Modelling Contrast Adaptation and Contextual Effects in Primary Visual Cortex

Orientation selective cortical neurons show contrast-invariant orientation tuning curves, contrast saturation and contrast adaptation. Additionally, their responses are modulated by oriented stimuli outside the classical receptive eld. We characterize the extent, to which these eeects can be caused by a recurrent network of simple neurons, and which properties are probably due to plasticity of geniculocortical synapses. For this we compare the predictions of a mean-eld-model based on cortical hypercolumns with those of a model that incorporates spiking neurons with detailed time-resolved synap-tic dynamics. We nd that in a mean-eld model contrast-invariant orientation tuning and contextual eeects can be generated by local and long-range intracortical circuitry respectively. However, they are incompatible with contrast-saturation and contrast-adaptation. Using a recurrent neu-ral network of excitatory spiking neurons with adapting synapses, we then show that within that model framework (i) contrast saturation can be explained by fast adaptation of the lateral or the aaerent synapses, (ii) phase advance with increasing contrast is due to fast adaptation of the aaerent synapses, and (iii) adaptation of the synaptic transmitter release probability of the geniculocortical synapses is a necessary and suucient mechanism for contrast adaptation. In summary, we suggest that recurrent lateral connections play a key role in orientation processing, while contrast processing is mainly a feed-forward phenomenon.