A Lifespan Brain-Based Signature of Sustained Attention
Loading...
Date
2023-05
Authors
Journal Title
Journal ISSN
Volume Title
Publisher
The Ohio State University
Abstract
Introduction: Sustained attention is involved in most cognitive processes, and is beginning to be conceptualized as the result of synchronous whole-brain activity. Recent studies in network neuroscience have found that data-driven machine learning methods like connectome-based predictive models (CPM) can be used to predict sustained attention using whole brain data in independent age groups, and can be used to predict related constructs like fluid cognitive ability and ADHD symptom severity. However, no study has generated whole-brain predictive models which are applicable across the entire adult lifespan.
Methods: The neural and behavioral data of 594 subjects across the adult lifespan (36-105 years old) were entered into the CPM framework. We derived a set of connections across the entire brain that are associated with performance (d') on a Go/NoGo task, controlling for age. We hypothesized that this model would retain within-sample predictive utility across the lifespan, and predict fluid cognitive ability as well.
Results: These models were significantly (Combined: ρ = .183, p < .001) predictive of sustained attention indexed by d'. Models associated with better (High: ρ = .216, p < .001) and poorer (Low: ρ = .130, p = .002) sustained attention also demonstrated significant predictive accuracy. The high sustained attention model remained predictive when applied to NIH Toolbox fluid cognition composite scores (ρ = .138; p < .001). The combined (ρ = .007; p = .873) and Low (ρ = -.377; p < .001) models did not predict fluid cognition composite scores.
Discussion: Our results show that whole-brain functional connectivity is predictive of sustained attention across the adult lifespan. It also appears that connectivity associated with better sustained attention may partially serve as a cornerstone of fluid cognitive neural processing. Future work might apply this model to novel subjects of healthy and clinical populations.
Description
Keywords
Functional Connectivity, Connectome Based Predictive Modeling, Sustained Attention, Fluid Cognition, Lifespan