Fig. 2: Single-neuron tuning correlates with input variance.
From: Cortical recurrence supports resilience to sensory variance in the primary visual cortex
Additional examples are shown in Supplementary Fig. 1. a Tuning curves of two neurons responding to Motion Clouds of increasing variance Bθ. Dots indicate the mean firing rate across trials (baseline subtracted, 300 ms average), error bars are the standard error and lines represent a fitted von Mises function. b Variance-tuning functions (VTF), relating the change of orientation tuning measured by the circular variance (CV, dots) as a function of input variance Bθ, fitted with a Naka-Rushton (NKR) function (dashed curves, parameters shown in light gray). Parameters of the VTF are \(\log (n)=8.4,{B}_{\theta 50}=14.{7}^{\circ },{f}_{0}=0.4\) for neuron A and \(\log (n)=2.4,{B}_{\theta 50}=35.{0}^{\circ },{f}_{0}=0.3\) for neuron B. The CV identity curve is shown in solid gray. c Histograms of the NKR parameters (in the [5%;95%] range of possible NKR fitting values) for the 249 recorded units. Median values are indicated by a black arrow (\(\log (n)=3.6,{B}_{\theta 50}=19.{2}^{\circ },{f}_{0}=0.75\)).
