An increasing number of studies is currently focusing on ‘personality neuroscience’, a term labeling the research aimed at neuroimaging correlates of inter-individual temperament and character variability. Among other methods, a graph theoretical analysis of the functional connectivity in resting state functional magnetic resonance imaging data was applied in a recent study by Gao et al. (2013, Frontiers in Human Neuroscience). This paper aims to replicate this study and extends the original statistical methods in order to demonstrate the effect of multiple comparison problem. In contrast to the original study, five personality dimensions were obtained in the revised ‘Big Five’ Personality Inventory. Using a larger sample (84 subjects) and an equivalent data analysis procedure, we obtained widely disagreeing results compared to the original study. While the Gao et al. reported a range of significant correlations between personality dimensions and some of the network metrics, we failed to replicate any significant correlations when FDR testing was applied. These results demonstrate that as with other neuroimaging studies, appropriate control of multiple comparison problem should be meticulously applied in order to prevent such false alarms in research into neurological substrates of personality differences. Of course, we do not attempt to disprove the existence of some link between personality and brain’s intrinsic functional architecture. Nevertheless, this link is very likely much more subtle and elusive than was suggested in previous studies.
On seeking souls in brain graphs: fancy neuroimaging analysis gone wrong?
