Search

Human behavior in many situations involving uncertainty and risk depends on the acquisition of information concerning event probabilities. A family of statistical models, growing out of general mathematical learning theory, accounts for probability learning in terms of the accumulation in memory of weighted ensembles of associations between recurring situations and subsequent events. These models provide rather detailed quantitative accounts of probability learning in some especially simplified experimental situations. Also they provide vehicles for applying theoretical interpretations of probability learning to the problems of choice and decision making in social and economic contexts.

In the studies the emphasis is on the applied mathematical statistics and model construction of simple Markov processes. The models are very simple ones and in their present form are intended only to provide rough estimates of certain epidemiological parameters that are not directly observable or that can be secured only by expensive and time-consuming field surveys. (Author)

The well-known discussion of the principles of experimentation, illustrated by a taste-testing problem, in R. A. Fisher's Design of Experiments, is the basis of this expository paper. The notion of a hypothetical population of identical experiments is defended. It is argued that attention must be paid to non-null cases (in testing theory) if a satisfactory probabilistic model for sensory sorting tests is to be built, and if the efficiency of various experimental designs is to be considered. Finally, some remarks are made on the role of randomization, and on the problem of “inexact” acceptance regions in discrete distributions.