Intrinsic Network Static and Dynamic Functional Connectivity Associated With Induced Affect

The brain’s processing of affective states involves complex interactions across brain regions, rather than localized to a specific region. The examination of the neural mechanisms of affect has shifted from a functional localization approach towards a distributed network neuroscience. Participants (N=73) underwent fMRI while engaging in a novel affect-induction task where they mentally visualized neutral, positive, and negative self-nominated acquaintances. We examined the topology of brain networks relevant to affective processing (i.e., the default mode [DMN], salience [SN], ventral attention [VAN], dorsal attention [DAN], and frontoparietal [FPN] networks). For each network in each task condition, we used graph-theory metrics to compute their integration with the rest of the brain (using participation coefficient; PC) and within-network coherence (using within-module degree; WD). Both static (i.e., mean) and dynamic (i.e., standard deviation) PC and WD were computed for each network. Several networks showed differences in static PC between affective conditions, while the VAN displayed differences in dynamic PC between conditions. FC of the networks during the tasks were also associated with subjective affect reported immediately after each task. These results are consistent with the constructionist account that emotions emerge from flexible network configurations and contribute to growing network-based understandings of affective processing.