Abstract
Studies of clinical problem solving show that physicians form diagnostic hypotheses on the basis of minimal clinical findings and that these hypotheses create a context within which to gather additional relevant data.1 , 2 By an iterative process, hypotheses are either eliminated or confirmed, and those that survive are made progressively more specific. This concept of clinical problem solving is supported by extensive research in cognitive psychology and artificial intelligence.3 , 4 Nonetheless, these observations have not been exploited in the teaching of clinical medicine. Instead, students are expected to learn diagnostic problem solving by reading textbooks, studying clinicopathological conferences, attending conferences dominated by . . .

This publication has 11 references indexed in Scilit: