Analysis of short-term retention: Models for Brown-Peterson experiments.

Abstract

A family of models, based on a weighting function that decreases as the reciprocal of elapsed time, is developed for the principal Brown-Peterson experiments. The models accommodate, within the limits of experimental error, the decrease in recall with the passage of time, the effects of covert rehearsals, of proactive interference, and of repeated presentations of the stimulus, and (but with less accuracy) the rate of responding when latencies are recorded. It is concluded that (a) the accessibility of memories decreases as the reciprocal of elapsed time up to, at least, 27 s; (b) failure to recall is almost entirely due to proactive interference; (c) repeated presentations of the stimulus create distinct records in memory, each with its separate weight; and (d) the reciprocal weighting function simultaneaously determines both the distribution of recall latencies and the cumulative proportion of correct recalls ultimately produced