Our long- term goal is to develop comprehensive theories of learning which unify currently disparate approaches to the behavioral and neural analyses of learning and memory. In particular, we seek a functional understanding of the hippocampal region in both animal and human learning. As a technique for evaluating theories of brain function, we use computer models of artificial neural networks. These network models provide a framework for identifying theoretical correspondences between animal and human learning behaviors and between behavioral and biological levels of explanation (see Gluck & Granger, 1993, """"""""Computational models of the neural bases of learning and memory,"""""""" Annual Review of Neuroscience) The basis of much of our theoretical work is a computational theory of hippocampal region function in associative learning recently proposed by the Principal Investigator (Gluck & Myers, 1993) in which the hippocampal region is assumed to develop novel stimulus representations which compress and differentiate stimuli based on both stimulus-stimulus redundancies and stimulus-outcome predictiveness. Other brain regions, including cerebral and cerebellar cortices, are presumed to use these hippocampal-dependent representations to recode their own stimulus representations for long-term memory storage. In the absence of a functioning hippocampal region, these other brain regions are presumed to learn based only on their preexisting (and now fixed) representations. Computational modeling has shown that this theory accounts for a wide range of classical conditioning behaviors in intact and hippocampal-lesioned animals (Gluck & Myers, 1993; Myers & Gluck, 1994/ in press). The work in our lab currently includes both experimental studies of normal and hippocampal-impaired humans as well as theoretical modeling studies of human cognition and the neural substrates of animal learning. MBRS students will play important roles at each phases. Both undergraduate and graduate MBRS students will be involved in neuropsychological testing of memory function in both normal control populations as well as with patients with memory disorders due to stroke or Alzheimer's Disease. Although we are not now directly running any animal studies ourselves, we are collaborating on animal studies with other labs around the country; in the near future, however, we anticipate that we will set up our own animal behavior and lesioning lab at Rutgers, Students in the MBRS program will thus be able to learn valuable animal experimentation skills that have broad biomedical application.
Showing the most recent 10 out of 15 publications
