EEG rhythms in long-lasting spatio-temporal feature integration

Pre-stimulus brain activity is thought to modulate visual perception. However, the nature of this modulation remains debated. In particular, the use of paradigms with very brief and weak stimuli complicates a univocal understanding. Here we focus on phenomena of long-lasting feature integration. We analyzed electroencephalography recordings (EEG) from a sequential metacontrast paradigm (SQM). In the SQM, participants discriminate the offset of a central line followed by flanking lines. Because of metacontrast masking, the central line itself is invisible but its offset can be perceived in the motion stream created by the lines. When a second line in the stream has an offset in opposite direction, the offsets integrate. We investigated whether pre- and post-stimulus brain activity can predict the perceived offset. When comparing trials in which the first or the second offset was reported, we observed no influence of pre-stimulus activity. However, by applying a linear classifier to post-stimulus activity, we were able to decode the relative dominance between the two offsets. The decoded signal was linearly related to pre-stimulus power increases in the alpha-theta band (7-12 Hz). This initial result suggests that low-frequency pre-stimulus power may modulate the relative weighting of visual stimuli embedded in a continuous stream.