The experimental study of clinical guessing.

One of the chronic points of tension between experimental psychologists and clinical psychologists has been on the question of prediction. The experimentalist contends that genuine science leads to prediction, and doubts that the clinician can achieve it. Clinicians who assert that they can make valid forecasts of the behavior of their patients rarely publish materials which prove this can be done. When Adler described his clinical approach, he implied that he modifi.ed his impression of the patient as he gained data which disproved his initial, or general, diagnosis (4, pp. 326-327). According to his description, he tested his initial hypothesis, as any scientist should, and discarded it for another as necessary. He left little doubt that the proper application of this method should lead to prediction, asserting as early as 1923 (4, p. 193) that prediction is "the sole criterion of understanding." The bulk of published materials on clinical prediction suggests (~3), that the process is either not very accurate or is not superior to much simpler methods (for example, mechanical predictions made on the basis of psychological tests). The purpose of this paper is to describe an experimental method for the study and improvement of clinical inference-making, termed "guessing" by Adler, "predicting" by some investigators, "judging" by others. Experimental work with this method over the past few years will be reported, with the warning that the findings are at best only preliminary.