Neuroticism, a stable personality trait marked by heightened negative affect and emotional volatility, is a wellestablished transdiagnostic risk factor for internalizing psychopathology. While early research emphasized amygdala hyperreactivity as a core neural correlate, emerging evidence suggests that neuroticism may be more accurately characterized by dysfunctional connectivity between the amygdala and broader regulatory networks involved in emotion processing and cognitive control. In this cross-sectional fMRI study, 115 healthy adults completed a classification task involving negative emotional facial expressions. Neuroticism was assessed using a latent factor score derived from five validated self-report instruments. Brain activity and psychophysiological interaction analyses were conducted using both region-of-interest and whole-brain approaches. Associations between neural measures and neuroticism were tested using robust regression, controlling for age and sex. No evidence was found for an association between neuroticism and regional brain activity. However, higher neuroticism was associated with increased task-dependent functional connectivity between the amygdala and both the hippocampus and dorsolateral prefrontal cortex. Whole-brain analyses further revealed associations between neuroticism and amygdala coupling with regions implicated in emotion regulation and salience processing, including the anterior insula and dorsal cingulate cortex. These findings support the conceptualization of neuroticism as a network-level phenomenon, characterized by dysregulated interactions within fronto-limbic and salience circuits, rather than by localized changes in brain activity. Specifically, increased amygdala-hippocampal and amygdala-prefrontal connectivity may underlie the persistence and regulation difficulties of negative emotions that characterize the neurotic phenotype.

The ability to experience pleasure, known as hedonic capacity, is closely linked to reward anticipation and is regarded as a fundamental component of human motivation. The striatum plays a central role in neural reward processing and has further been linked to extraversion, a personality trait associated with positive affect and approach-related sensitivity to reward. This multimodal study thereby combined neural, behavioral and self-report data to examine whether hedonic capacity and extraversion are associated with striatal reactivity during social reward anticipation. To this end, 120 healthy participants completed the Social Incentive Delay paradigm during functional magnetic resonance imaging, in which anticipatory cues indicated the magnitude of social reward represented by smiling and laughing faces. Behavioral responding was indexed by target reaction times, and self-report measures included the Dimensional Anhedonia Rating Scale (DARS) for hedonic capacity and the Adult Temperament Questionnaire (ATQ) for extraversion. The results suggest an overlapping neural substrate of hedonic capacity and extraversion anchored in the left ventral and bilateral dorsal striatum. Hedonic capacity was most strongly associated with the left caudate and putamen, whereas extraversion was most strongly associated with the bilateral putamen. Together, these findings extend current models of striatal reward processing by indicating that both hedonic function and personality-related reward sensitivity are reflected in overlapping neural systems involved in social reward anticipation.

Resilience can be broadly defined as positive functioning despite adversity. Although often conceptualized as a stable trait-like characteristic, its neural correlates remain incompletely understood. Here, we present evidence from two large-scale neuroimaging studies capturing complementary aspects of resilience: a specific interpersonal form of resilience, defined as secure attachment despite childhood maltreatment, and a more general, outcome-based approach defining resilience as better-than-expected depressive symptom severity relative to cumulative risk. In a first study (N=1,317), we examined a specific interpersonal form of resilience, defined as secure attachment despite childhood maltreatment. Here, we found that secure attachment buffered the adverse effects of maltreatment on mental health and brain structure, with interpersonally resilient individuals showing more favorable clinical outcomes and greater gray matter volume in the left supramarginal gyrus, a region implicated in social cognition and self–other processing. In a second study (N=1,804; longitudinal subsample N=808), we used an integrative framework incorporating multiple risk and protective factors to derive individual resilience scores as better-than-expected depressive symptom severity relative to cumulative risk. While resilience showed no cross-sectional associations with brain structure, higher baseline resilience prospectively predicted lower gray matter volume in orbitofrontal and temporal regions at two-year follow-up, potentially reflecting greater neural efficiency or delayed biological costs. Ongoing analyses extend this framework to functional network organization. Together, these findings suggest that different aspects of resilience are linked to distinct neural correlates, which may partly emerge over time rather than reflecting stable predispositions.

Genome-wide association studies (GWAS) of personality traits are often underpowered, limiting insight into their molecular architecture. One approach to address this is to leverage genetic overlap with large-scale psychiatric GWAS to enhance statistical power. Here, we applied this framework to Hypomanic Personality, a dimensional trait marked by elevated mood, energy, and impulsivity that has long been proposed as vulnerability marker for bipolar disorder and schizophrenia. We conducted a GWAS of Hypomanic Personality in 7,004 individuals of European ancestry, identifying a genome-wide significant locus at 2p13.3 (rs79707808; p = 7.8e-10) and estimating SNP-based heritability at 12% (SE = 6.6%), indicating a measurable polygenic basis. We observed substantial genetic correlations with schizophrenia (rg = 0.40, SE = 0.15) and bipolar disorder (rg = 0.24, SE = 0.12), consistent with shared genetic architecture. Leveraging this overlap, we conducted pleiotropy-informed analyses, increasing discovery to 10 loci, including intronic variants in CACNA1C, a cornerstone gene in psychiatric genomics, as well as loci implicated in synaptic and affective processes. To extend these findings, we computed polygenic scores in 341,335 UK Biobank participants and conducted a phenome-wide scan across >6,700 health traits. Hypomanic Personality polygenic risk was associated with risk-taking behavior, substance use, cognitive performance, affective symptoms, and cardiometabolic traits, underscoring its broad phenotypic footprint. Together, these findings position Hypomanic Personality as a biologically grounded, transdiagnostic construct capturing shared genetic liability across severe mental illness and normative behavioral variation. More broadly, they show how leveraging genetic overlap with large psychiatric cohorts enhances discovery in underpowered personality traits.

An increasing number of studies report null or weak associations between personality-related constructs—including traits and personality pathology—and biological or behavioral markers across diverse biomarker approaches. This talk examines potential reasons for these recurring null findings and argues that they are often more informative about conceptual and methodological constraints than about the absence of underlying relations. Two conceptual issues are highlighted. First, many experimental paradigms constitute strong situations that constrain behavior or physiological responses, thereby limiting individual variability across measurement domains. Second, consistent with Brunswik’s lens model, systematic mismatches may exist between the level of generality of predictors (e.g., broad personality constructs) and criteria (e.g., highly specific biological or behavioral indicators). In addition, two technical factors are discussed: limited reliability of commonly used measures and variance restriction due to typical sampling strategies. Together, these factors can substantially attenuate observable associations and contribute to the widespread occurrence of null findings in personality–biomarker research. While these issues are well known, the talk argues that they remain insufficiently integrated into research design, analysis, and interpretation, and calls for renewed attention to these constraints across biopsychological research.