Stability of specific personality network features corresponding to openness trait across different adult age periods: A machine learning analysis

Shengwen Zhi; Wentao Zhao; Ruiping Wang; Yue Li; Xiao Wang; Sha Liu; Jing Li; Yong Xu

doi:10.1016/j.bbrc.2023.06.012

Stability of specific personality network features corresponding to openness trait across different adult age periods: A machine learning analysis

Biochem Biophys Res Commun. 2023 Sep 10:672:137-144. doi: 10.1016/j.bbrc.2023.06.012. Epub 2023 Jun 14.

Authors

Shengwen Zhi¹, Wentao Zhao¹, Ruiping Wang², Yue Li¹, Xiao Wang¹, Sha Liu¹, Jing Li¹, Yong Xu³

Affiliations

¹ Department of Psychiatry, First Hospital/First Clinical Medical College of Shanxi Medical University, Taiyuan, Shanxi, China; Shanxi Key Laboratory of Artificial Intelligence Assisted Diagnosis and Treatment for Mental Disorder, First Hospital of Shanxi Medical University, Taiyuan, Shanxi, China.
² Science and Technology Information and Strategy Research Center of Shanxi, China.
³ Department of Psychiatry, First Hospital/First Clinical Medical College of Shanxi Medical University, Taiyuan, Shanxi, China; Shanxi Key Laboratory of Artificial Intelligence Assisted Diagnosis and Treatment for Mental Disorder, First Hospital of Shanxi Medical University, Taiyuan, Shanxi, China. Electronic address: xuyongsmu@vip163.com.

PMID: 37352602
DOI: 10.1016/j.bbrc.2023.06.012

Abstract

The functional connectivity patterns of the brain during resting state are closely related to an individual's cognition, emotion, behavior, and social interactions, making it an important research method to measure personality traits in an unbiased way, replacing traditional paper-and-pencil tests. However, due to the dynamic nature of the brain, whether the changes in functional connectivity caused by age can stably map onto personality traits has not been previously investigated. This study focuses on whether network features that are significantly related to personality traits can effectively distinguish subjects with different personality traits, and whether these network features vary across different periods of adulthood. The study included 343 healthy adult participants, divided into early adulthood and middle adulthood groups according to the age threshold of 35. Resting-state functional magnetic resonance imaging (fMRI) and the Big Five personality questionnaire were collected. we investigated the relationship between personality traits and intrinsic whole-brain functional connectome. We then used support vector machine (SVM) to evaluate the performance of personality network features in distinguishing subjects with high and low scores in the early-adulthood sample, and cross-validated in the mid-adulthood sample. Additionally, edge-based analysis (NBS) was used to explore the stability of personality networks across the two age samples. Our results show that the network features corresponding to openness personality trait are stable and can effectively differentiate subjects with different scores in both age samples. Furthermore, this study found that these network features vary to some extent across different periods of adulthood. These findings provide new evidence and insights into the application of resting-state functional connectivity patterns in measuring personality traits and help us better understand the dynamic characteristics of the human brain.

Keywords: Functional connectome; Network-based statistic; Personality neuroscience; Support vector machine.

Publication types

Research Support, Non-U.S. Gov't

MeSH terms

Adult
Brain*
Cognition
Connectome* / methods
Emotions
Humans
Magnetic Resonance Imaging / methods
Nerve Net
Personality