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

Abstract

Multiple adjacent, roughly mirror-image topographic maps are commonly observed in the sensory neocortex of many species. The cortical regions occupied by these maps are generally believed to be determined initially by genetically-controlled chemical markers during development, with thalamocortical afferent activity subsequently exerting a progressively increasing influence over time.Here we use a computational model to show that adjacent topographic maps with mirror-image symmetry can arise from activity dependent synaptic changes whenever the distribution radius of afferents sufficiently exceeds that of horizontal intracortical interactions. Which map edges become adjacent is strongly influenced by the probability distribution of input stimuli during map formation. Our results suggest that activity-dependent synaptic changes may play a role in influencing how adjacent maps become oriented following the initial establishment of cortical areas via geneticallydetermined chemical markers. Further, the model unexpectedly predicts the occasional occurrence of adjacent maps with a different rotational symmetry. We speculate that such atypically oriented maps, in the context of otherwise normally interconnected cortical regions, might contribute to abnormal cortical information processing in some neurodevelopmental disorders.2

An unsupervised learning method for representing simple sentences

“…Earlier work established a set of parameter values (many of which we borrow directly) suitable for learning to produce unique activation patterns for differing sequences of phonemes representing single words. This past work also demonstrated that a trained layer can become a topologically clustered map with respect to its phoneme inputs [2], and that mirror-image topographical maps also occur naturally in the model [6], which provides convincing support for the analogy between this type of neural network and biological cortex.…”

“…They are thus recurrent neural networks that learn via competitive, unsupervised Hebbian training rules (described below). This type of layer has been found to mimic some organizational properties of biological cortex [1], [6].…”

Plasticity-Induced Symmetry Relationships Between Adjacent Self-Organizing Topographic Maps

“…We adopted the same multiwinner self-organizing map architecture and methodology used in the earlier study for maps occurring along one dimension (Schulz & Reggia, 2005), but now modified to permit maps to occur along two dimensions. The output or cortical layer nodes over which map formation occurs are arranged in a regular rectangular grid.…”

“…Mirror-symmetric topographic maps are common in the neocortices of many species (Schulz & Reggia, 2005), but it remains unknown to what extent such symmetries are due to genetic encoding versus thalamocortical activity and synaptic plasticity. Multiple theories have been put forward, the most widely accepted being that the initial partitioning of the cortex into different areas during development is primarily due to genetically determined chemical markers (Levitt, 2000;Sur & Leamey, 2001;Zhou & Black, 2000).…”

“…Many experimental studies have demonstrated remarkable plasticity in cortical maps due to altered probabilities of sensory stimuli, peripheral nerve lesions, or cortical damage (Buonomano & Merzenich, 1998;Pantev et al, 1998;Raineteau & Schwab, 2001). Recent modeling work established that such mirror-symmetric maps can arise along a single dimension due to activitydependent synaptic changes, leading to linear sequences of pairwise mirror-symmetric maps, each having one or two neighbor maps (Schulz & Reggia, 2005).…”