More generally, over the last century there have been economic developments with public broadcasts of competitions and performances that generate sufficient revenue for a number of domains of expertise, such as sports and chess, to support professional full-time performers as well as coaches, trainers, and teachers. In these new domains, along with the traditional professions, current and past expert performers continue to be the primary teachers at the advanced level (masters), and their professional associations have the responsibility of certifying acceptable performance and the permission to practice. Accordingly, they hold the clout in thus influencing training in professional schools, such as law, medical, nursing, and business schools \ufffd \ufffdtesting is the tail that wags the dog\ufffd (Feltovich, personal communication) \ufffd as well as continuing education training (see Evetts, Meig, & Felt, Chapter 7 on sociological perspectives on expertise). The accumulation of knowledge about the structure and acquisition of expertise in a given domain, as well as knowledge about the instruction and training of future professionals, has occurred, until quite recently, almost exclusively within each domain with little cross-fertilization of domains in terms of teaching, learning methods, and skill-training techniques.

It is not immediately apparent what is generalizable across such diverse domains of expertise, such as music, sport, medicine, and chess. What could possibly be shared by the skills of playing difficult pieces by Chopin, running a mile in less than four minutes, and playing chess at a high level? The premise for a field studying expertise and expert performance is that there are sufficient similarities in the theoretical principles mediating the phenomena and the methods for studying them in different domains that it would be possible to propose a general theory of expertise and expert performance. All of these domains of expertise have been created by humans. Thus the accumulated knowledge and skills are likely to reflect similarities in structure that reflect both human biological and psychological factors, as well as cultural factors. This raises many challenging problems for methodologies used to describe the organization of knowledge and mechanisms and reveals the mediating expert performance that generalizes across domains.

Once we know how experts organize their knowledge and their performance, is it possible to improve the efficiency of learning to reach higher levels of expert performance in these domains? It should also be possible to answer why different individuals improve their performance at different rates and why different people reach very different levels of final achievement. Would a deeper understanding of the development and its mediating mechanisms make it possible to select individuals with unusual potential and to design better developmental environments to increase the proportion of performers who reach the highest levels? Would it be possible even to facilitate the development of those rare individuals who make major creative contributions to their respective domains?

Individual Differences in Mental Capacities

A widely accepted theoretical concept argues that general innate mental capacities mediate the attainment of exceptional performance in most domains of expertise. In his famous book, \ufffdHeriditary Genius,\ufffd Galton (1869/1979) proposed that across a wide range of domains of intellectual activity the same innate factors were required to attain outstanding achievement and the designation of being a genius. He analyzed eminent individuals in many domains in Great Britain and found that these eminent individuals were very often the offspring of a small number of families \ufffd with much higher frequency than could be expected by chance. The descendents from these families were much more likely to make eminent contributions in very diverse domains of activity, such as becoming famous politicians, scientists, judges, musicians, painters, and authors. This observation led Galton to suggest that there must be a heritable potential that allows some people to reach an exceptional level in any one of many different domains. After reviewing the evidence that height and body size were heritable Galton (1869/1979) argued: \ufffdNow, if this be the case with stature, then it will be true as regards every other physical feature \ufffd as circumference of head, size of brain, weight of grey matter, number of brain fibres, &c.; and thence, a step on which no physiologist will hesitate, as regards mental capacity\ufffd (pp. 31\ufffd32, emphasis added).

Galton clearly acknowledged the need for training to reach high levels of performance in any domain. However, he argued that improvements are rapid only in the beginning of training and that subsequent increases become increasingly smaller, until \ufffdmaximal performance becomes a rigidly determinate quantity\ufffd (p. 15). Galton developed a number of different mental tests of individual differences in mental capacity. Although he never related these measures to the objective performance of experts on particular real-world tasks, his views led to the common practice of using psychometric tests for admitting students into professional schools and academies for arts and sports with severely limited availability of slots. These tests of basic ability and talent were believed to identify the students with the capacity for reaching the highest levels.

In the 20th century scientists began the psychometric testing of large groups of experts to measure their powers of mental speed, memory, and intelligence. When the experts\ufffd performance was compared to control groups of comparable education, there was no evidence for Galton\ufffds hypothesis of a general superiority for experts because the demonstrated superiority of experts was found to be limited to specific aspects related to the particular domain of expertise. For example, the superiority of the chess experts\ufffd memory was constrained to regular chess positions and did not generalize to other types of materials (Djakow, Petrowski, & Rudik, 1927). Not even IQ could distinguish the best among chess players (Doll & Mayr, 1987) or the most successful and creative among artists and scientists (Taylor, 1975).

In a recent review, Ericsson and Lehmann (1996) found that (1) measures of basic mental capacities are not valid predictors of attainment of expert performance in a domain, (2) the superior performance of experts is often very domain specific, and transfer outside their narrow area of expertise is surprisingly limited, and (3) systematic differences between experts and less proficient individuals nearly always reflect attributes acquired by the experts during their lengthy training.