Effects of two retention interval components on performance differences between study-test and anticipation methods in paired-associate learning

Given that the retention interval consists of two qualitatively different and operationally distinguishable components, both separate and joint effects of them were necessary to assess their relative dominance in controlling both significant and nonsignificant performance differences between anticipation and study (reinforcement)-test methods; thus illuminating results thought inconsistent for about two decades. Three paired-associate learning experiments, with a total of 167 university students, were conducted with both massed (no intercycle or Type I interval) and spaced (30 sec Type 1 intervals) practice within each experiment. Both significantly and nonsignificantly superior performance resulted for the study-test method vis-à-vis the anticipation method with massed practice without practice. Findings support the retention interval hypothesis (Izawa, 1972) which is further strengthened when augmented with the capacity to account for the absolute performance differences between methods.