Categorization and rule induction in clinical diagnosis and assessment

This chapter examines the role of categorization and rule induction in clinical evaluation. First, general characteristics of categorization and rule induction processes in clinical evaluation are examined and compared to those of laboratory categorization experiments. The implications of these differences are discussed in terms of the nature of the models which need to be developed to understand categorization in clinical evaluation. It reviews a summary of the major characteristics and purposes of diagnosis, recent advances in understanding clinicians' mental representations of psychiatric disorders. Lower-level categorization in clinical diagnosis is also examined. This is the categorization occurring prior to the disorder classification and involves decisions about the diagnostic criteria of a diagnostic disorder. Issues which are addressed include (1) the decomposition, context embeddedness, and multi-determination of features in clinical categorization, (2) the contribution of the clinician's domain knowledge to successful categorization, and (3) mechanisms whereby both clinical theory and previous exemplar and category-label information can interfere with successful diagnosis. These issues are formalized in a preliminary model of lower-level categorization in clinical diagnosis. The chapter also reviews that the clinical assessment is differentiated from clinical diagnosis in terms of the nature of the categorization goal and process.