Diagnostic classification by experts and novices

The present study examines differences between experts and novices in classifying symptoms and the effect of the nature of the task on classifying. The study involved three groups of subjects, two expert groups (n = 12, n = 10) and one novice group (n = 12). Thinking-aloud protocols were collected for two classification tasks: sorting of behavioural symptoms into predefined categories and intuitive clustering of behavioural symptoms. In a third task, experts and novices were asked to make typicality ratings of behavioural symptoms. The protocols were analyzed with respect to seven cognitive operations: (a) asking or giving information, (b) associating, (c) abstracting or labelling, (d) explaining, (e) neutral matching, (f) identifying, and (g) differentiating. Results showed an effect of experience and an effect of the task on the relative frequencies of these operations. No differences were found in typicality rating of experts versus novices. These contradictory findings are discussed in relation to Kolodner's model about the evolution of expertise.