Abstract
Sustaining visual attention requires numerous dynamic cognitive processes acting in concert. Consequently, a diverse set of factors underlie fluctuations in performance during a sustained attention task and identifying a specific cognitive cause from fMRI-based measures like mean activation or functional connectivity (FC) is rife with ambiguity. In the present study, we provide richer characterization of how large-scale neural networks relate to attentional fluctuations by considering novel information-rich measures that use Representational Similarity Analysis to quantify the fidelity and connectivity of stimulus-specific representations across a set of brain regions. Participants (N=145) performed the gradual onset Continuous Performance Task (gradCPT) during an fMRI scan. This entailed viewing a series of city or mountain scenes, responding to cities (90% of trials) and withholding responses to mountains (10%). Representational similarity matrices (RSMs), reflecting the similarity structure of the set of city exemplars (n=10) were used to quantify the representational fidelity (RF) and connectivity (RC) of the stimulus representations by computing the inter-participant reliability of RSMs within (RF) and across (RC) each brain network. Critically, we computed how changes in behavioral measures of attentional state related to changes in univariate activation, FC, RF, and RC. Examining the visual network, we found that better performance was associated with greater mean activation, greater RF, greater RC with the Dorsal Attention Network (DAN) but lower FC with the Default Mode Network (DMN; all p < 0.05). These results provide direct support for the notion that visual-DAN connectivity reflects the communication of stimulus-specific information while visual-DMN connectivity reflects distracting stimulus-unrelated information. Across various brain networks and measures of task performance, we demonstrated that considering multiple dynamic measures of neural processing--particularly representational similarity-based measures--can greatly enrich conclusions regarding the neural and cognitive processes that underlie attentional fluctuations.
Acknowledgement: This work was supported by the U.S. Department of Veteran Affairs through a Clinical Science Research and Development Career Development Award (Grant Number 1IK2CX000706- 01A2) to M. E.
URL:Confirm Conflict Disclosure Policy