The alternated brain states in resting state after immoral decisions

Abstract

Studies have shown that the pattern of the brain in tasks can be effectively predicted from resting brain features. However, few studies explore how task could, in turn, influence resting state connectivity. This work attempts to understand this through the lens of immoral decisions. Using resting-state and task-state fMRI data collected before, during, and after an information-passing task involving dishonest choices with rewards, we employed the Hidden Markov Model (HMM) to capture the changes in the brain dynamics state. First, the HMM results identified 4 intrinsic brain states across both resting scans and task scans, including one reward network (state 1) and one control network (state 3). The reward network are associated with high dishonesty, while the control network are associated with low dishonesty. Importantly, task engagement in immoral behavior altered the post-task brain state. As participants engaged in persistent dishonesty, the network reconfigures itself to priorities the reward network and suppress the control network. Such suppression persists beyond the immediate task timeframe and leaves lingering signatures in the post-task resting scan. Contrary to the view of resting-state as a static baseline, our analysis showed the "non-resting" nature of post-task resting scans, where dishonesty could have an "after-effect" on neural dynamics. Together, these findings suggest that the post-moral decision brain operates not as a static system but as a set of dynamically shifting states, whose adaptive trajectories may underlie the lasting effect of decision behavior in the prior task.