Skip to main page content
Access keys NCBI Homepage MyNCBI Homepage Main Content Main Navigation
, 12, 109

Focusing on the Differences of Resting-State Brain Networks, Using a Data-Driven Approach to Explore the Functional Neuroimaging Characteristics of Extraversion Trait


Focusing on the Differences of Resting-State Brain Networks, Using a Data-Driven Approach to Explore the Functional Neuroimaging Characteristics of Extraversion Trait

Feng Tian et al. Front Neurosci.

Erratum in


In recent years, functional magnetic resonance imaging (fMRI) has been widely used in studies that explored the personality-brain association. Researches on personality neuroscience have the potential to provide personality psychology with explanatory models-that is, why people differ from each other rather than how they differ from each other (DeYoung and Gray, 2009). As one of the most important dimensions of personality traits, extraversion is the most stable core and a universal component in personality theory. The aim of the present study was to employ a fully data-driven approach to study the brain mechanism of extraversion in a sample of 111 healthy adults. The Eysenck Personality Questionnaire (EPQ) was used to measure the personality characteristics of all the subjects. We investigated whether the subjects can be grouped into highly homogeneous communities according to the characteristics of their intrinsic connectivity networks (ICNs). The resultant subjects communities and the representative characteristics of ICNs were then associated to personality concepts. Finally, we found one ICN (salience network) whose subject community profiles exhibited significant associations with Extraversion trait.

Keywords: data-driven; extraversion trait; personality traits; resting-state fMRI; salience network.


Figure 1
Figure 1
Demonstration of data analysis flow. For simplicity, we assume that there are only three subjects (denoted as S1, S2, and S3). First, the fMRI data of subjects are decomposed individually by using spatial independent component analysis (ICA) into spatial components (ICs). Assume that for each subject we can get three ICs that are color coded to indicate which subject they are from. The resultant ICs maps are presented in the green layer. Second, all of the ICs from individual subjects were pooled in gRAICAR (as presented in the purple layer). We present a distance space depicting the similarity between all ICs in the yellow layer. The intention of gRAICAR in this part is to identify ICs that are from different individuals but are close to each other (as marked with green dashed circles). The group-level aligned components (ACs) were formed by these clustered ICs sequentially, and a community detection algorithm can be applied to each AC to identify homogeneous subject communities among all subjects. Third, we try to seek a kind of personality trait, according to which the subjects could be grouped into communities that maximally agreed with the brain neural activity derived communities.
Figure 2
Figure 2
gRAICAR reveal the salience network is associated with Extraversion classification. (A) The salience network dominated by the anterior cingulate and bilateral anterior insula. (B) Combined with the Extraversion scores, the similarity matrix change into a regular distribution. Compared to the LES subjects, the HES subjects have a higher inter-subjects average similarity. For visualization purpose, the subjects are grouped into LES and HES groups, and the blue solid lines mark the boundary between the two groups. (C) Map of the salience network showing inter-regional connections exhibiting significant (green line) differences in connectivity strength (Fisher's Z) between LES and HES groups. (D) Bar graphs showing separately statistical details comparing functional connectivity strength between LES and HES groups. Labels along the horizontal axis correspond to the connections marked on (C). All three inter-ROI connections show significant difference between LES and HES groups. The surface maps are rendered in BrainNet Viewer (Xia et al., 2013).
Figure 3
Figure 3
Correlations between extraversion scores and connectivity strength in salience network. The numbers on the y, x coordinate indicate extraversion scores and connectivity strength (Fisher's Z) between each ROI of salience network. (A) Connection 2 (between ACC and right insula), r = 0.304, p < 0.01; (B) connection 1 (between ACC and left insula), r = 0.302, p < 0.01; (C) connection 3 (between bilateral insula), r = 0.282, p < 0.01.

Similar articles

See all similar articles

Cited by 1 article


    1. Adelstein J. S., Shehzad Z., Mennes M., Deyoung C. G., Zuo X. N., Kelly C., et al. . (2011). Personality is reflected in the brain's intrinsic functional architecture. PLoS ONE 6:e27633. 10.1371/journal.pone.0027633 - DOI - PMC - PubMed
    1. Avants B., Gee J. C. (2004). Geodesic estimation for large deformation anatomical shape averaging and interpolation. Neuroimage 23(Suppl. 1), S139–S150. 10.1016/j.neuroimage.2004.07.010 - DOI - PubMed
    1. Barrett L. F. (2005). Feeling is perceiving: core affect and conceptualization in the experience of emotion, in Emotions:Conscious and Unconscious, eds Barrett L. F., Niedenthal P. M., Winkielman P., editors. (New York, NY: Guilford; ), 255–284.
    1. Barrett L. F. (2006). Solving the emotion paradox: categorization and the experience of emotion. Pers. Soc. Psychol. Rev. 10, 20–46. 10.1207/s15327957pspr1001_2 - DOI - PubMed
    1. Barrett L. F. (2009). The future of psychology: connecting mind to brain. Perspect. Psychol. Sci. 4, 326–339. 10.1111/j.1745-6924.2009.01134.x - DOI - PMC - PubMed

LinkOut - more resources