Gain modulation of recurrent networks

Gain modulation is an important mechanism by which attentional and other inputs modify the amplitude of neuronal responses without changing their selectivity. Gain modulation has been studied previously in feedforward circuits but not in recurrent neural networks. We show how gain modulation modi"es the response of a recurrent network to feedforward inputs. Even modest gain modulation of the recurrent network can cause downstream neurons to switch from a state in which they are unresponsive to a stimulus to a state where they respond selectively. Funneling the recurrent connections of a network through gain modulated neurons allows the selectivity within the network to be modi"ed by modulatory inputs. ( 2000 Elsevier Science B.V. All rights reserved.