Homeostatic plasticity and ocular dominance diversity can account for the differential expansion of the left- and right-eye receptive fields of V1 neurons after monocular retinal lesions
conference contribution
posted on 2023-05-24, 13:10authored byYoung, JM, Wimmer, K, Calford, MB, Obermayer, K
When V1 neurons are subjected to a loss of subcortically-mediated feedforward input they respond by expanding their receptive fields, apparently due to an increase in the efficacy of the input arriving via their intracortical connections. This is likely to reflect a generalised strategy that allows neurons to recover from deficiencies in synaptic drive by uniformly potentiating subthreshold input connections to suprathreshold levels at which these connections can engage in activity-dependent Hebbian competition. It is plausible that this sub- to supra-threshold transformation is achieved by some kind of homeostatic modulation of gain, where the efficacies of all synapses are scaled such that excitatory neurons become more responsive and inhibitory neurons become less responsive to input (effectively producing disinhibition). However, after being partially deafferented by a monocular retinal lesion, binocular neurons in V1 appear to show receptive field expansion of their lesioned eye receptive fields only. Such a bias appears to be inconsistent with the hypothesis that a non synapse-specific form of homeostatic plasticity underlies this receptive field expansion. In this study we examined experimental results from monocular lesion experiments and compared them to the behaviour of a network model of V1. The feedforward and horizontal connectivity of each modeled neuron population (or ‘column’) had a specific ocular bias, and the distribution of ocular dominance within the network was based on in vivo data. We found that if the modeled columns underwent homeostatic plasticity at an intensity proportional to their input loss the distribution of the ratios of lesioned eye to non-lesioned eye receptive field expansion matched the distributions observed in vivo. Our results support the hypothesis that neurons are able to undergo homeostatic plasticity that is proportional to changes in their total synaptic input. In addition, the results indicate that the spatial specificity of this plasticity is at least below that of the scale of ocular dominance columns.