Effects of category length and strength on familiarity in recognition.

Abstract

In most recognition models a decision is based on a global measure often termed familiarity. However, a response criterion is free to vary across lists varying in length and strength, making familiarity changes immeasurable. We presented a single list with a mixture of exemplars from many categories, so that the criterion would be unlikely to vary with length or strength of the category of the test item. False alarms rose with category length but not category strength, suggesting that familiarity does not change much with changes in strength of other items but grows when additional items are studied. The results were well fit by an extension of the search of associative memory (SAM) model presented by R. M. Shiffrin, R. Ratcliff, and S. E. Clark (1990).

A Hybrid-Similarity Exemplar Model for Predicting Distinctiveness Effects in Perceptual Old-New Recognition.

“…In the present research, we started by operationalizing distinctiveness in terms of the degree of isolation of an object in the similarity space of other studied objects. In agreement with some past work (Shiffrin et al, 1995;Zaki & Nosofsky, 2001), in Experiment 1 of this article, we found that old items located in isolated regions of the similarity space were not recognized with higher probability than were more typical, densely located objects. Furthermore, the absence of such old-item distinctiveness effects was even observed in conditions in which highsimilarity items never served as foils for the distinctive targets.…”

“…First, averaged across all three experiments and conditions of testing, hit rates to old items from dense categories (typical old items) slightly exceeded hit rates to old items from sparse categories (distinctive old items; .780 vs. .756). Thus, as was found in the experiments reported by Zaki and Nosofsky (2001) and Shiffrin et al (1995), there was no overall evidence of a distinctiveness effect in which distinctive old items had higher hit rates than typical old items. The most important question concerned whether the relative hit rates to typical and distinctive old items depended on the structure of the testing conditions.…”

“…In Experiment 1, and previous experiments like it (Shiffrin et al, 1995;Zaki & Nosofsky, 2001), distinctive objects were defined simply as those objects located in isolated regions of the similarity space of studied items. However, such objects did not tend to possess individuating discrete properties that were not also possessed by typical objects located in dense regions.…”

“…If a highly distinctive face is presented at time of study, there are unlikely to be any foils that are similar to that face at time of test; whereas typical faces presented during study are likely to have highly similar foils at time of test. By contrast, in the designs conducted by Shiffrin et al (1995) and Zaki and Nosofsky (2001), different testing situations were used. In Zaki and Nosofsky's experiment, subjects were tested in a uniform manner with all colors from all regions, regardless of which particular colors were presented at time of study.…”

“…In the complete-set testing condition (analogous to the method used by Zaki & Nosofsky, 2001), subjects were tested on all items from all categories. In the fixed-subset testing condition (analogous to the method used by Shiffrin et al, 1995), subjects were tested on a single old item and either one or two new items from each category, regardless of whether the study category was dense or sparse. And in the correlated-subset testing condition, subjects were tested on all items from the dense categories (including both old and new items), but on only the single old item from the sparse categories.…”