Abstract
The early visual evoked potentials (VEPs) used as indices of early visual processing are traditionally parsed into three components, namely the C1, P1, and N1, with peak latencies between 60-90ms, 100-130ms, and 150-180ms respectively (all measured via scalp electrodes in the central occipito-parietal region). The polarity of the C1 component’s peak is dependent on whether the experimental stimuli are presented peripherally in the upper or lower visual field, but has been argued to be negative for stimuli presented foveally. However, preliminary work from our group has shown a sizable peak latency difference (~30ms) between the C1 for peripheral stimuli compared to the C1 measured foveally, suggesting that the two may arise from different cortical processes. Here, we sought to provide more conclusive support for a functional separation of the C1 (relative to itself and the subsequent P1 and N1) by measuring VEP component tuning functions for spatial frequency (SF), using foveal and peripheral stimuli. The stimuli were high contrast Gabors (2.5 degrees in diameter) at 6 peak SFs (0.5, 1.0, 2.0, 4.0, 8.0, 12.0 cpd). During EEG recording, stimuli were presented for 50ms and followed by a spatial attention distractor task. Participants maintained central fixation throughout the experiment while the stimuli were presented at foveal (0 degrees eccentricity) and peripheral (6 degrees eccentricity: upper left or right, lower left or right) visual field locations. The results show a clear functional separation between the peripheral C1 and the foveal C1, with the former exhibiting no clear SF preference, while the latter exhibited a strong and systematic preference for increasing SF. Such a separation calls into question recent attempts to characterize general C1 processes measured solely at fixation, and argues for a careful re-examination of the C1 as it pertains to foveal or peripheral stimulus locations.
Meeting abstract presented at VSS 2015