Purchase this article with an account.
Yih-Shiuan Lin, Chien-Chung Chen, Mark W. Greenlee; Neural Correlates of Perceptual Filling-In as Measured by Functional Magnetic Resonance Imaging. Journal of Vision 2020;20(11):279. doi: https://doi.org/10.1167/jov.20.11.279.
Download citation file:
© ARVO (1962-2015); The Authors (2016-present)
Perceptual filling-in describes the phenomenon when our visual system replaces the central missing information with the surround features. We investigated such center-surround modulation by using functional magnetic resonance imaging (fMRI). Three radial sinusoidal grating stimuli were used in an event-related fMRI experiment: a full-field grating (center-surround), a filling-in grating with a crescent artificial scotoma in the periphery on each of the visual field (surround-only), and a pedestal grating containing radial pattern only in the scotoma locations (center-only). We doubled the trial number of the filling-in grating and adjusted the pattern luminance contrast for each observer until they perceived filling-in in about 50% of the trials, making it possible to compare the BOLD signal with and without reported filling-in. On each trial, one of the three gratings flickered in counterphase for 10 seconds, followed by a 14 seconds blank period. Observers were to press a button whenever the fixation point changed color, and to report whether filling-in was experienced during the presentation of the filling-in gratings. The functional regions of interest (ROIs) of the scotoma in the visual cortex (V1-3) were determined in an independent localizer session in which an on-and-off flickering checkerboard were presented in the “scotoma” location. The univariate analysis results showed no BOLD activation difference in the ROIs between the filling-in and non-filling-in trials, suggesting that the overall activation level did not correspond to the subjective percept. Interestingly, the BOLD activation was stronger in the pedestal condition compared to the full-field condition, indicating that adding the surround reduced the center activation. This finding supports that lateral inhibition was introduced from the surround. The multivariate pattern analysis (MVPA) results showed that response patterns in the visual cortex can significantly differentiate between filling-in and non-filling-in trials, suggesting that the neural activation pattern depends on the subjective percept of the observers.
This PDF is available to Subscribers Only