This repository contains R scripts that used partial least squares correlation (PLSC) (Krishnan et al., 2011; McIntosh and Lobaugh, 2004 to identify latent, multivariate patterns of associations between the adolescent sociocultural environment (at the individual, family, and community levels) and resting state functional connectivity (rsFC). rsFC was chosen as a neurobiological metric of interest given (i) prior work linking resting state fluctuations and cultural variables (Constante et al., 2023; Luo et al., 2022; Meca et al., 2023), (ii) the known functional relevance of rsFC to self-referential and affiliative processing (Pintos Lobo et al., 2020; Raichle, 2015, and (iii) the flexibility of rsFC to examine the full range of brain systems in absence of external stimuli and/or task demands (Smith et al., 2009.
Additionally, for each significant latent dimension derived from PLSC, we conducted a regression analysis to determine how the identified sociocultural brain patterns may relate to behavioral risk/resilience factors, which will include measures of both physical health (e.g., physical inactivity, weight/obesity, sleep) and mental health (e.g., internalizing/externalizing symptoms).
Variables that were included in the PLSC analysis:
| Sociocultural Measures | Gordon Networks |
|---|---|
| MEIM-R youth | Auditory Network |
| VIA youth | Cingulo-opercular Network |
| MACV youth | Cingulo-parietal Network |
| MEIM-R caregiver | Default Network |
| VIA caregiver | Dorsal Attention Network |
| MACV caregiver | Fronto-parietal Network |
| Perceived discrimination | Retrosplenial Temporal Network |
| Community cohesion | Sensorimotor Hand Network |
| LED: Child opportunity index | Sensorimotor Mouth Network |
| LED: NaNDA disadvantage index | Salience Network |
| LED: Getis-Ord Gi* Statistic | Ventral Attention Network |
| Visual Network |