Global cortical arousal effects in fMRI reveal brain markers of state and trait anxiety.

To personalize the diagnosis and treatment of anxiety, there is a need to identify biological constructs that underlie self-reported symptoms. Notably, physiological responses and levels of arousal are constituents of anxiety, and have widespread ("global") effects on functional magnetic resonance imaging (fMRI) signals that may provide novel brain markers of anxiety. Here, we modeled autonomic physiological and cortical arousal signatures in fMRI data and determined whether these global fMRI components relate to measures of state and trait anxiety. Additionally, we tested if these global effects impact relationships between brain network connectivity and anxiety levels. We found significant relationships between fMRI global components and state/trait anxiety scores, identifying brain regions in which the strength of the global mean fMRI signal and the cortical arousal-related fMRI signal were associated with both state and trait anxiety. Notably, the resulting patterns exhibited substantial overlap with established large-scale functional networks, including the default-mode network. These findings indicate that global effects in fMRI signals hold valuable information that may reveal how cortical arousal is expressed in anxiety, and may provide a source of information that has previously been treated as a confound.