Structural properties of V1 in achromats and controls.

(A) Mean V1 cortical thickness of each eccentricity based V1 segment in millimetres (mm). Cortical thickness was increased in achromats (n=31) compared to controls (n=100) only for foveal projection segments 0°-2° and not in more eccentric segments 2°-8°. This pattern was trending in V2 and significant in V3. Error bars are 95% confidence intervals (B) Visualisation of regions of interest in one control participant. Segmentation of V1 to 1° eccentricity bins, and the foveal projection zone (0°-2°) of V2 and V3 was based on the Benson retinotopy atlas (Benson and Winawer,2018). The half circle on the right shows, in visual space, the area represented by each segment.

pRF properties of V1 in achromats and controls.

(A) Foveal projection zone (0°-2°) coverage. Coverage is the fraction of vertices within the foveal confluence that contain visual information (i.e., pRF R2>0.1). Achromats (red markers) are not significantly different from controls (black markers) in the rods only condition (square markers), and do not show complete filling in even in the non-selective condition with higher contrast (triangle markers). Triangle markers: group mean for the non-selective condition, square markers: group mean for the rods only condition. Black markers: controls group mean, red markers: achromat group means. Black/red vertical lines around the mean: 95% confidence interval, grey shaded lines: 2.5%-97.5% percentile interval (only in controls). For achromats, participants’ dots are colour-coded to match across conditions and measures (B) Group eccentricity plots on a flat surface for each condition (in standard fsaverage space), presenting the group median value of each vertex. For visualisation purposes, vertices that were absent in more than 30% of the group were left blank in each condition. Black lines denote delineations of V1,2,3; white lines mark the foveal projection zone (0°-2°) in V1. As shown in (A) we see no evidence of ‘filling in’ of the rod scotoma in achromats. (C) pRF sizes along the atlas V1 segments. Achromats show a consistent increase in pRF size compared to controls. (D) Eccentricity of modelled pRFs along the atlas V1 segments. Achromats show an increase in eccentricity compared to controls, however that shift is not consistent across the two stimuli types. Vertical lines in C&D represent 95% confidence intervals.

Connective field modelling (V3) in achromats and controls – sampled eccentricities.

(A) Right: The fraction of all vertices within V3 that sample from V1, binned by sampled V1 atlas eccentricity. In achromats there is a shift towards greater sampling from the most foveal area that contains rods (1°-4°) of V1 relative to the normal-sighted rod and cone systems. Shaded areas: standard errors of the mean (s.e.m). All V1 eccentricities are derived from the Benson Atlas. Left: A schematic illustration of the sampled eccentricities in visual space. Dashed area represents the area where group differences in sampling were found, 1°-4°. (B) Visualisation of V3 sampled eccentricities of V1. From left to right: Half circle represents the visual field, where each colour represents 1 visual degree. V1 is colour-coded into segments based on the atlas-assigned eccentricities. Each vertex in V3 is then coloured based on the centre of Gaussian connective field within the V1 surface. Values in V3 are based on group medians. To further highlight the group differences in the amount of cortical surface devoted to the 1°-4° V1 segments we visualise V3 again on the right, colouring in all vertices with a CF centre at the 1°-4° segments.

Connective field modelling (V3) in achromats and controls – connective field size and relationships with pRF size.

(A) CF sizes in V3 in achromats and controls across the rod-selective and non-selective condition. Controls show an expected increase in CF size with sampled eccentricity while achromats show a flat pattern, with smaller CFs between 4°-6°. In both groups results are consistent across the two stimulation conditions. Ratio of pRF size between achromats and controls in V1 (circles) and V3 (triangles). A ratio of 1 indicates the mean pRF size of achromats is equivalent to that of controls. Across both non-selective (left) and rod-selective (right) stimulus conditions, the ratio of pRF sizes in V3 is closer to 1 than in V1, pointing towards a possible normalising effect of CF size on pRF size. (C) Illustrations depicting the differences in pRF and CF sizes across the groups. (D) A summary of the proposed model of how smaller CF sizes in V3 of achromats might operate as a compensatory mechanism for increased pRF size in V1, resulting in normalized pRF size at the level of V3.