Dissertation, Otto-Friedrich-Universität Bamberg, 2019

Research on the biological basis of personality has a long tradition. Although the popular and widespread Five Factor Model (FFM) of personality is not biologically based, there is evidence that differences in brain structures are associated with individual differences in personality traits that are part of the FFM. To date, there is only sparse research on differences in task-related brain activity linked with differences in FFM factors such as conscientiousness. This is surprising because the study of differences in event-related potentials (ERPs) in the context of conscientiousness appears to be a promising approach. This is shown, for example, by evidence that conscientiousness is related to the error-related negativity (ERN).
According to cognitive theories the size of the ERN should be related to variations in behavior. However, recent research indicates that the ERN and behavioral measures are at least partly unrelated. It seems evident that individual difference measures, psychopathology, and motivational factors impact mechanisms generating the ERN and therefore moderate the size of the ERN. The CRN is the counterpart of the ERN occurring after correct responses. Based on the assumption that the CRN is at least partly reflecting an equal cognitive process as the ERN, the size of the CRN should be moderated by individual difference measures as well.
I conducted two experiments addressing the impact of conscientiousness on the task-related ERP after both errors and correct responses. In both experiments, I used simple-choice reaction tasks in different variations to investigate ERN and CRN amplitudes in relation to conscientiousness and deliberate rule violations. I observed variations in the ΔERN amplitude (which is the difference amplitude of ERN and CRN) related to conscientiousness that might indicate a greater focus on the task and a motivational salience of correct responses. I also observed differences in the CRN amplitude that are due to response conflict moderated by conscientiousness. The results showed an apparent association of conscientiousness and the task-related response-locked ERP.
Finally, I examined whether differences in the ERP can be used as biomarkers for the classification of conscientious individuals through the application of linear discriminant analysis (LDA) and linear regression (LR), machine learning methods popular for classification of single-trial and aggregated ERP data. I propose a classification procedure that can also be used for classification of other features related to ERP data. Both LDA and LR embedded in the procedure demonstrate their suitability and usefulness by reanalyzing the ERP datasets of both experiments. The classification performance was evaluated using cross validation and showed that both methods successfully discriminated above chance between two groups of individuals scoring low and high on a conscientiousness scale.
Future research should concentrate on moderating effects on the association of conscientiousness and CRN and ΔERN.