Neurocomputing 44–46 (2002) 467–472 www.elsevier.com/locate/neucom Evolving eld models for inhibition eects in early vision Christian Igel a ; ∗ , Werner von Seelen a , Wolfram Erlhagen b , Dirk Jancke c a Institut f ur Neuroinformatik, Ruhr-Universit at Bochum, 44780 Bochum, Germany b Departamento de Matematica, Universidade do Minho, 4800 Guimar˜ aes, Portugal c Center for Research of Higher Brain Functions, The Weizmann Institute of Science, 76100 Rehovot, Israel Abstract We present a nonlinear eld model based on linear couplings for inhibition eects in early vision. The model is tted to data from single unit recordings in the primary visual cortex of the cat. We focus on the prominent eect that responses to second stimuli are amplied, reduced, or unaected depending on the temporal and spatial distance of the stimuli. The model is adjusted using an elaborated self-adaptive evolution strategy resulting in an accurate, easy to interpret, and well generalizing model. c 2002 Elsevier Science B.V. All rights reserved. Keywords: Neural elds; Evolutionary optimization; Population representation; Early vision; Wiener system 1. Introduction We propose evolutionary “analysis by synthesis” guided by neurobiological knowl- edge as a powerful tool in computational neuroscience. The challenge is to force ar- ticial evolution to generate solutions that are plausible from the biological point of view. Such solutions are only likely to evolve if as much neurobiological knowledge as possible is incorporated into the design process. This can be achieved by provid- ing sucient experimental data to evaluate the evolved systems and by a considerate choice of the basic structure. Additional background information can be coded in the tness function and in constraints that ensure biological plausibility. * Corresponding author. Tel.: +49-2343225558; fax: +49-2343214209. E-mail addresses: christian.igel@neuroinformatik.ruhr-uni-bochum.de (C. Igel), wolfram.erlhagen@ eng.uminho.pt (W. Erlhagen), dirk.jancke@weizmann.ac.il (D. Jancke). 0925-2312/02/$-see front matter c 2002 Elsevier Science B.V. All rights reserved. PII:S0925-2312(02)00402-2