Please use this identifier to cite or link to this item: https://doi.org/10.1016/S0925-2312(01)00701-9
Title: A computational model for the development of simple-cell receptive fields spanning the regimes before and after eye-opening
Authors: Rishikesh, N.
Venkatesh, Y.V. 
Keywords: Computational model
Correlation
Critical period
Orientation selectivity
Receptive field (RF) development
Selective rearing
Self-organization
Spontaneous neural activity
Visual experience
Issue Date: Jan-2003
Citation: Rishikesh, N., Venkatesh, Y.V. (2003-01). A computational model for the development of simple-cell receptive fields spanning the regimes before and after eye-opening. Neurocomputing 50 : 125-158. ScholarBank@NUS Repository. https://doi.org/10.1016/S0925-2312(01)00701-9
Abstract: Motivated by the possible existence of post-natal cortical plasticity, we analyze Miller's (J. Neurosci. 14 (1994) 409-441) correlation-based plasticity dynamics using sinusoidal patterns of varying frequency and orientation. After demonstrating that this leads to the formation of a cluttered receptive field (RF), and analyzing the reasons therefor, we propose a Kohonen-type, response-dependent modulation of Miller's dynamics. We analyze the simulation outputs-the RF profiles and preference maps-arising from changes in the model parameters. Further, in an attempt to quantify the hypothesis that (i) spontaneous activity and (ii) visual experience play prominent roles in the (a) establishment and (b) maturity of orientation selectivity, respectively, we initialize the plasticity dynamics with developing Miller-type RFs. We interpret such an initialization to form a combined pre-natal-post-natal model, and quantify the relative effects of spontaneous activity and visual experience on developing RFs and their preference organization. As a next step, we analyze a possible quantification of the critical period phenomenon in the proposed model, and discuss the biological implications of such a quantification. Further, we subject the model to selective rearing by presenting it with biased visual environments. By analyzing the results, and calibrating the output using its appropriate biological counterparts, we show that the model measures up to biological realities. We also fix bounds for certain model parameters by comparing the results with biological data. © 2002 Elsevier Science B.V. All rights reserved.
Source Title: Neurocomputing
URI: http://scholarbank.nus.edu.sg/handle/10635/68082
ISSN: 09252312
DOI: 10.1016/S0925-2312(01)00701-9
Appears in Collections:Staff Publications

Show full item record
Files in This Item:
There are no files associated with this item.

Google ScholarTM

Check

Altmetric


Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.