Homotopic Functional Affinity Unraveling Regional Specialization and Cross-Cultural Diversity
Poster No:
1763
Submission Type:
Abstract Submission
Authors:
Li-Zhen Chen1, Xi-Nian Zuo1
Institutions:
1Beijing Normal University, Beijing, CN
First Author:
Co-Author:
Introduction:
The integration/specialization of the two brain hemispheres can be characterized by the functional similarity of homotopic areas, providing important perspectives for the neural correlates of cognition and behavior. Despite the decisive role of spatial geometric constraints and homophilic attachment on the human connectome (Betzel et al., 2016), traditional practices in mapping functional homotopy only considered the correlations of functional timeseries between homotopic areas, disrespecting homophily factors and interactions between non-homotopic brain areas. Here, we proposed a new method on functional homotopy analysis, namely Homotopic Functional Affinity (HFA), depicted a new prospect for studies investigating hemispheric integration/specialization through the lens of functional homotopy.
Methods:
HFA evaluates the affinity between the full-brain Functional Connectivity Profiles (fbFCP) of homotopic areas, simultaneously captures geometric constraints (homotopic location) and homophily (fbFCP affinity) (Fig.1). Using high-precision resting state fMRI data from the Human Connectome Project (HCP, Van Essen et al., 2013) and the Chinese Human Connectome Project (CHCP, Ge et al., 2023), we constructed the human HFA map and evaluated its test-retest reliability with Intraclass Correlation Coefficient (ICC). Next, we validated HFA through ROI-based analysis, including ROIs' fbFCP comparison, meta-analysis of cognitions involved in ROIs, and the correlation analysis between activations of ROIs under different tasks and their HFA. Finally, we validated HFA at global level by calculating the correlation between the HFA map and multimodal brain maps of evolution, gene expression, myelination, functional gradient and cognitive association.
Results:
The HFA maps were highly similar to the principal functional gradient map (Margulies et al., 2016), with an ICC of the global mean greater than 0.5, and vertex-level ICCs generally greater than 0.4, while some vertices with almost perfect reliability (Fig.1).
Low HFA of the Inferior Parietal Lobule (IPL) and their clear differentiable subregions were observed. Based on the HFA map of HCP, three adjacent IPL subregions were defined as ROIs, namely aIPL, mIPL, and pIPL. The fbFCP of aIPL and pIPL showed cross-dataset consistency. The left aIPL strongly connected with the default mode network (DMN), while the right aIPL tended to connect with attention networks. Importantly, the right aIPL negatively connected with the DMN. The left pIPL also had the strongest connectivity with DMN, but many negative connectivity appeared in the attention networks, while the right pIPL had fewer negative connectivity globally. The fbFCP of mIPL displayed transition between aIPL and pIPL. The fbFCP of left mIPL was similar to that of the left aIPL, and the fbFCP of right mIPL was similar to that of the right pIPL (Fig.2). Compared with the connectivity in HCP, the left mIPL in the CHCP had more negative connectivity with the ventral attention network.
The results of meta-analysis showed a gradual functional shift from attention to social cognition, and then to language between aIPl, mIPL, and pIPL. Functional specialization of ROIs in left hemisphere was stronger, and their right counterparts engaged in more extensive cognitions. The results of task activation correlation were consistent with those of the meta-analysis. Notably, there were significant differences across datasets in social cognition tasks. Furthermore, multimodal brain maps' correlation analysis illustrated a close relationship between HFA map and multimodal brain maps (Fig.2).
Low HFA of the Inferior Parietal Lobule (IPL) and their clear differentiable subregions were observed. Based on the HFA map of HCP, three adjacent IPL subregions were defined as ROIs, namely aIPL, mIPL, and pIPL. The fbFCP of aIPL and pIPL showed cross-dataset consistency. The left aIPL strongly connected with the default mode network (DMN), while the right aIPL tended to connect with attention networks. Importantly, the right aIPL negatively connected with the DMN. The left pIPL also had the strongest connectivity with DMN, but many negative connectivity appeared in the attention networks, while the right pIPL had fewer negative connectivity globally. The fbFCP of mIPL displayed transition between aIPL and pIPL. The fbFCP of left mIPL was similar to that of the left aIPL, and the fbFCP of right mIPL was similar to that of the right pIPL (Fig.2). Compared with the connectivity in HCP, the left mIPL in the CHCP had more negative connectivity with the ventral attention network.
The results of meta-analysis showed a gradual functional shift from attention to social cognition, and then to language between aIPl, mIPL, and pIPL. Functional specialization of ROIs in left hemisphere was stronger, and their right counterparts engaged in more extensive cognitions. The results of task activation correlation were consistent with those of the meta-analysis. Notably, there were significant differences across datasets in social cognition tasks. Furthermore, multimodal brain maps' correlation analysis illustrated a close relationship between HFA map and multimodal brain maps (Fig.2).
Conclusions:
The consistency of results in different analyzes demonstrated the feasibility of using HFA to explore the rs-fMRI and t-fMRI associations and the sensitivity of HFA in detecting brain functional specialization mechanisms underlying cultural-related differences. This new method will provide a reliable and effective tool for future population neuroscience research.
Modeling and Analysis Methods:
fMRI Connectivity and Network Modeling 1
Methods Development 2
Keywords:
Other - Hemispheric Functional Integration and Specialization; Connectome; Affinity; Inferior Parietal Lobule; Cross-Cultural Difference
1|2Indicates the priority used for review
·HFA diagram
·Validation of HFA from HCP dataset at ROl level and global level
Provide references using author date format
Ge, J. (2023), 'Increasing diversity in connectomics with the Chinese Human Connectome Project', Nature Neuroscience, vol. 26, no. 1, pp. 163-172.
Margulies, D.S. (2016), 'Situating the default-mode network along a principal gradient of macroscale cortical organization', Proceedings of the National Academy of Sciences of the United States of America, vol. 113, no. 44, pp. 12574-12579.
Van Essen, D.C. (2013), 'The WU-Minn Human Connectome Project: an overview', Neuroimage, vol. 80, pp. 62-79.