Identifying Neural Signatures of Racial Discrimination in Black individuals with a Multivariate Data Fusion Approach

Emory University, Atlanta GA

Investigators

Abstract

Project Summary The current proposal utilizes an innovative, multivariate data fusion approach that combines ecological momentary assessment of racial discrimination (RD) and multimodal human neuroimaging to characterize neural network alterations related to RD exposure and mental coping in Black Americans. Racial discrimination (RD) is embedded in the daily lives of Black Americans and consistently linked to brain health disparities. However, the neural mechanisms through which RD—both exposure to RD and mental coping responses—increases risk for these brain health disparities are unclear. Few studies have investigated relationships between RD exposure and cognitive and emotional facets of response and variance in brain networks. Further, extant research has used univariate methods and retrospective RD reports. Ecological Momentary Assessment (EMA) is a direct, ecologically-valid way to assay RD experiences and mental coping responses in real time. Multimodal, multivariate approaches such as multiset canonical correlation analysis + joint-independent component analysis (mCCA + jICA) with reference can incorporate EMA data with different neuroimaging methods in the same analysis, permitting the identification of cross-modal neural network differences that distinguish unique features of RD. This approach is ideal for capturing high-dimensional relationships between RD exposure and emotional/cognitive coping responses, modeling variance likely to be missed by unimodal techniques. As such, mCCA + jICA is a comprehensive data fusion method that can be used to identify unique RD-related brain signatures. These biomarkers may be used as targets for neuromodulation. Our objective is to use this data fusion approach, merging EMA with several magnetic resonance imaging (MRI) modalities, including task-based and resting-state functional MRI as well as structural MRI assessing features of white and gray matter morphology (diffusion-weighted imaging, T1-weighted imaging) to investigate associations between RD exposure and mental coping and structural and functional covariance in neural networks. We will recruit 220 Black Americans from the metropolitan Atlanta, Georgia area. Given that many participants will have experienced a range of other stressors, structural inequities, prior trauma and post-trauma psychopathology, we will include these variables in analyses as covariates; we will also examine interactions of RD with structural inequities/stressors in secondary analyses. The findings of this study can provide reliable and valid RD biomarkers and targets for neuromodulation. This may contribute significantly to our current understanding of how RD gets “under the skin,” to affect brain networks and enhance risk for brain health disparities.