Nature of the Representation for Picture-Recognition Memory

Abstract

When distracting pictures are randomly selected from the same pool of pictures as the target pictures, performance with large numbers of pictures (1,000) is much poorer than has previously been reported. Further, performance drops quite substantially after 1 wk., a finding that differs from the conclusion that picture memory is relatively stable over time. Under conditions where the distractors were selected to reduce similarity between targets and distracting pictures, performance approximated levels previously reported in the literature. The results of the experiments seem to demand some type of categorization model of picture-recognition memory.