Interacting Brain Modules for Memory: An Adaptive Representations Architecture; Final technical rept. Sep 2005-Jan 2008

摘要

An innovative and novel biologically-based computational model of interacting brain modules for memory, using the adaptive representations architecture of Gluck & Myers (1993; see also, 2001, Gateway to Memory: An Introduction to Neural Network Models of the Hippocampus and Learning, MIT Press) has been developed. The approach began with a connectionist-level architecture for the hippocampal region (medial temporal lobe) as a central system for creating optimal and adaptive stimulus representations, and then worked outwards from the hippocampal region to the brain systems that it modulates, including the cerebellum, cerebral cortex, basal ganglia, as well as other structures which, themselves, reciprocally modulate the hippocampus (ventral tegmental area/VTA, medial septum of the basal forebrain). Ultimately, this defined a novel biologically-inspired and constrained architecture for the neural substrates of a broad range of learning and memory behaviors and capabilities.