Document Type


Degree Name

Master of Arts (MA)




Faculty of Science

First Advisor

Donald Morgenson

Advisor Role

Thesis Supervisor


The aim of the present research was to examine the nature of the memory code that contributes to performance in a serial feature-positive discrimination. In order to test the hypothesis that a reinforce expectancy based on the first-order association between feature element and reinforce defines the content of the code, different groups of rats received various forms of pretraining involving either the feature element or the common element, or both. For some groups the feature element was trained as a CS+, while for others it was trained as a CS-. In addition, some groups received training establishing the common element as a CS+. The effects of these forms of element pretraining on the development of a serial feature-positive discrimination were then examined. Behavioral observations of the animals suggested that the existence of a first-order association between either the feature element of the common element and reinforcement resulted in a decrement in serial feature-positive discrimination performance, a result that was incompatible with the original hypothesis. The finding was interpreted in terms of a modified memory code that may involve a CS-specific associative learning component.

Convocation Year


Convocation Season