INFORMATION HYPOTHESIS AND NON-REPETITIONS

The information hypothesis specifies that equi-information conditions will yield the same mean RT provided that S can and does take advantage of all the information available to him prior to the occurence of each new signal. KORNBLUM'S recent experiment (1968) demonstrated RT was not the same for equi-information conditions differing in proportion of repetitions. One can conclude with Kornblum that his findings disconfirm the information hypothesis or one can look for factors in the experimental situation that may have impeded Ss ability to profit from information in alternations. When we repeated three of Kornblum's conditions in a choice reaction experiment designed to give S more opportunity to use the information from alternations, the overall results were more compatible with the information hypothesis. This result suggest that a fruitful way to study the microstructure of choice RT is to isolate the factors that differentiate situations in which the information hypothesis holds from those in which it does not. © 1969.