direkt zum Inhalt springen

direkt zum Hauptnavigationsmenü

Sie sind hier

TU Berlin

Page Content

All Publications

Simultaneous Stabilization of On-Off and Ocular Dominance Patches by Modeled Unspecific LTP
Citation key Stetter1998b
Author Stetter, M. and Lang, E.W., and Weih, P. and Obermayer, K.
Title of Book European J. Neurosc, 10(S10):pg 332
Year 1998
Publisher European Forum of Neuroscience ENA
Abstract We analyze the conditions, under which both elongated ON- and OFF-patches of cortical receptive fields and ocular dominance (OD) are stabilized simultaneously within a correlation-based learning (CBL) model. For this we consider a feed forward-neural network with four input layers modeling ON- and OFF-subpopulations of left-eye and right-eye-driven LGN-neurons, and one output layer as a model area 17. LGN-activities are assumed to be purely generated by mexican-hat-filtered spontaneous photoreceptor activities of both retinae, i.e. no LGN-circuitry is assumed. For a single linear output neuron we show analytically, that plain Hebbian learning cannot stabilize OD and ON-OFF-segregation at the same time. Proceeding from that, we include the recently observed unspecific LTP into the learning rule, where Hebbian LTP of a given synapse evokes synaptic modification of neighboring synapses irrespective of their history (F. Engert and T. Bonhoeffer, Nature 388:279-284 (1997)). It is shown analytically that unspecific LTP stabilizes both OD and ON-OFF-patches simultaneously, as soon as left-eye and right-eye driven synapses tend to form groups on the postsynaptic neuron. Numerical simulations using nonlinear, iterated networks confirm the analytical results in showing, that unspecific LTP and synaptic grouping evoke stable OD and ON-OFF patches, while Hebbian LTP alone fails to stabilize both structures at the same time. We conclude that details of synaptic plasticity on a cellular level may considerably influence the prediction of developmental models on a systems level pointing towards the requirement of more detailed models of synaptic plasticity.
Download Bibtex entry

Zusatzinformationen / Extras

Quick Access:

Schnellnavigation zur Seite über Nummerneingabe

Auxiliary Functions