Structural and functional connectivity of the inferior temporal numeral area

Abstract

A growing body of evidence suggests that in adults, there is a spatially consistent "inferior temporal numeral area" (ITNA) in the occipitotemporal cortex that appears to preferentially process Arabic digits relative to non-numerical symbols and objects. However, very little is known about why the ITNA is spatially segregated from regions that process other orthographic stimuli such as letters, and why it is spatially consistent across individuals. In the present study, we used diffusion-weighted imaging and functional magnetic resonance imaging to contrast structural and functional connectivity between left and right hemisphere ITNAs and a left hemisphere letter-preferring region. We found that the left ITNA had stronger structural and functional connectivity than the letter region to inferior parietal regions involved in numerical magnitude representation and arithmetic. Between hemispheres, the left ITNA showed stronger structural connectivity with the left inferior frontal gyrus (Broca's area), while the right ITNA showed stronger structural connectivity to the ipsilateral inferior parietal cortex and stronger functional coupling with the bilateral IPS. Based on their relative connectivity, our results suggest that the left ITNA may be more readily involved in mapping digits to verbal number representations, while the right ITNA may support the mapping of digits to quantity representations.

Keywords: brain connectivity; numerical cognition; object processing; occipitotemporal cortex; ventral stream.

Fig. 1

fMRI task paradigm. Figure adapted from Pollack and Price (2019). Dotted lines under the target stimuli are included here for presentation purposes only. Detection tasks were performed separately and counterbalanced across participants.

Fig. 2

ITNA and Letter seed areas in relation to a non-exhaustive sample of reported literature coordinates.

Fig. 3

Schematic of the homotopic mapping procedure. This process was used to estimate an anatomically informed transformation of data between surfaces of each hemisphere. This allowed for direct, whole-surface comparisons of the ipsilateral and contralateral hemisphere connectivity profiles of the left and right ITNAs. Images in the final panel demonstrate the mapping result on a single participant’s data.

Fig. 4

Structural Connectivity of ITNAs and Letter area. A–C) Group-average connectivity to ipsilateral and contralateral cortical surface of each seed ROI, measured in units of fiber cross-section area (mm²), and masked according to the consistency-based thresholding procedure (Supplementary Fig. 1). Connectivity to lateral wall of the contralateral hemisphere was relatively weak, so only the medial wall of the contralateral hemisphere is depicted here. See Supplementary Fig. 1 for views of the contralateral hemisphere and these results at more liberal thresholds.

Fig. 5

Comparison of Structural Connectivity of ITNAs and Letter area. A, B) Contrast and conjunction statistical maps involving the ITNA L and Letter area connectivity profiles. Contrast Z-scores derived from the TFCE-correction procedure. Z-scores ranged from −4.26 to 4.26, corresponding to the depicted P-values of 10e⁻⁶. C, D) Contrast and conjunction maps involving the left and right ITNA connectivity profiles. These comparisons involve ITNA R connectivity mapped to the contralateral hemisphere, such that data on the left and right hemisphere represent connectivity to the ipsilateral and contralateral hemisphere, respectively (see Methods for our homotopic mapping procedure).

Fig. 6

Structural profile of ITNA and Letter area projections. A) Fiber density, i.e. estimated total (weighted) cross-sectional area of white matter fibers innervating each region, is plotted. Subject-specific measurements are represented by black dots. B) Average streamline length, incorporating streamline weights for improved biological validity, is plotted for each region. Streamlines were those that had at least one termination point within the region.

Fig. 7

White Matter Bundle Overlap. Mean volumetric overlap (Dice coefficient) of bundle and track-density images (TDI), across participants. MNI-space bundles taken from the XTRACT subset of the Pandora 4D population atlas (Warrington et al. 2020; Hansen et al. 2021). We focus only on connections to the ipsilateral cortical surface, due to the relatively sparse degree of axonal projections to the contralateral hemisphere. Though we report our primary results on both hemispheres, restriction of structural connectivity results to the ipsilateral hemisphere is common in the literature (Bouhali et al. 2014; Grotheer et al. 2019). The color scale was fit to the range of observed values, with white to red representing the lowest to highest degree of overlap, respectively.

Fig. 8

Functional Connectivity of ITNAs and Letter area. A–C) Group-average task-general functional connectivity to ipsilateral and contralateral cortical surface from each seed ROI. Task-general connectivity referred to the correlation in the beta-series across all task events, regardless of condition. Pearson r values were then converted to Fisher Z.

Fig. 9

Comparison of Functional Connectivity of ITNAs and Letter area. A, B) Contrast and conjunction statistical maps involving the left ITNA and Letter area connectivity profiles (task-general connectivity). Contrast Z-scores derived from the TFCE-correction procedure. Z-scores range from −4.26 to 4.26, corresponding to the depicted P-values of 10e⁻⁶. For conjunction analyses, only positive connections were assessed. C, D) Contrast and conjunction maps involving the left and right ITNA. All comparisons here involve ITNA R connectivity mapped to the contralateral hemisphere, such that data on the left and right hemisphere represent connectivity to the ipsilateral and contralateral hemisphere, respectively (see Methods for our homotopic mapping procedure). E) Condition-specific connectivity analyses, where the beta-series correlation across trials of one condition was contrasted with another. Only one cluster survived TFCE-correction, for the Lp-La contrast of Letter area connectivity. F) Contrasts in condition-specific connectivity (difference in preferential connectivity) between regions. No clusters survived TFCE-correction for either contrast.