At the same time, our explication of the expertise approach is applicable to
any phenomenon involving reliably superior performance that can be captured
in the laboratory. We believe that an attempt to encompass phenomena nor-
mally labeled as perceptual (e.g., chicken sexing), motoric (e.g., typing), or
knowledge-based (e.g., physics) within the same overall approach will allow us
to identify common methodological and theoretical issues and to consider a
common and more differentiated set of learning mechanisms in accounting for
achievement of superior performance in any one of these different domains.
Such an approach will have the additional advantage of allowing us to consider
the many different perceptual, memory, motoric, and knowledge-based aspects
of superior performance in domains like chess (Charness, 1991), physics (Anzai,
1991), medicine (Patel & Groen, 1991), performing arts and sports (Allard &
Starkes, 1991), and music (Sloboda, 1991).
Capturing Superior Performance: The First Step
The first step in the expertise approach involves finding or designing a collec-
tion of tasks to capture the superior performance in the appropriate domain.
If one is able to identify such a collection of tasks, the following important
advantages will accrue: First, the performance of the designed tasks will reflect
the stable characteristics of the superior real-life performance. More important,
the availability of such a collection of tasks will allow us to study the perfor-
mance of the experts extensively in order to accumulate sufficient information
on the mediating processes to make a detailed assessment and analysis. During
these extensive observations of performance, we should not expect significant
changes due to learning and practice, as we shall be monitoring stable processes
that have been adapted and perfected over a long period of time. The period
during which performance will be observed will be negligible in comparison.
Finally, these collections of tasks will provide us with an excellent testing
ground for studying how rapidly the various identified characteristics can be
acquired through practice. In fact, one could argue that with an adequate col-
lection of tasks, the rates of acquisition should be comparable for practice with
the collection of tasks and in real life. If, on the other hand, the collection of
designed tasks does not elicit the mechanisms that mediate superior real-life
performance, or does so only partially, then we are likely to see substantial
learning and changes in the processes as a result of further practice. Collections
of tasks that lead to rapid rises in levels of performance by experts with further
practice are unlikely to yield an adequate representation of superior perfor-
mance. Even more devastating evidence against the claim that such a collection
can capture superior performance comes from situations in which novices have
matched or surpassed the performance levels of experts after only a few weeks
or months of practice.
For some types of expertise it is easy to identify such a collection of tasks, but
in most cases it is the most difficult step. We shall first describe some simple
cases and then turn to the difficult issues involved in designing a collection of
tasks to characterize real-life expertise. We shall also consider the advantages
and problems of designing a collection of memory tasks to study superior
memory performance by experts, as opposed to studying directly the superior
performance of experts.
528 K. Anders Ericsson and Jacqui Smith