Time-compressed preplay of anticipated events in human primary visual cortex

Abstract

Perception is guided by the anticipation of future events. It has been hypothesized that this process may be implemented by pattern completion in early visual cortex, in which a stimulus sequence is recreated after only a subset of the visual input is provided. Here we test this hypothesis using ultra-fast functional magnetic resonance imaging to measure BOLD activity at precisely defined receptive field locations in visual cortex (V1) of human volunteers. We find that after familiarizing subjects with a spatial sequence, flashing only the starting point of the sequence triggers an activity wave in V1 that resembles the full stimulus sequence. This preplay activity is temporally compressed compared to the actual stimulus sequence and remains present even when attention is diverted from the stimulus sequence. Preplay might therefore constitute an automatic prediction mechanism for temporal sequences in V1.

Evaluating the neurophysiological evidence for predictive processing as a model of perception

“…For example, while both macaque V1 and V2 neurons respond to the illusory contour of a Kanizsa triangle, V2 neurons consistently respond earlier than those in V1 130 . Several fMRI studies have also reported that patterns of neural activity evoked in early visual cortex following the omission of an expected stimulus resemble the activity evoked by the veridical stimulus 49,51,132 and this result has been replicated in recordings from rodent V1 52 . Muckli et al 132 .…”

“…Interestingly, the activity was present up to 600 ms after stimulus offset and was temporarily erased by the presence of a visual mask, but reinstated on mask removal, demonstrating that the activity could not be explained through iconic memory mechanisms 133 . In a human fMRI study, Ekman et al 51 . repeatedly exposed participants to a dot stimulus rapidly moving across a screen.…”

“…Interestingly, the activity was present up to 600 ms after stimulus offset and was temporarily erased by the presence of a visual mask, but reinstated on mask removal, demonstrating that the activity could not be explained through iconic memory mechanisms. 133 In a human fMRI study, Ekman et al 51 repeatedly exposed participants to a dot stimulus rapidly moving across a screen. Following the exposure period, a flash of the dot stimulus at the starting position produced a time-compressed sequence of BOLD activity across the retinotopic locations of V1 corresponding to the previously observed stimulus trajectory.…”

“…67,79 Indeed, most of the studies that report PP-like dynamics in V1 use simple stimuli like gratings or flashes. 51,78,107,135,201 PP's stance on this problem is somewhat unclear. PP does not predict that the evidence for perceptual inference should be modular, siloed in an area of the brain specialized for the particular stimulus in question.…”

Mapping sequences can bias population receptive field estimates

“…Moreover, recent 57 studies indicate that pRF size and eccentricity vary in concert when the task requires to move the 58 focus of attention from fixation to the mapping stimulus, suggesting that processing resources are 59 Specifically, the position of a coherently moving stimulus can be anticipated based on its current 75 location and the direction of motion. The predictability of the stimulus location could induce an 76 anticipatory response in such locations (Ekman, Kok, & de Lange, 2017). Moreover, knowledge 77 of the upcoming stimulus location can provide spatial cues to direct attention to the relevant portion 78 of the screen affecting pRF estimates accordingly (Kastner, Pinsk, De Weerd, Desimone, & 79 Ungerleider, 1999).…”