Subscribe to RSS
DOI: 10.1055/s-0045-1807310
A genome-wide association study of brain function across multiple cognitive domains
Authors
Task-based fMRI is widely used to study the neurobiological basis of behavior, cognition, and emotion. Previous studies disagree on whether statistics derived from task-based fMRI are heritable – estimates range from approximately five percent to more than forty percent. Here we present the largest and most diverse genome-wide association study of task-based fMRI to date that uses a single, harmonized data analysis pipeline across all contributing sites. This abstract reports the SNP-based heritability results obtained from the current sample.
We invited researchers with access to relevant data to contribute through the ENIGMA consortium and public postings on social media. We chose three tasks that have been widely used for inclusion in the study. These are emotional faces, working memory, and reward tasks.
SNP-based analyses of seven datasets show moderate heritability across a wide range of brain regions for emotional faces and reward tasks. The amygdala is known to have a large effect size in the emotional faces task. However, we find greater heritability in cortical regions not commonly associated with the task. For reward, we found the maximum heritability in the striatum, which is consistent with brain maps found by imaging-only studies. We did not find significant heritability for working memory, likely due to a lack of statistical power. At time of writing, not all sites planned for inclusion in the meta-analysis have completed data analysis. We expect to increase statistical power by including these datasets.
Our results are consistent with previous findings of SNP-based heritability for the amygdala in the emotional faces task. This demonstrates the feasibility of genome-wide association studies for investigating individual differences in task-based fMRI. The results presented here will inform secondary analyses including genetic correlations and annotation. These may provide important insights into the relation of genes, molecules, cells, and circuits to psychological domains.
Publication History
Article published online:
30 April 2025
© 2025. Thieme. All rights reserved.
Georg Thieme Verlag KG
Oswald-Hesse-Straße 50, 70469 Stuttgart, Germany
