Personality and local brain structure: Their shared genetic basis and reproducibility

Abstract

Local cortical architecture is highly heritable and distinct genes are associated with specific cortical regions. Total surface area has been shown to be genetically correlated with complex cognitive capacities, suggesting cortical brain structure is a viable endophenotype linking genes to behavior. However, to what extend local brain structure has a genetic association with cognitive and emotional functioning is incompletely understood. Here, we study the genetic correlation between personality traits and local cortical structure in a large-scale twin sample (Human Connectome Project, n ​= ​1102, 22-37y) and we evaluated whether observed associations reflect generalizable relationships between personality and local brain structure two independent age-matched samples (Brain Genomics Superstructure Project: n ​= ​925, age ​= ​19-35y, enhanced Nathan Kline Institute dataset: n ​= ​209, age: 19-39y). We found a genetic overlap between personality traits and local cortical structure in 10 of 18 observed phenotypic associations in predominantly frontal cortices. However, we only observed evidence in favor of replication for the negative association between surface area in medial prefrontal cortex and Neuroticism in both replication samples. Quantitative functional decoding indicated this region is implicated in emotional and socio-cognitive functional processes. In sum, our observations suggest that associations between local brain structure and personality are, in part, under genetic control. However, associations are weak and only the relation between frontal surface area and Neuroticism was consistently observed across three independent samples of young adults.

Fig. 1.. Distribution of personality traits in the full HCP sample.

Distribution of NEO-FFI personality traits in the HCP dataset, score on x-axis, number of occurrences on the y-axis, as well as the correlation between NEO-FFI traits in the HCP sample (A = Agreeableness, C = Conscientiousness, E = Extraversion, N = Neuroticism, O = Openness).

Fig. 2.. Relation between personality traits and local brain structure in the full HCP sample.

A) Mean cortical thickness of each parcel and the distribution of average cortical thickness across participants; B) Regional associations between personality traits and cortical thickness; C) Average surface area sum per parcel and the distribution of total surface area across participants; D) Regional associations between surface area and personality traits. Positive associations between local brain structure and each personality trait are displayed in red and negative associations displayed in blue. Multiple comparisons were accounted for by using FDR corrections at q < 0.05 correcting for the number of parcels (200) and only significant associations are displayed.

Fig. 3.. Heritability of local cortical structure and personality traits.

A) Heritability of local cortical thickness; B) Heritability of surface area; C) Heritability of NEO-FFI: A = Agreeableness, C=Conscientiousness, E = Extraversion, N=Neuroticism, O=Openness.

Fig. 4.. Quantitative functional decoding of consistent associations between personality and local brain structure.

Both forward inference and reverse inference of activation-domain and paradigm-domain contrasts are reported for the right medial frontal cortex which showed evidence of successful replication in two samples.