Abstract
Purpose: :
Given the large input range under which the visual system must operate, the retina must adjust its light sensitivity to indicate deviations in luminance level despite its narrow spiking range. Prior research has explored the effects of luminance and contrast gain mechanisms on the sensitivity of retinal ganglion cell responses. Our research uses a linear-nonlinear model of neural firing to validate the changes in response sensitivity to luminance changes in retinal ganglion cells.
Methods: :
We investigated luminance and contrast adaptation mechanisms and their relationship to retinal sensitivity changes in anesthetized rat optic tract using a binary white noise paradigm. The white noise was presented with and without sinusoidal modulations, the latter of which was implemented with varying temporal frequencies. We then calculated the spike-triggered average (STA) stimulus, estimating the linear dynamics of the system, and mapped any residuals to a nonlinear filter.
Results: :
Retinal ganglion cells show luminance adaptation when binary white noise sequences were presented with sinusoidal modulation. Using spikes fired at different phases of the sinusoid, we constructed different linear-nonlinear models, each with their own STA and static nonlinearity. By scaling the non-linear stages of the system and re-scaling the STA amplitudes accordingly, we observed deviations from a baseline (non-modulated) binary noise model; specifically, amplitudes were increased when the modulations brought the mean luminance below baseline, and were decreased when the modulations brought the mean luminance above baseline. Increases in the temporal frequency of the sinusoidal modulation caused these amplitudes to decrease and increase, respectively.
Conclusions: :
Our findings validated previous experimental results. Probing a cell with sinusoidally-modulated binary white noise activated a rapid timescale (<100ms) luminance gain control that decreases response amplitude in the presence of a relatively high mean background luminance and vice versa. This suggests that retinal sensitivity to luminance shapes ganglion cell spiking as the eyes absorb visual information.
Keywords: electrophysiology: non-clinical • ganglion cells