Position coding mechansims for binocular disparity revealed by fMRI
Date Issued
2015
Date
2015
Author(s)
Yu, Wei-Hsiang
Abstract
Electrophysiological evidence suggests that the visual cortical neurons encode disparity through either position or phase difference between left and right receptive fields. Since Anzai, Ohzawa, & Freeman (1999a) proposed that the visual cortical neurons encode binocular disparity mainly rely on the receptive fields phase difference but less position difference, how group of neurons encode binocular disparity remains unclear. Here, we investigated the position disparity model at the population level with fMRI. Retinotopic BOLD activations were measured under binocular and two monocular (the left and right eye) viewing conditions with low contrast checkerboard patterns and fitted to a population receptive field (pRF) model (Dumoulin & Wandell, 2008). The position disparity was measured by the distance between centers of left and right pRFs. Also, a bootstrap method was used to generate simulated data to estimate variations of pRFs of each viewing condition of each voxel. This allows us to test binocular disparity of each neural population statistically. Our results showed that distributions of horizontal and vertical disparities of visual cortical neurons do not significant deviate from zero, suggesting that visual system does not prefer to any disparity. Moreover, simulations showed that a bunch of binocular voxels (30% on average) have a significant pRFs position shift (at α =.01 level), suggesting a position coding of disparity at the population level. Also, both horizontal and vertical disparities increase with the eccentricity in all visual cortices, which consistent with psychophysics’ results that the fusion limit is positive correlate to the eccentricity. Together, our results revealed (A) a position coding of binocular disparity and (B) neural evidence for increasing fusion limit with the eccentricity at the population level.
Subjects
Binocular disparity
Type
thesis
File(s)![Thumbnail Image]()
Loading...
Name
ntu-104-R01227113-1.pdf
Size
23.54 KB
Format
Adobe PDF
Checksum
(MD5):8cfd15f6044a0aef44c7eee4e4b2ca41