Familiarity, Item recognition picture similarity experiment

Abstract

How familiar does an item or event have to be before it is thought recollected? We examined the relationships between familiarity, recollection, recognition accuracy, and confidence in a picture-similarity two-alternative forced-choice item recognition task using natural scene photographs as targets and lures. Participants rated their confidence in their recognition decision on a 0–100 scale, answered yes or no to whether they had recollected the picture identified as old, and finally rated how familiar the recognised image had seemed on a 0–100 scale. Decisions made when the target was recollected were significantly faster than decisions made in the absence of recollection. Critically, with no requirement for source memory reporting, recollection was significantly faster than familiarity unless familiarity was highest (≥ 76), when mean reaction times were not significantly different. Rather than assuming a dual-process explanation, and consistent with other research, if familiarity and recollection represent memory strength on a continuous scale, there must be a point where the memory strength of a recognition decision because an item is familiar equates to the memory strength of a recognition decision because an item is recollected. As familiarity increased, accuracy, the proportion of recollected trials, and confidence in the recognition decision also increased. Recollection improved the accuracy of recognition, but only when familiarity was low; if the target was recollected, increasing familiarity had no additional effect on accuracy, with no significant interaction effect between recollection and familiarity on recognition accuracy.