Surface area and cortical magnification of V1, V2, and V3 in a large sample of human observers

Abstract

How variable is the topological structure of early visual areas in human cortex? The Human Connectome Project 7T Retinotopy Dataset (Benson et al., 2018) can potentially answer this question as it contains the world’s largest collection of retinotopic measurements. We trained two anatomists to trace iso-eccentricity and iso-angle contours on the retinotopic maps of each subject. The contours followed the representations of 5 eccentricities (spanning 0.5–7°) and the horizontal and vertical meridians. These contours were used to calculate the mid-gray surface areas of V1, V2, and V3. Surface area estimates were extremely reliable across anatomists (r=0.95). Analyses were performed on 166 subjects (15 of 181 subjects with poor data quality were excluded). The median bilateral surface area of V1 (0–7° eccentricity) was 15.66 cm² with a standard deviation of 2.91 cm², and the maximum V1 size was 3.01-fold larger than the minimum. For V2, these values were 14.95, 2.56, and 3.16, respectively; for V3 they were 12.82, 2.27, and 2.93. The sizes of V1 and V2 were highly correlated across subjects (r=0.74) while V2 and V3 (r=0.57) and V1 and V3 (r=0.39) were less correlated. All surface areas (V1–V3) were only weakly correlated with total cortical surface area (r≈0.2–0.3). In V1, cortical magnification was, on average across subjects, well fit by m_LH=(16.9/(0.50+e))² and m_RH=(18.0/(0.63+e))² where e is eccentricity. Our measurements of V1, V2, and V3 sizes are broadly consistent with past reports. However, our estimates of cortical magnification near the fovea are slightly higher than found by Horton and Hoyt (1991) in lesion observations. The three-fold difference in minimum and maximum area of these regions mirrors the three-fold differences found in photoreceptor density (Curcio et al., 1987). These results provide a robust population benchmark against which new individuals and developmental and clinical populations can be compared.

Acknowledgement: R01 EY027401, R01 EY027964, R01 MH111417, NSF SMA1757390