Faster is better: Visual responses to motion are stronger for higher refresh rates
Although neural responses with a millisecond precision were reported in the retina, lateral geniculate nucleus and visual cortex of multiple species, the presence and role of such a fine temporal structure is still debated at the cortical level and the general belief remains that early visual system encodes information at slower timescales. In this study, we used a new stimulation platform to generate visual stimuli that were very precisely encoded in time and we characterized in human subjects the EEG responses to moving patterns that shared the same global motion but differed in their fine scale spatio-temporal properties. In two experiments, we manipulated the information within temporal windows that corresponded to the frame duration in conventional (1/60 = 16.7ms, experience 1) and more recent (1/120 = 8.3ms, experience 2) visual displays. Our results demonstrate that EEG responses to temporally dense and coherent trajectories are significantly stronger than those to control conditions without these properties. They extend previous results from our group that showed that accurate temporal information (<10ms) significantly improve perceptual judgments on spatial discrimination, digit recognition and sensitivity for speed [Kime et al., 2016]. Altogether, our results suggest that instead of low-pass filtering the temporal information it receives from its thalamic afferents, the visual cortex may actually exploit its richness to improve visual perception.\n\nSignificance statementMost visual experiments use frame-based stimulation screens which are synchronous and slow. This might not engage the full encoding capacity of the visual system, as in natural condition, visual information is asynchronously acquired by the retina with a very high temporal resolution. In the present study, we used an unconventional and highly configurable stimulation platform to determine the EEG responses to moving patterns that shared the same global motion but differed in their fine scale (<10ms) spatio-temporal properties. Our results demonstrate that brain responses in human are significantly stronger for motion patterns that are refreshed at very high frame rate (360Hz) and thereby provide a strong support for the use of very precise temporal stimulations in vision studies.