Gracie Grimsrud


Polyneuro risk scores: testing a multivariate approach to study functional connectivity involved in executive functioning

Executive function (EF) encompasses processes involved in the control of rule-driven behavior and cognition involved in task navigation. Variation in EF is associated with developmental disorders like ADHD. Understanding how the brain manifests EF could unveil potential mechanisms that can aid in developing new interventions. Most prior research focuses on finding specific circuits or regions associated with EF via mass univariate testing; these study results vary considerably. Because of this, we suspect that EF mechanisms may be distributed across the whole brain rather than localized to a specific circuit or region. To test this hypothesis, we developed a polyneuro risk score (PNRS) of EF that can be used to study functional connectivity distributed as a whole across the brain in relation to EF. PNRSs are generated using multivariate testing and can leverage the whole brain. The PNRS was developed using a split-half sample from the Adolescent Brain Cognitive Development (ABCD) study (N=11,877). Across all motion criteria thresholds and brain parcellations for which the PNRS was assessed, we show an effective model of calculating such scores to predict executive function. In all models, a greater variance was explained using combined data compared to individual features. This suggests brain-wide manifestation of EF across multiple circuits. These findings support the notion that an individual brain region or circuit does not fully instantiate EF control. Rather, the polyneuro risk score approach for brain-wide associations predicts that small relationships that span connections across the whole brain account for the most variance in EF.

Video file