Visual attention using spiking neural maps

Visual attention is a mechanism that biological systems have developed to reduce the large amount of visual information in order to efficiently perform tasks such as learning, recognition, tracking, etc. In this paper, we describe a simple spiking neural network model that is able to detect, focus on and track a stimulus even in the presence of noise or distracters. Instead of using a regular rate-coding neuron model based on the continuum neural field theory (CNFT), we propose to use a time-based code by means of a network composed of leaky integrate-and-fire (LIF) neurons. The proposal is experimentally compared against the usual CNFT-based model.

[1]  D. Spalding.  The Principles of Psychology , 1873, Nature.

[2]  Gerben G. Meyer,et al.  ANTICIPATORY BEHAVIOR IN ADAPTIVE LEARNING SYSTEMS: FROM BRAINS TO INDIVIDUAL AND SOCIAL BEHAVIOR , 2007 .

[3]  Nicolas P. Rougier,et al.  Synchronous and asynchronous evaluation of dynamic neural fields , 2011 .

[4]  Bernard Girau,et al.  Massively distributed digital implementation of an integrate-and-fire LEGION network for visual scene segmentation , 2007, Neurocomputing.

[5]  L. Busse,et al.  The spread of attention across modalities and space in a multisensory object. , 2005, Proceedings of the National Academy of Sciences of the United States of America.

[6]  Claudio Castellanos Sánchez,et al.  On-chip visual perception of motion: A bio-inspired connectionist model on FPGA , 2005, Neural Networks.

[7]  Sylvain Chevallier,et al.  Visual focus with spiking neurons , 2008, ESANN.

[8]  Bernard Girau,et al.  Bio-inspired visual sequences classification , 2010, BICS 2010.

[9]  Leslie G. Ungerleider.  Two cortical visual systems , 1982 .

[10]  A. Treisman,et al.  A feature-integration theory of attention , 1980, Cognitive Psychology.

[11]  Sylvain Chevallier,et al.  Covert Attention with a Spiking Neural Network , 2008, ICVS.

[12]  C. Koch,et al.  Computational modelling of visual attention , 2001, Nature Reviews Neuroscience.

[13]  J. G. Taylor,et al.  Neural ‘bubble’ dynamics in two dimensions: foundations , 1999, Biological Cybernetics.

[14]  B. Cessac.  A discrete time neural network model with spiking neurons , 2007, Journal of mathematical biology.

[15]  John H. R. Maunsell,et al.  Feature-based attention in visual cortex , 2006, Trends in Neurosciences.

[16]  Martin V. Butz,et al.  Anticipatory Behavior in Adaptive Learning Systems, From Brains to Individual and Social Behavior [the book is a result from the third workshop on anticipatory behavior in adaptive learning systems, ABiALS 2006, Rome, Italy, September 30, 2006, colocated with SAB 2006] , 2007, ABiALS book.

[17]  Nicolas P. Rougier,et al.  Emergence of attention within a neural population , 2006, Neural Networks.

[18]  S Ullman,et al.  Shifts in selective visual attention: towards the underlying neural circuitry. , 1985, Human neurobiology.

[19]  Jérémy Fix,et al.  A Dynamic Neural Field Approach to the Covert and Overt Deployment of Spatial Attention , 2011, Cognitive Computation.

[20]  Jean-Charles Quinton,et al.  Exploring and optimizing dynamic neural fields parameters using Genetic Algorithms , 2010, The 2010 International Joint Conference on Neural Networks (IJCNN).

[21]  S. Amari.  Dynamics of pattern formation in lateral-inhibition type neural fields , 1977, Biological Cybernetics.