Dynamic Normalization

Abstract The normalization model has been applied to explain neural activity in diverse neural systems including primary visual cortex (V1). The model’s defining characteristic is that the response of each neuron is divided by a factor that includes a weighted sum of activity of a pool of neurons. In spite of the success of the normalization model, there are 3 unresolved issues. 1) Experimental evidence supports the hypothesis that normalization in V1 operates via recurrent amplification, i.e., amplifying weak inputs more than strong inputs. It is unknown how nor-malization arises from recurrent amplification. 2) Experiments have demonstrated that normalization is weighted such that each weight specifies how one neuron contributes to another’s normalization pool. It is unknown how weighted normalization arises from a recurrent circuit. 3) Neural activity in V1 exhibits complex dynamics, including gamma oscillations, linked to normalization. It is unknown how these dynamics emerge from normalization. Here, a new family of recurrent circuit models is reported, each of which comprises coupled neural integrators to implement normalization via recurrent amplification with arbitrary normalization weights, some of which can reca-pitulate key experimental observations of the dynamics of neural activity in V1. Significance Statement A family of recurrent circuit models is proposed to explain the dynamics of neural activity in primary visual cortex (V1). Each of the models in this family exhibits steady state output responses that are already known to fit a wide range of experimental data from diverse neural systems. These models can recapitulate the complex dynamics of V1 activity, including oscillations (so-called gamma oscillations, ∼30-80 Hz). This theoretical framework may also be used to explain key aspects of working memory and motor control. Consequently, the same circuit architecture is applicable to a variety of neural systems, and V1 can be used as a model system for understanding the neural computations in many brain areas.