Nathaniel G. Harnett

Objective: Polygenic risk scores (PRS) for posttraumatic stress disorder (PTSD) often account for a low amount of variance. Ancestry-related differences in PRS scale and variance limit cross-group comparisons. This methodological challenge further complicates gene-by-environment (GxE) analyses, given that socioenvironmental exposures are inequitably distributed across ethnoracial groups. We constructed an ancestry-calibrated polygenic risk score (AC-PRS) for PTSD in the largest longitudinal study of trauma survivors to date and investigated GxE interactions. Method: Recent trauma survivors (N=1,801) provided a blood specimen for genotyping. Six PTSD trajectories were previously identified from PTSD Checklist for DSM-5 (PCL-5) scores at 2-weeks, 8-weeks, 3-months, and 6-months post-trauma. Greenspace (normalized difference vegetation index [NDVI) and socioeconomic disadvantage (area deprivation index [ADI]) were derived from residential addresses. Logistic regressions examined interactions between newly developed AC-PRS and neighborhood factors on trajectories after adjusting for sociodemographic and trauma-related covariates. Secondary linear models considered GxE interactions on 6-month PCL-5 scores. Results: AC-PRS performed well across ethnoracial groups, explaining significant variability in PTSD trajectories (R ² =.053). ADI moderated the association between AC-PRS and the likelihood of assignment in a high nonremitting trajectory of PTSD symptoms and severity of symptoms at 6-months ( ps < .05). There were no NDVI x AC-PRS interactions in any models. Conclusions: AC-PRS captures genetic risk for PTSD in admixed trauma survivors, demonstrating good discrimination between nonremitting and resilient courses of PTSD. However, neighborhood disadvantage may modify utility of PRS for PTSD, warranting careful consideration when applying these scores across contexts.

Residential segregation is associated with differential exposure to air pollution. Hippocampus structure and function are highly susceptible to pollutants and associated with posttraumatic stress disorder (PTSD) development. Therefore, we investigated associations between residential segregation, air pollutants, hippocampal neurobiology, and PTSD in recent trauma survivors. Participants (N = 278; 34% non-Hispanic white, 46% Non-Hispanic Black, 16% Hispanic) completed multimodal neuroimaging two weeks after trauma. Yearly averages of air pollutants (PM_2.5 and NO₂) and racial/economic segregation (Index of Concentration at the Extremes) were derived from each participant's address. Linear models assessed if air pollutants mediated associations between segregation and hippocampal volume, threat reactivity, or parahippocampal cingulum fractional anisotropy (FA) after covarying for age, sex, income, and 2-week PTSD symptoms. Further models evaluated if pollutants or segregation prospectively predicted PTSD symptoms six months post-trauma. We found that non-Hispanic Black participants lived in neighborhoods with significantly greater segregation and air pollution compared to Hispanic and non-Hispanic white participants (ps < 0.001). PM_2.5 concentration was positively correlated with threat reactivity (r(276) = 0.16, p < 0.006), while NO₂ concentration was positively correlated with hippocampus volume (r(276) = 0.17, p < 0.005) and negatively correlated with white matter tract FA (r(276) = -0.18, p < 0.003). There was a significant indirect effect of NO₂ between segregation and FA values (β = 0.08, 95% CI[0.01, 0.15]), and an indirect effect of PM_2.5 between segregation and threat reactivity (β = -0.08, 95% CI[-0.14, -0.01]). There was no direct effect of segregation on hippocampal features. Pollutants and segregation were not associated with PTSD symptoms. In conclusion, residential segregation is associated with greater air pollution exposure, which is in turn associated with variability in hippocampal features among recent trauma survivors. Further research is needed to assess relationships between other environmental factors and trauma and stress-related disorders.

Visual cortex morphology is associated with posttraumatic stress disorder (PTSD) symptoms. However, limited research has investigated whether trauma exposure itself relates to visual cortex structure. Distinguishing trauma-related variability from PTSD-specific variability has important implications for identifying potential neural signatures of the disorder. We compared thickness of early visual areas (V1-V4v) in two independent samples (total n = 99) of recently trauma-exposed (RTE) individuals (Discovery n = 38, Replication n = 20) and individuals without recent (NRTE, Discovery n = 23) or no reported trauma exposure (NTE, Replication n = 18). We further investigated if visual cortex thickness was related to acute PTSD symptoms, and if trauma-related variability in thickness was specific to visual regions. We observed greater V1 thickness in the RTE compared to the NRTE/NTE group [Discovery: d = 0.84, p = 0.002; Replication: d = 0.67, p = 0.024]. V1 differences remained when including covariates such as sex, trauma load, current PTSD symptoms, and global cortical thickness. Ventral V4 thickness was positively associated with PTSD symptom severity in the discovery sample however this association was not observed in the replication dataset [Discovery: r = -0.49, p = 0.002; Replication: r = 0.08, p = 0.733]. The findings suggest trauma may acutely impact visual cortex morphology and may be partially independent of PTSD symptoms. Trauma-related variability in thickness was most observed in V1 which suggests this visual area may be key to understanding potential resilience or susceptibility to trauma-related psychopathology.

Background: Racial discrimination (RD) is a chronic stressor associated with increased risk for post-traumatic stress disorder (PTSD), a disorder associated with disruptions in neural network organization. However, the neural mechanisms linking RD to PTSD remain unclear. We examined whether RD is associated with network organization metrics, including modularity and clustering coefficient (CC), and whether network metrics influenced associations between RD and PTSD symptoms. Methods: Ninety adult (age range, 18-62) Black American women recruited for the Grady Trauma Project completed resting-state MRI along with measures of RD, trauma exposure and PTSD symptom severity. Network topology was examined for each of seven resting-state networks; adjacency matrices of each network were used to derive network modularity and CC. Partial correlations were conducted with RD and network metrics with covariates of age, trauma exposure and systemic inequities. Metrics that showed significant associations with RD were entered into moderation analyses with PTSD symptom clusters. Results: Greater RD exposure was associated with lower CC of the somatomotor network (SMN, r=-.318, p = .003). Moderation analysis revealed that RD associated with PTSD re-experiencing symptom severity at relatively lower (≤.48) [B=.58, CI (.22, .95), t = 3.20, p =.002] SMN CC values; this relationship was not observed at higher CC values ( p s>.05). Conclusion: Greater RD linked to lower clustering within the SMN, reflecting a shift toward more distributed network organization. Lower SMN clustering moderated associations between RD and PTSD re-experiencing symptoms. Findings suggest that more frequent RD may disturb the organization of networks responsible for the integration of external sensory and internal visceral signals, which, in turn, may influence the development of PTSD reliving phenomena.

Black individuals are disproportionately exposed to adverse childhood experiences (ACEs) in comparison to White individuals, including greater violence exposure, neighborhood disadvantage, and poverty. Neural circuitry that includes the amygdala, hippocampus, prefrontal cortex (ventromedial, dorsomedial, and dorsolateral), and inferior parietal lobule support stress‐related emotional processes. ACEs may modify emotion‐related activity within these brain regions, which, in turn, may modulate emotional behavior. However, the extent to which ACEs underlie race‐related differences in emotional function remains to be determined. Therefore, this study investigated whether stress‐elicited brain activity varies with race‐related differences in ACEs. Functional magnetic resonance imaging (fMRI) data from 301 Black and White participants were acquired during the Montreal Imaging Stress Task. Violence exposure, neighborhood disadvantage, and family income were measured during adolescence to assess ACEs, whereas stress‐elicited brain function was assessed in emerging adulthood. Behavioral (stress ratings) and psychophysiological data (skin conductance and heart rate) were collected alongside fMRI. Race‐related differences were observed in behavioral (stress ratings), psychophysiological (heart rate), and neural (fMRI) responses to stress. Further, a significant relationship was observed between stress reactivity and ACEs. Importantly, adjusting for ACEs reduced race‐related differences in stress reactivity (stress ratings and brain function), suggesting that the neurobehavioral response to stress may be shaped, in part, by ACEs. These findings provide new insight into the socioenvironmental factors that influence emotional function.

Stress-related conditions disrupt cognition and emotion regulation and can result in psychiatric illness, but the neural circuit-level changes that can explain these broad effects remain unclear. To address this issue, we paired population-grounded discovery with out-of-sample testing. Using resting-state fMRI from > 14,000 healthy adults in the UK Biobank, we derived connectivity profiles tied to cognition (reaction time, numeric memory), and proxies of emotion dysregulation (neuroticism, anhedonia). We then applied the profiles to a trauma-exposed cohort (N = 306) to assess symptom relevance. Associations with stress, anxiety, and depression concentrated in a subset of circuit motifs, two of which recurred: (1) hyperintegration between the default mode and control/limbic interfaces, as well as (2) hypointegration between the default mode and visual interfaces, both tracked higher symptom burden. Static and dynamic analyses converged on the precuneus as a critical hub: stronger precuneus–visual coupling and greater occupancy of a precuneus-engaged dynamic state were related to lower symptoms, whereas the opposing state was related to higher burden. This novel hybrid approach—discover-then-project—thus yielded interpretable markers of circuit dysfunction that generalized to post-trauma psychopathology. Furthermore, the approach identified the precuneus as a potential target for mechanistically informed interventions.

Racial discrimination is a common and potentially chronic psychosocial stressor that influences affective processing. Prior research suggests experiences of discrimination (EOD) can enhance anterior cingulate cortex (ACC) and amygdala reactivity to negative stimuli. It is unclear if frequent discrimination influences positive emotional processing especially during passive engagement with emotionally salient stimuli. This study explored EOD's influence on neural processing of positive stimuli, predicting a positive association with ACC response to rewarding images. 59 Black women, ages 18–65, from a community‐based sample were enrolled in a study assessing trauma and its influence on mental health. Participants completed fMRI scans viewing positive, negative, and neutral images. They reported subjective emotional responses to the affective images. Participants completed the Experiences of Discrimination interview, measuring frequency of racial discrimination, as well as reporting on nondiscrimination‐related traumatic events. EOD frequency was not associated with subjective ratings of emotional arousal or valence, nor amygdala, nucleus accumbens (NAc) or ACC responses to negative or positive stimuli. There was a significant positive association between EOD and the ventral lateral prefrontal cortex response to positive scenes, p = 0.04, but this did not withstand false discovery rate correction. Whole‐brain analyses revealed EOD frequency predicted lower right fusiform gyrus responses to positive > neutral images (p_FDR0.05). Findings suggest that greater racial discrimination frequency was associated with lower response in the fusiform gyrus‐ which is involved with processing of face stimuli and specifically sensitive to race‐ in response to positive images. Results indicate that past EOD influences visual sensitivity to positively valenced visual stimuli.

Emotion processing engages multiple large-scale brain networks. However, prior investigations relying on a priori, contrast-based models of brain activity obscure networks’ distinct temporal dynamics and roles in task performance. Here, we performed tensor independent component analysis to identify and track concurrent brain processes, including those with non-canonical dynamics, during the emotional face matching task (EFMT) in healthy young adults (n = 413; n = 416 replication). Ten EFMT-recruited large-scale brain networks were identified, reflecting flexible recoupling of visual association cortex to diverse non-visual networks. These networks collectively engaged 74% of cortex and more strongly explained variability in cognition and a performance-based index of emotion interference than contrast-based amygdala activation/connectivity. Variability in EFMT-recruited network activity was more strongly linked to variability in cognition than affect. Findings reveal a rich landscape of brain activity under the surface of contrast-based fMRI analyses and deepen insights into the distinct brain processes underlying subcomponents of emotional face processing.