the domain, which can be used in mental simulation. Using Holyoak's expression, the key to adaptive expertise is the development of deeper conceptual understanding of the target domain. Needless to say, such conceptual understanding must be connected to procedural competencies and meta-cognitive awareness and monitoring of one's own understanding.

It is hypothesized that if people ask themselves why a skill works or why each step is needed during its application, this question will tend to lead them to form some conceptual knowledge about the object. This was similar to what Donald Schoen in 1983 called "reflection-in-action" as against technical problem-solving in his attempt to characterize professionals. Although experts are seldom taught conceptual knowledge in the verbalized form, they may construct it in the process of solving problems or performing tasks in the domain.

Motivational and Contextual Conditions

When are people likely to gain adaptive expertise? Identifying particular kinds of learning experiences that develop adaptive expertise is a serious challenge for educational researchers. In 1992 Hatano and Inagaki proposed four conditions that would promote sustained comprehension activity that is likely to lead to adaptive expertise. Their proposal is based on the assumption that cognitive incongruity (a state of feeling that current comprehension is inadequate; for example, wondering why a given procedure works) induces enduring comprehension activity, including seeking further information from the outside, retrieving another piece of prior knowledge, generating new inferences, examining the compatibility of inferences more closely, and so forth. The first two of the proposed conditions are concerned with the arousal of cognitive incongruity and the last two with the elicitation of committed and persistent comprehension activity in response to induced incongruity. The four conditions are: (1) encountering fairly often a novel problem to which prior knowledge is not readily applicable or a phenomenon that disconfirms a prediction based on prior knowledge;(2) engaging in frequent dialogical interaction, such as discussion, controversy, and reciprocal teaching;(3) being free from urgent external need (e.g., material rewards or positive evaluations), and thus able to pursue comprehension even when it is time consuming; and (4) being surrounded by reference group members who value understanding.

These conditions can be rephrased in terms of the nature of the practice in which people participate. For example, when a practice is oriented toward skillfully solving a fixed class of problems (e.g., making the same products for years), participants tend not to encounter novel problems, and thus they are likely to become experts distinguished in terms of speed, accuracy, and automaticity (i.e., routine experts). In contrast, when successful participation in a practice requires meeting varied and changing demands (e.g., making new, fashionable products), participants' prior knowledge must be applied flexibly, and they are likely to acquire adaptive skills. From sociocultural perspectives, adaptive experts may not be characterized only by their domain-specific knowledge; in order to invent new procedures, for example, in addition to deeper conceptual understanding, people have to be able to participate in discourse, offer valuable suggestions, evaluate others' suggestions, and so on.

Educational Implications

At school students are seldom expected to become experts in a particular domain. Rather they are expected to learn in many subject-matter domains. Thus the extent to which school instruction and vocational expertise share similar goals and processes of learning is debatable. Trying to build a classroom as a community of adaptive experts is challenging, but practically it may be too ambitious or require too much effort on the part of educators.

Whatever the ultimate goal for instruction should be, the concept of adaptive expertise "provides an important model of successful learning" (Bransford, Brown, and Cocking, p. 36). It is thus encouraging to see attempts to apply this notion to teaching and learning in mathematics, science, history, and other subjects.

Expertise refers to the psychological mechanisms underlying the superior achievement of an expert and the social forces that designate the status of being an expert, that is, "one who has acquired special skill in or knowledge of a particular subject through professional training and practical experience" (Webster's Third New International Dictionary, p. 800). The term expert has been used to describe highly experienced professionals, such as medical doctors, accountants, teachers, and scientists, but has been expanded to include any individual who has attained superior performance by instruction and extended practice, ranging from bird-watchers to pianists, golfers to chess players.

Theoretical Frameworks

Experts' behavior looks so effortless and natural that it is often attributed to special talents, though knowledge and training are necessary. The role of acquired skill for the highest levels of achievement has traditionally been minimized. However, when scientists began measuring the experts' presumed superior powers of speed of thought, memory, and intelligence with psychometric tests, no general superiority was found–the demonstrated superiority was domain specific. For example, the chess experts' vastly superior memory was constrained to regular chess positions and did not generalize to other types of materials. Not even IQ could distinguish the chess masters among chess players nor the most successful and creative among artists and scientists. In K. Anders Ericsson and Andreas C. Lehmann's 1996 review, it was found that (1) measures of general basic capacities do not reliably predict success in a domain; (2) the superior performance of experts is often very domain specific and transfer outside their narrow area of expertise is limited; and (3) systematic differences between experts and less proficient individuals nearly always reflect attributes acquired by the experts during their lengthy training.

Thought processes. In a pioneering empirical study first published in 1946 of the thought processes at the highest levels of performance, Adrian de Groot instructed expert and world-class chess players to think aloud while they selected their next move for an unfamiliar chess position. The world-class players did not differ in the speed of their thoughts or the size of their basic memory capacity, and their ability to recognize promising potential moves was based on their extensive experience and knowledge of patterns in chess. In their influential 1973 theory of expertise, Herbert A. Simon and William G. Chase proposed that experts with extended experience acquire and remember a larger number of complex patterns and use these new patterns to store new knowledge about which actions should be taken in similar situations.

According to this influential theory, expert performance is viewed as the result of skill acquired with gradual improvements of performance during extended experience in a domain. Furthermore, this theory assigned a central role of acquired knowledge and encouraged efforts to elicit experts' knowledge to build computer models of experts, that is, expert systems.

It is tempting to assume that the performance of experts improves as a direct function of increases in knowledge through training and extended experience. However, there are many demonstrations that extensive domain knowledge does not necessarily entail superior performance. For example, the outcome of psychological therapy does not increase as a function of the length of training and professional experience of the therapist. Similarly, the accuracy of decision-making–such as medical diagnosis for common diseases and the quality of investment decisions–does not improve with further professional experience. More generally, the number of years of work and experience in a domain is a poor predictor of attained performance.

The development of expert performance. In a 1985 pioneering study, Benjamin S. Bloom and his colleagues studied the developmental history of scientists, athletes, and artists who had attained international awards for their outstanding achievements. These elite performers did not attain their performance from regular experience in their respective domains but were given access to superior instruction in the best educational environments. Their families provided them substantial financial and emotional support to allow them to focus fully on the development of their performance. Bloom's influential research demonstrated the necessity for extended training in the best training environments to reach the highest levels of performance.

Effects of practice and experience. Subsequent research published in 1993 by Ericsson, Ralf Krampe, and Clemens Tesch-Römer analyzed the effects of different types of experience on the improvement of performance. They found that in activities where individuals had attained an acceptable level of performance, such as recreational golf and many professions, even decades of continued experience was not associated with any improvement of performance. The researchers proposed that in those domains where performance consistently increases, aspiring expert performers seek out particular kinds of experience, that is, deliberate practice–activities designed, typically by a teacher, for the sole purpose of effectively improving specific aspects of an individual's performance. In support of a critical role of deliberate practice, expert musicians differing in the level of attained solo performance also differed in the amounts of time they had spent in solitary practice during their skill development, which totaled around 10,000 hours by age twenty for the best experts, around 5,000 hours for the least accomplished expert musicians, and only 2,000 hours for serious amateur pianists. More generally, the accumulated amount of deliberate practice is closely related to the attained level of performance of many types of experts, such as musicians, chess players, and athletes.

Conclusions

Advances in the understanding of the complex representations, knowledge, and skills that mediate experts' superior performance derive from studies where experts are instructed to think aloud while completing representative tasks and from studies using methods of cognitive task analysis and cognitive field research. These process-tracing studies have shown that the difference between experts and less-skilled individuals is not merely a matter of the amount and complexity of the accumulated knowledge; it also reflects qualitative differences in strategies, the organization of knowledge, and the representation of problems. During the acquisition of their performance, experts acquire domain-specific memory skills that allow them to rely on long-term memory to dramatically expand the amount of information that can be kept accessible during planning and during reasoning about alternative courses of action. The superior quality of the experts' mental representations allows them to adapt to changing circumstances as well as anticipate future events in advance, so the expert performers can respond with impressive speed. The same acquired representations appear to be essential for experts' ability to monitor and evaluate their own performance so they can keep improving by designing their own training and assimilating new knowledge.