Overview = This project combines data, code, and documentation for ***Research Objective #2*** of the National Science Foundation funded study (Award# BCS-173582) titled "Cognitive Control Theoretic Mechanisms of Real-time fMRI-guided Neuromodulation". ***Note: This project extends the work of NSF Award #BCS-173582 Research Objective #1, "Action-value processing underlies the role of the dorsal anterior cingulate cortex in performance monitoring during self-regulation of affect", which may be found at https://osf.io/jwv6c/*** Research Objective = We sought to characterize the neural mechanisms driving affect processing dynamics in the resting (i.e., untasked) human brain. Methods Summary = - We constructed high-accuracy affect (valence and arousal) decoding models (using SVMs) from fMRI data recorded during affect induction via IAPS image stimuli. - We then applied these decoding models out-of-sample to resting state fMRI data. - We computed 1st and 2nd derivative time-series of the resting state affect processing. - We constructed and validated neural decoding models of these derivatives - We then simulated resting data affect processing dynamics via numerical integration of decoded affective processing derivatives. - We applied the Haufe transform to the decoding models to characterize the neural encodings of these derivatives. Results = - We showed that simulated resting state affect processing is significantly more accurate in reflecting affect processing dynamics than surrogate dynamics (simulation using randomly sampled derivatives) up to 4 volumes (TR=2 s) into the future. - We showed that affect processing dynamics are broadly encoded across the human brain in regions associated with affect processing and regulation.