A psychophysically-tuned computational model of human primary visual cortex produces geometric optical illusions

Abstract

<jats:title>Abstract</jats:title><jats:p>Geometrical optical illusion (GOIs) are mismatches between physical stimuli and perception. GOIs provide an access point to study the interplay between sensation and perception, yet there is scant quantitative investigation of the extent to which different GOIs rely on similar or distinct brain mechanisms. We addressed this knowledge gap. First, 30 healthy adults reported quantitatively their perceptual biases with three GOIs, whose physical properties parametrically varied on a trial-by-trial basis. Biases observed with one GOI were unrelated to those observed with another GOI, suggestive of (partially) distinct underlying mechanisms. Next, we used these psychophysical results to tune a computational model of primary visual cortex that combines parameters of orientation, selectivity, intra-cortical connectivity, and long-range interactions. We showed that similar biases could be generated <jats:italic>in-silico</jats:italic>, mirroring those observed in humans. Such results provide a roadmap whereby computational modelling, informed by human psychophysics, can reveal likely mechanistic underpinnings of perception.</jats:p>

A Cortical-Inspired Sub-Riemannian Model for Poggendorff-Type Visual Illusions

“…This phenomenon, as well as a more detailed study on how the choice of the parameters used to generate Figure 3 a (such as the incidence angle, the width of the central gray bar, the distance between lines) in a similar spirit to [ 69 ] where psysho-physicis experiments were performed on analogous images, is an interesting topic for future research.…”

A cortical-inspired sub-Riemannian model for Poggendorff-type visual illusions

“…As shown by these preliminary examples, the prosecutions computed by both (LHE) models agree with our perception as the reconstructed connection in the target region links the two misaligned segments, while somehow 'stopping' the connection of the collinear one. This phenomenon as well as a more detailed study on how the choice of the parameters used to generate Figure 2a (such as the incidence angle, the width of the central gray bar, the distance between lines), in a similar spirit to [66] where psyshophysicis experiments were performed on analogous images, is an interesting topic for future research.…”