Purchase this article with an account.
Michael Pitts, Jan Nerger, Clea Stalmaster; The role of spatial and selective attention in the perception of bistable images. Journal of Vision 2007;7(9):186. doi: https://doi.org/10.1167/7.9.186.
Download citation file:
© ARVO (1962-2015); The Authors (2016-present)
Recent studies (Pitts, Nerger, & Davis, 2007; Kornmeier & Bach, 2004; 2005) report early event-related potentials (ERP) associated with perceptual reversals of bistable images — enhanced P1 and N1 amplitudes and a broad reversal-related negativity (180–400msec post-stimulus). Pitts et al argue these results support a model of bistable perception (see Slotnick & Yantis, 2005) in which changes in early spatial and selective attention, indicated by P1 & N1 enhancements and the selection negativity (see Anllo-Vento, Luck, & Hillyard, 1998), modulate perceptual reversals.
In the present study, a Necker-type stimulus was presented for 800msec with a 400msec ISI to allow time-locking of ERPs to stimulus onset. Observers pressed one of two response buttons after each trial to indicate left or right perceptual orientation of the stimulus. Unlike previous studies, this method permits us to correlate ERPs to what the observer perceived on each trial rather than simply whether a reversal had occurred or not. Recordings were partitioned into four categories based on reported percepts in each trial and the preceding trial: Left stable, right stable, left-to-right reversal, right-to-left reversal. By making comparisons among these four waveforms we can directly assess multiple competing models of bistable perception.
Preliminary analyses reveal a frontal positivity (≈140–380msec) and occipital/temporal/parietal negativities (≈170–400msec) at sites contralateral to the perceived foreground of the stimulus for reversal vs. stability trials, e.g. by comparing left-to-right reversal trials with right stability trials and vise versa. A comparison of the two different reversal waveforms (L to R vs. R to L) revealed a contralateral negativity (≈220–400msec) at parietal and temporal sites, whereas a comparison between the two types of stability waveforms (L vs. R) revealed no differences at the equivalent sites. Taken together, these results support a model of bistable perception whereby reversals are initiated by shifts in spatial and/or selective attention.
This PDF is available to Subscribers Only