Abstract
Human stereopsis has relatively poor spatio-temporal resolution, due to the way disparity is encoded in V1 [1], [2]. We are investigating this with EEG. Our stimuli are dynamic random-dot stereograms. At trial onset, they depict a zero-disparity plane, but after a random latency of at least 1s, they change into a horizontal disparity square-wave grating, whose disparity reversed sign with temporal frequency TF. We randomly interleaved trials with different disparity amplitude D (0.1,0.2,0.3,0.4 deg), spatial frequency SF (0, 0.25, 0.5,1 cpd) and temporal frequency TF (2.1, 4.2, 8.3, 12.5 Hz) . After a random time between 5 and 6s, the screen was blanked and participants were asked to report whether they had perceived the grating (i.e. the disparity alternation), or whether they had perceived two solid planes. They were most likely to perceive the grating at intermediate SFs and low TFs. We measured the ERP at the Oz electrode when transitioning from the zero-disparity plane into the depth grating. We calculated the 'ERP magnitude' by integrating the absolute value of the EEG trace relative to baseline, from 0 to 1.5 seconds from the transition. Pooling from 44 sets of data, we found that the ERP magnitude is higher for stimulus parameters where participants perceive the grating (r=0.5244, p< 0.0001, Pearson correlation between ERP and probability of seeing grating, N=64 conditions). This may reflect the visual system's use of disparity cue for structure detection. [1] D. Kane, P. Guan, and M. S. Banks, "The limits of human stereopsis in space and time," J Neurosci, vol. 34, no. 4, pp. 1397–1408, 2014. [2] H. Nienborg, H. Bridge, A. J. Parker, and B. G. Cumming, "Neuronal computation of disparity in V1 limits temporal resolution for detecting disparity modulation," J Neurosci, vol. 25, no. 44, pp. 10207–10219, 2005.
Meeting abstract presented at VSS 2016