Cortical Map Reorganization as a Competitive Process

Abstract

Past models of somatosensory cortex have successfully demonstrated map formation and subsequent map reorganization following localized repetitive stimuli or deafferentation. They provide an impressive demonstration that fairly simple assumptions about cortical connectivity and synaptic plasticity can account for several observations concerning cortical maps. However, past models have not successfully demonstrated spontaneous map reorganization following cortical lesions. Recently, an assumption universally used in these and other cortex models, that peristimulus inhibition is due solely to horizontal intracortical inhibitory connections, has been questioned and an additional mechanism, the competitive distribution of activity, has been proposed. We implemented a computational model of somatosensory cortex based on competitive distribution of activity. This model exhibits spontaneous map reorganization in response to a cortical lesion, going through a two-phase reorganization process. These results make a testable prediction that can be used to experimentally support or refute part of the competitive distribution hypothesis, and may lead to practically useful computational models of recovery following stroke.

Success and failure of new speech category learning in adulthood: Consequences of learned Hebbian attractors in topographic maps

“…Moreover, objects in the world are cohesive along dimensions such as time, space, and frequency. A topographic map "projects" this cohesiveness directly into the perceptual system and can thereby guide its development, as demonstrated by models of the maturation of the early visual pathways (Linsker, 1986;Sirosh & Miikkulainen, 1997) and by models of somatotopic map reorganization (Grajski & Merzenich, 1990;Sutton, Reggia, Armentrout, & D'Autrechy, 1994). …”

Mirror Symmetric Topographic Maps Can Arise from Activity-Dependent Synaptic Changes

“…Multi-winner SOMs have been used in past, more biologically realistic and more complex and computationally expensive cortical models (von der Malsburg, 1973;Sutton et al, 1994;Sirosh and Miikkulainen, 1994). However, these previous multi-winner SOMs typically do not compute their output using Kohonen's algorithm (or a generalization thereof, like we do here).…”

“…First, the net input h to each cortical node i is computed as h i = w Cho and Reggia, 1994;Pearson et al, 1987;Reggia et al, 1992;Sutton et al, 1994;von der Malsburg, 1973) by a one-step selection of winners. However, unlike in the standard SOM, multiple winners occur where each cortical node i which receives a net input greater than that to each of the N neighboring cortical nodes closest to i (ties resolved arbitrarily) is taken to be a winner.…”

“…This result raises the possibility that activity-dependent synaptic changes, which are generally assumed to be a central force in individual topographic map selforganization during development (Kohonen, 2001a;Ritter et al, 1992;Sutton et al, 1994), may play a more important role in forming the orientations of adjacent cortical maps than is currently recognized. At the least, our results indicate the existence of an influence on the relative orientations of adjacent maps that may have affected the evolution of their genetically-guided afferent connectivity.…”

“…There has been less speculation as to why such maps often exhibit reflection symmetry, and the relative contributions during map formation of activity-dependent versus activity-independent mechanisms remain the source of some debate, even for individual maps (Grove and Tomomi, 2003;Cohen-Cory, 2002). Past computational models of self-organizing neocortical topographic maps (Kohonen, 2001a;Ritter et al, 1992;Sutton et al, 1994;Sirosh and Miikkulainen, 1994) have generally been limited to single maps and thus do not shed substantial light on the issue of forming multiple mirror-image maps.…”