The Effects of Manipulating the Degree of Belief in a Diagnostic Hypothesis on Feature Detection

Abstract

In Experiment 1, the degree of belief in a focal hypothesis was manipulated using priming as well as the principle of unpacking of Tversky and Koehler (1994). The effects of these manipulations on feature detection was measured. It was found that regardless of the degree of belief in the focal hypothesis, novice diagnosticians who have it in mind will call more of its features than those who do not have it in mind. It is believed that this is due to the fact that having a diagnosis in mind seems to focus the attention of diagnosticians to the relevant features. Also, our manipulation of suggesting alternatives to the diagnosticians did not have the effect of decreasing the diagnosticians' belief in the focal hypothesis, contrary to what is predicted by Tversky and Koehler's unpacking principle (1994). The results from Experiment 1 suggest, and those from Experiment 2 confirm the hypothesis that in order to decrease the degree of belief in the focal hypothesis when it is presented with alternatives, the alternatives must be plausible. If the focal hypothesis is extremely dominant over the alternatives, a reversal of the unpacking principle will occur.