|
|
Project 6: Basic Mechanisms and Cooperating Systems in Learning Memory
This project seeks to foster the further development of a parallel-distributed processing approach to learning and memory through the incorporation of findings and insights from neurosciences (Center Aim 3), informed by a parallel-distributed processing perspective on the roles of different brain areas in information processing. There are three key tenets of the effort: (1) Learning and memory depend fundamentally on an essentially Hebbian principle of learning (Center Aim 2): The brain tends to strengthen whatever response it makes to given inputs. (2) This tendency to strengthen elicited responses can lead to progress in learning, but it can also lead to failures; and if it is left to operate without modulation or control, a mechanism of this sort may be too weak to lead to sophisticated cognitive competencies. This leads to the idea that acquisition of cognitive competencies in a Hebbian system depends on the modulation of the Hebbian learning process in target brain regions in response to task demands, outcome information, and constraining input from other brain regions, so that overall learning functions of the system as a whole emerge from cooperation among many brain areas (Center Aim 1). (3) Overt manifestations of successes and failures in learning or memory reflect both the basic mechanisms involved and their integration within a system of interacting brain structures. Three specific aims will be pursued: Aim 1: Phonological learning in adulthood. This work extends a simple preliminary model that captures aspects of phonological category learning in adulthood, in an effort to provide a more detailed and more biologically constrained account of some of the factors that contribute to successes and failures in the acquisition of new phonological distinctions in adult. The work will incorporate specific assumptions embodying the first and second tenets above to address data from behavioral and functional imaging experiments. Aim 2: Priming of perceptual processing in normal and amnesic populations. This part of the work will investigate patterns of findings seen in a range of tasks that have been used with normal and amnesic individuals, in hopes of further exploring the nature of the basic mechanisms underlying learning as well as clarifying the nature of the contributions of different brain regions in these types of tasks. Aim 3: Acquisition of semantic information with and without amnesia. This part of the work will be similar to the second, in that it will investigate patterns of success and failure in learning in both normal and amnesic persons, but within the context of tasks that appear to tap the acquisition of factual or semantic information.
Key Personnel:
| Name | Organization | Role on Project |
| James L. McClelland | Carnegie Mellon/CNBC | Principal Investigator |
| Julie A. Fiez | Univ. of Pittsburgh/CNBC | Consultant |
| Randall C. O'Reilly | Univ. of Colorado, Boulder | Consultant |
|
