Towards a rational theory of human information acquisition
Jonathan D. Nelson · 2008
Abstract This chapter considers the crucial question of the usefulness of data, with respect to a set of hypotheses under consideration. Some data are clearly irrelevant to the hypotheses of concern; some decisively favour one hypothesis or another. But, in general, data modifies the probability of the hypotheses, but to a modest degree. Only with a measure of the usefulness of data is it possible to formulate a theory of which experiments should be carried out, or which observations should be made. It provides a thorough analysis of the strengths and weaknesses of a range of measures of evidential usefulness, drawn from a variety of fields of cognitive science.