Functional Connectome Correlates of Laterality Preferences: Insights into Hand, Foot, and Eye Dominance across the Lifespan

Abstract

Humans exhibit laterality preferences, with handedness being the most extensively studied. Accordingly, brain-handedness associations are well documented. However, laterality preferences extend beyond handedness to include other limbs, such as footedness and eyedness. Despite these distinctions, brain-footedness and brain-eyedness associations using resting-state functional connectomes remain largely unexplored. We utilize two large datasets, the Human Connectome Project-Development (HCP-D) and Human Connectome Project-Aging (HCP-A), to study the associations between sidedness (i.e., handedness, footedness, and eyedness) and whole-brain functional connectomes. While hand and foot preferences were correlated significantly, they explained <40% of the variance, suggesting some distinctions between measures. For both cohorts, significant associations between handedness connectivity were observed [p < 0.05, network-based statistics (NBS) corrected]. Notable patterns include increased connectivity for left-handedness in the posterior temporal areas and right-handedness in cerebellar regions. In contrast, significant associations between footedness and handedness connectivity were observed only in the HCP-A (p < 0.05, NBS corrected) and not the HCP-D. No significant associations between eyedness and connectivity were observed for either group. Finally, we compared the effect size between brain-handedness and brain-footedness associations. A greater difference was found in the HCP-D. The two cohorts primarily differed in edge distribution in the prefrontal lobe, temporal lobe, and cerebellum. Overall, in adults, brain-handedness and brain-footedness associations were similar. However, in children to adolescents, brain-handedness and brain-footedness associations diverge, suggesting a developmental shift. Characterizing sidedness associations with whole-brain connectomes may provide important insights into understanding the motor and visual systems, rehabilitation and occupational therapy, and benchmarking normative variations in the connectome.

Keywords: brainwide associations; eyedness; footedness; functional connectivity; handedness.

Figure 1.

Behavioral correlations between sidedness measures. While handedness and footedness were correlated significantly, they explained <40% of the variance of the other one, suggesting some distinctions between the measures. Correlation heatmaps are shown as variance explained (r²) for each pair of sidedness measures in both cohorts.

Figure 2.

Edges significantly differ between left- and right-sided individuals as calculated by network-based statistics. Circular bar plots showing the percent of significant edges belonging to each canonical network for each sidedness measure (rows) as separated by datasets (columns). Each circular bar plot depicts percentages for left- (cyan for HCP-D and salmon for HCP-A) and right-sided edges (blue for HCP-D and red for HCP-A). Handedness yielded a significance of p < 0.001 for both the HCP-A and the HCP-D, whereas footedness only yielded a significance of p < 0.001 for the HCP-A. The remainder of the results show no significant differences.

Figure 3.

Effect size quantifications and correlations for all edges in each sidedness measure. Top, Effect sizes (Cohen's d) for all 35,778 were calculated and plotted on a histogram for each sidedness measure and cohort separately. A wider histogram indicates that more edges show a larger effect size. Bottom, Pairs of sidedness measures from the top were further compared individually and accompanied with calculations of covariances for the HCP-D and the HCP-A, separately.

Figure 4.

Edges with significantly different effect sizes in handedness and footedness for each cohort separately. A, The number of edges with significantly different effect sizes for brain–handedness and brain–footedness associations, separated by cohort. As expected, the number of edges was significantly greater in the HCP-D than in the HCP-A (x² = 8.1, p < 0.001). B, These edges were located throughout the brain. Results for all visualizations were thresholded as follows to aid with interpretability (70 for HCP-D and 50 for HCP-A). C, Edges with significantly different effect sizes for brain–handedness and brain–footedness associations split by hemispheres and networks. * indicates networks where the HCP-D exhibited significantly more edges that differed in effect size between handedness and footedness.