Adaptation to Changes in Higher-Order Stimulus Statistics in the Salamander Retina

Gašper Tkačik,Anandamohan Ghosh,Ronen Segev,Elad Schneidman,Sanjoy Bhattacharya

doi:10.1371/journal.pone.0085841

Abstract

Adaptation in the retina is thought to optimize the encoding of natural light signals into sequences of spikes sent to the brain. While adaptive changes in retinal processing to the variations of the mean luminance level and second-order stimulus statistics have been documented before, no such measurements have been performed when higher-order moments of the light distribution change. We therefore measured the ganglion cell responses in the tiger salamander retina to controlled changes in the second (contrast), third (skew) and fourth (kurtosis) moments of the light intensity distribution of spatially uniform temporally independent stimuli. The skew and kurtosis of the stimuli were chosen to cover the range observed in natural scenes. We quantified adaptation in ganglion cells by studying linear-nonlinear models that capture well the retinal encoding properties across all stimuli. We found that the encoding properties of retinal ganglion cells change only marginally when higher-order statistics change, compared to the changes observed in response to the variation in contrast. By analyzing optimal coding in LN-type models, we showed that neurons can maintain a high information rate without large dynamic adaptation to changes in skew or kurtosis. This is because, for uncorrelated stimuli, spatio-temporal summation within the receptive field averages away non-gaussian aspects of the light intensity distribution.

Highlights

Adaptation is ubiquitous in the nervous system, from synaptic depression [1,2] and single neuron spiking [3,4], to the activity of neural modules (e.g. [5])
The authors reported that contrast-gain control responds to spatial root-mean-square contrast but not to the higher moments in the pixel luminance distribution. These results raised a number of important questions that we address here: (i) are there any signatures of adaptation to higher-order statistics, especially if spatially uniform stimuli that match the naturalistic range of skew/kurtosis are used instead of the spatial textures, which cannot accommodate the same effective range of skewness/kurtosis values; (ii) do changes in higher-order stimulus statistics affect the cells’ rate of information coding; and (iii) what would be theoretically expected changes for LNtype neurons in response to changes in higher-order stimulus statistics if the neurons were maximizing the amount of transmitted information
To characterize how the retina encodes higher-order statistics (HOS) of the luminance distribution, we presented it with a set of 9 synthetic spatially homogenous stimuli S, where the light intensity of each stimulus frame was drawn independently from distributions PS(L) that were matched in mean L

Summary

Introduction

Adaptation is ubiquitous in the nervous system, from synaptic depression [1,2] and single neuron spiking [3,4], to the activity of neural modules (e.g. [5]). The retina is one of the most studied highly adaptive neural circuits, in which the mapping between stimuli and neural response changes to match the statistics of the mean light intensity [11], temporal and spatial contrast and spatial scale [12,13,14], pattern [15], relative motion [16] and periodicity [17]. It has been further shown that neural systems adapt to various stationary stimuli, and to dynamic changes in stimulus distributions taking place across multiple timescales [23,24,25]

Methods

Results

Conclusion