A SIMPLE PROGRAMMING TECHNIQUE FOR COMPARING AVERAGE RESPONSES.

Abstract A technique is described for presenting stimuli in irregular order and entering response samples into appropriate subdivisions of a computer memory for averaging. The approach is conceptually simple and offers advantages of low cost and considerable flexibility in programming. An illustration is given of one program that can be applied to the control of sampling errors. This method, which is somewhat analogous to a split-half reliability test, imposes limits on differences between averages that can be confidently interpreted. One additional application to the study of sensory interaction effects is shown.