Mallory Gleason

University Honors and Honors in Psychology

Majors: Psychology and Biology

Supervisor: Katherine Bruce, Psychology

Effect of Exemplar Set Size on Identity Concept Learning in Rats Using an Olfactory Match-to-Sample Procedure

Abstract concepts are rules based on relations between stimuli divorced from any particular instance, rather than individual features of the stimuli. The ability to form such concepts is therefore higher order, and is undoubtedly adaptive in significance. A commonly used tool to train and test concepts is the Match-to-sample (MTS) procedure, specifically, the concept of “identity” or “sameness”. One feature of the MTS procedure is the number of example stimuli used during training. In previous studies, systematically increasing the set size resulted in greater accuracy when generalized to novel samples, the commonly held indication of concept mastery. Additionally, stimulus modality is an important feature to demonstrate concepts in specific species. In the current procedure, we explored the effect of set size on rat’s ability to learn identity using olfactory stimuli. Rats were trained in MTS with either 2 (n=2) or 10 stimuli (n=4). After reaching a certain accuracy criterion in training, subjects were tested for concept learning by their ability to generalize to 10 novel stimuli. Rats trained with 10 stimuli performed significantly better at this task than rats in the 2 stimuli group. Subsequently, all subjects were trained with the10 stimuli used in the probe and re-tested with 10 more novel exemplars. Performance on the second probe test significantly increased in the 2 stimuli group, suggesting that multiple exemplars may facilitate learning. Accuracy remained high in the group trained with 10 stimuli. Results were consistent with previous experimental findings on the effect of expanding the set size, and may add to the growing body of data about the underlying mechanisms of concept learning.