Disrupted Anti-correlation Between the Default and Dorsal Attention Networks During Hyperthermia Exposure: An fMRI Study.

Environmental hyperthermia is a common risk factor for occupational safety in many situations due to decreased vigilance performances. Previously, we have reported that decreased resting-state functional connectivity within the default mode network (DMN) and decreased activations in dorsal attention network (DAN) such as dorsolateral prefrontal cortex (DLPFC) were correlated with selective attention deficits during hyperthermia. However, whether the inherent functionally organized anti-correlation between the DMN and DAN would contribute to the behavioral deficits remains unclear. In this study, we collected the resting-state fMRI data of 25 participants during two simulated thermal conditions: normothermic condition (25°C for 1 h) and hyperthermic condition (50°C for 1 h). Using group independent component analysis (ICA), we investigated the functional connectivity within the DMN and DAN, as well as the anti-correlations between both networks. Paired comparisons revealed that decreased intranetwork functional connectivity in the medial prefrontal cortex (mPFC)/anterior cingulate cortex (ACC) in the DMN contributed to executive control performance during hyperthermia using multivariate linear regression analysis. Paired comparison on the DAN showed that increased one in the posterior part of the middle and inferior temporal gyrus nearby the temporal-parietal junction area contributed to preserved alerting performance. Lastly but most importantly, we found that decreased correlation between mPFC in the DMN and intraparietal sulcus (IPS) area in the DAN contributed to the executive control deficit, suggesting a weaker intrinsic anti-correlation between DMN and DAN during hyperthermia. These findings indicated that a functional reorganized architecture of DMN and DAN might provide a potential neural basis of the selective deficits for different cognitive-demand attention tasks in high-temperature environments.