Fig. 1: Observer-independent border detection and profiling of PM areas.

A Region of interest of a cell-body stained section with a border (orange line) between areas 4a and 6d1 (top left). The black rectangles mark the positions of the zoom-ins below, showing the cytoarchitecture of 4a (bottom row, left) and 6d1 (bottom row, right), as well as the cell packing density over the cortical layers II–VI quantified by a grey level index (GLI) profile (black and green line). Roman numerals (I–VI) indicate cortical layers. Area 4a mainly differs from 6d1 by Betz cells in layer V, an unsharp layer VI-WM border and a broader layer VI, which is subdivided into two sublayers. For the observer-independent border detection, GLI profiles are extracted along transverses (1–162) along the cortical ribbon (top right). Averaged profiles of two adjacent blocks of profiles, including 12–24 profiles per block, are calculated to increase the signal-to-noise ratio. The Mahalanobis distances (MD) as a measurement for cytoarchitectonic differences is computed by calculating the distances between all pairs of neighbouring cortical blocks using a sliding window approach. These values are plotted as a function of the profile position. The upper graph shows this function using a block size of 23 profiles. A significant MD maximum (p < 0.001) is found at profile position 76. The graph below shows additional significant MD maxima at the profile position 76 using blocks of 16–24 profiles, indicating a border between neighbouring areas at this position. Additional borders between the PM areas and neighbouring areas are shown in Supplementary Fig. 1. B Cytoarchitecture of areas 6d2, 6d3, 6v1, 6v2, 6v3 and 6r1 and corresponding GLI profiles.