Mechanisms of Associative Learning.

My research is aimed at the development of a quantitative, mechanistic theory of associative learning. The several models that have come from our laboratory (e.g., the so-called Rescorla-Wagner Model, SOP and AESOP) are primarily based upon behavioral data from well-characterized animal learning situations, including habituation, classical conditioning and instrumental learning, and are most germane to such circumstances. However, they are also intended to be in contact with available neurobiological data from various model systems (especially eyeblink conditioning in the rabbit) that are currently being exploited in investigation of the architectural and cellular basis of learning and memory. The neural-network form of the models has favored the incorporation of certain of their tenants into "connectionist" treatments of more complex human cognition. A major focus of our current research is on the contextual control of associative learning, as seen in retrieval modulation, conditional discrimination, and "occasion setting." A resulting thrust is toward the development of a context-sensitive, real-time model of associative learning.

Sample Publications

Wagner, A.R. & Donegan, N.H. (1989) Some relationships between a computational model (SOP) and a neural circuit for Pavlovian (rabbit eyeblink) conditioning. In, R.D. Hawkins and G.H. Bower (Eds.) The Psychology of Learning and Motivation: Vol 22 Computational Models of Learning in Simple Neural Systems. (pp. 157-203) Orlando: Academic Press.

Wagner, A.R. (2003) Context-sensitive elemental theory. The Quarterly Journal of Experimental Psychology. Vol 56B, 7-29.