Purchase this article with an account.
VM Zemon, J Gordon; Luminance Contrast Mechanisms In Human Vision: Visual Evoked Potentials (VEPs) and A Nonlinear Model . Invest. Ophthalmol. Vis. Sci. 2002;43(13):3933.
Download citation file:
© ARVO (1962-2015); The Authors (2016-present)
Purpose: To investigate the ability of nonlinear mechanisms to explain the VEP functions obtained through manipulations of contrast parameters that are designed to emphasize activity from either magnocellular (M) or parvocellular (P) pathways and their ON or OFF subdivisions. Methods: Luminance of isolated-check patterns were modulated sinusoidally at ∼ 6 Hz under two types of contrast conditions: appearance-disappearance and pedestal. These conditions were presented in both one-minute and swept-parameter runs. Two contrast parameters were manipulated: mean contrast (Cmean) and depth of modulation (DOM). In appearance-disappearance conditions, Cmean=DOM to emphasize M contributions, and in pedestal conditions, Cmean is fixed at a high level and DOM is varied to emphasize P contributions to the VEP. Arrays of bright or dark checks were used with each condition to elicit responses from respective ON and OFF subdivisions of M and P pathways. Fourier analysis was applied to each epoch of EEG and a multivariate statistic was applied to the fundamental frequency component in each set of epochs to derive noise estimates. A model using a simplex algorithm and a least-squares criterion was used to fit amplitude and phase data. In this model, an inhibitory signal (linearly related to DOM) produces a proportional change in shunting conductance. This results in a change in the system's time constant. A threshold parameter is included, as is half-wave rectification, which serves to isolate activity from ON and OFF pathways. Results: Under appearance-disappearance conditions, the model yielded phase advance and compressive amplitude functions with increases in DOM, and under pedestal conditions, it yielded flat phase (or phase lag) and linear amplitude functions. Under both of these conditions, for small checks, dark-check stimuli yielded greater contrast gains than did bright-check stimuli. Conclusion: The model captured the data accurately, and the parameter values obtained under appearance-disappearance and pedestal conditions agreed with those expected from M and P pathways, respectively.
This PDF is available to Subscribers Only