Investigating domain-knowledge and reasoning as predictors of diagnostic performance with PATSy.

This study investigated the clinical reasoning of speech and language pathology students who used PATSy (a web-based, multimedia database of clinical cases for education and research) in the context of a cognitive neuropsychology course in aphasia. Theoretically, the course was oriented around an information processing account of normal and pathological language production. PATSy contains numerous virtual patient cases consisting of multimedia data on individuals who have language production disorders. The aims were a) to assess PATSy as a research tool in the study of clinical reasoning, b) to investigate students clinical reasoning and that of experts, and c) to learn more about the relative contributions of domain-specific knowledge and general reasoning ability to diagnostic performance on PATSy. Experts and students used PATSy under exam conditions to diagnose a previously unseen patient case. Subjects notes, hypotheses and conclusions were logged in the course of their reasoning, together with a trace of their assessment test selections. Students were pre-tested on domain-general and domain-specific knowledge. The results showed that end-of-course diagnostic reasoning performance was predicted by precourse domain knowledge, the number of PATSy cases accessed by students during the course, and deductive reasoning performance. The implications for a) teaching clinical reasoning skills and b) adding intelligent support to PATSy are discussed.