Personality traits are the core building blocks for understanding individual differences in daily functioning. While neurobiological theories of personality have a long history, their empirical validation remains challenging. A key challenge is the mapping of broad traits to much more fine-grained neuro-behavioral measures (or vice versa), neglecting their hierarchical organization. This symposium demonstrates how matching the specificity of traits, behaviors, temporal dynamics, and neural spatial resolution refines understanding of personality across the domains of emotion and cognition (N=150-450).

The first talk (emotional responses/trait resolution) presents a comprehensive machine learning multiverse analysis on a large fMRI dataset, demonstrating that neural responses to emotional stimuli robustly predict specific trait facets (e.g., vulnerability to stress) rather than broad domains (e.g., neuroticism).

The second talk (emotion regulation/behavioral resolution) distinguishes between the selection and implementation of emotion regulation strategies during EEG, showing differential associations to mental health-related personality traits and executive function.

The third talk (executive functions/temporal resolution) demonstrates a data-driven time-frequency decomposition of EEG responses to dissect cognitive flexibility across three tasks, refining the neural characterization of task switching.

The fourth talk (intelligence/spatial resolution) uses fine-grained white matter microstructure analyses to investigate fluid intelligence, integrating structural equation modeling with measures of structural integrity, axon density, and myelin content.

Lastly, a moderated discussion will integrate findings and future directions guided by models of personality neuroscience (e.g., Brunswick asymmetry, causal theory).

Together, this symposium highlights how methodological and statistical advances enhance the neurobiological mapping of personality, increasing precision across traits, processes, and neuroimaging techniques.

The neural basis of personality is often inferred from activity in single brain regions or networks, but these approaches frequently lack validity for broad trait constructs. This talk presents theory-guided models and a comprehensive machine learning multiverse analysis on a large fMRI dataset (N=458), evaluating whether trait negative affectivity can be decoded from neural responses to emotional stimuli. Contrary to longstanding theories, we found Bayesian evidence against associations between neuroticism and traditional region- or network-based measures (e.g., amygdala, salience network) as well as complex neuro-affective signatures. However, whole-brain machine learning models successfully predicted the neuroticism facet vulnerability to stress (r=.21, replication r=.19), highlighting the importance of specificity when linking personality to neural data. Exploratory analyses across 14 affective constructs and 1176 model specifications further illustrate the task-dependence of neural trait prediction. Findings challenge common neurobiological interpretations of broad traits while demonstrating the utility of fine-grained, data-driven approaches for personality neuroscience.

Previous studies suggest that different emotion regulation (ER) strategies vary in the effectiveness with which they reduce specific components of an emotional response. In this study, we investigate how individual differences in the effectiveness of distraction and reappraisal with regard to subjective experience as well as physiological responding are related to differences in (a) the selection of these strategies in adaptation to context factors like stimulus intensity, (b) executive functions (EF), and (c) mental health-related personality traits. In a non-clinical sample of N = 279 participants, we studied ER with an ER implementation and an ER selection task. For the ER implementation task, we analyzed individual differences in the capacity to reduce (a) ratings of subjectively experienced negative affect and (b) the amplitude of the late positive potential (LPP) in EEG with reappraisal vs. distraction. These measures of ER capacity were both positively correlated with the individual tendency to choose reappraisal as regulation strategy in the ER selection task – especially for emotional stimuli of high intensity. Regarding associations of ER with experimental measures of EF (inhibition, shifting, updating) and self-report measures of mental health-related personality traits findings were mixed and few correlations exceeded an effect size of .10. Overall, our findings suggest that in an experimental setting, individuals preferentially select the ER strategy that is most effective for them – both with respect to subjective experience as well as physiological responding. The talk will discuss why the empirical data do not consistently support theoretically expected associations with executive function and mental health.

Cognitive flexibility, the ability to adapt to changing task demands by switching between mental sets, is a key element of human behavior. It is discussed to contribute to individual differences in higher-order cognitive abilities. Studying individual differences in neural correlates of flexibility can help us unravel the mechanisms underlying cognitive flexibility. The present study investigated individual differences in event-related spectral perturbations via time-frequency analysis in three distinct cued task-switching paradigms (N = 148). We employed a data-driven approach, combining mass-univariate analyses with cluster-based permutation testing. Our analyses showed condition-specific differences in a cue- and target-related theta increase and alpha decrease, as well as in a target-related beta decrease. Using a latent change model, we found that only the enhanced parietal alpha decrease in switch compared to repeat trials in response to the cue and the target, loaded onto a common flexibility factor across tasks. Neither the greater theta increase nor the greater beta decrease in switch compared to repeat trials shared common variance across tasks, indicating that they do not reflect a consistent process. Furthermore, we observed no correlation between the latent flexibility factor and higher-order cognitive abilities. These findings challenge the notion that cognitive flexibility plays a significant role in individual differences observed in higher-order cognitive abilities. However, our analyses contribute to the broader understanding of cognitive flexibility by confirming the generalizability of the results across three distinct tasks and the specificity of the contributing frequency bands.

White matter (WM) microstructure is a candidate brain property underlying individual differences in fluid intelligence, possibly facilitating faster information transfer within and across brain regions. To date, it remains uncertain whether MRI-derived markers of WM microstructure generalize across different tracts, enabling the modeling of general factors related with fluid intelligence in a latent variable framework. Using data of N = 150 participants from the ongoing Dortmund Vital Study (Gajewski et al., 2022), we applied confirmatory factor analysis to derive measurement models for markers of WM integrity, neurite density, and myelin content. We characterized their factor structure across ten functional clusters of 52 WM tracts from the HCP-1065 tract atlas. Hierarchical models with bifactors for the hemispheres proved optimal for all markers. Fluid intelligence measured with Raven’s Progressive Matrices 2 was significantly explained by WM integrity (β = 0.46, p = .002) and myelin content (β = 0.20, p = .021). By establishing measurement models for WM, this study provides a framework for using MRI-derived markers in individual differences research while simultaneously shedding light on the biological basis of intelligence.

The panel discussion will integrate insights from the four symposium talks, focusing on how increased specificity in psychological constructs, neural measures, and analytic strategies enhances the mapping between brain and behavior across emotion and cognition. As a conceptual anchor, I will introduce the Brunswik Symmetry principle (Wittmann & Süß, 1988), which posits that empirical correlations are attenuated when measures are drawn from mismatched levels of hierarchical constructs. This framework highlights the importance of aligning levels of generalization – e.g., broad traits vs. specific neural responses – to accurately capture brain–behavior associations. We will explore how this principle applies to research on the neurocognitive basis of personality, and how it can guide the selection of appropriate levels of analysis. The panel will also address key methodological challenges, including multiverse analyses, measurement fidelity, and sample size constraints, and will outline future directions for integrating across levels of analysis. Audience participation will be encouraged to stimulate discussion around the conceptual and practical implications of high-resolution approaches in personality neuroscience.