Brain differences and their correlates

A key focus of our lab is understanding individual differences in the human brain, including:

• how brains differ from each other;

• how these differences develop across the lifespan; and

• how they relate to variability in behavior, cognition, and clinical outcomes.

Fine-grained brain functional organization¶

Fig 1. Coarse- vs. fine-grained information. Coarse-grained information is derived by spatially smoothing or averaging fMRI data within regions, reflecting their overall activity or connectivity (centimeter scale). Fine-grained information, in contrast, captures detailed vertex-by-vertex or voxel-by-voxel patterns within regions (millimeter scale).

We are particularly interested in individual differences in fine-grained brain functional organization --- the detailed spatial patterns of brain activity and connectivity at the voxel or vertex level. Compared to coarse-grained differences (e.g., region-level averages), these fine-grained differences are

1. more reliable across independent data and more congruent across functional indices Feilong et al., 2018;

2. more predictive of intelligence and other cognitive abilities Feilong et al., 2021;

3. more experience-driven, potentially reflecting learning and training effects Busch et al., 2024.

Fig 2. Predicting general intelligence (g). Fine-grained functional connectivity profiles (y-axis, blue histogram) are markedly better predictors of general intelligence than coarse-grained connectivity profiles (x-axis, green histogram) across different brain systems, accounting for twice as much variance in g. Adapted from Figure 3 in Feilong et al. (2021). Similar effects are observed for both task- and resting-state connectivity, and across different cognitive abilities Feilong et al., 2021.

These fine-grained differences are often obscured by individual variability in anatomical--functional correspondence and cannot be properly studied using traditional alignment methods. We use hyperalignment Haxby et al., 2020 and the Individualized Neural Tuning (INT) model Feilong et al., 2023 to resolve these idiosyncrasies and reveal the individual differences in fine-grained functional organization.

References¶