Dark control: The default mode network as a reinforcement learning agent

Abstract

The default mode network (DMN) is believed to subserve the baseline mental activity in humans. Its higher energy consumption compared to other brain networks and its intimate coupling with conscious awareness are both pointing to an unknown overarching function. Many research streams speak in favor of an evolutionarily adaptive role in envisioning experience to anticipate the future. In the present work, we propose a process model that tries to explain how the DMN may implement continuous evaluation and prediction of the environment to guide behavior. The main purpose of DMN activity, we argue, may be described by Markov decision processes that optimize action policies via value estimates through vicarious trial and error. Our formal perspective on DMN function naturally accommodates as special cases previous interpretations based on (a) predictive coding, (b) semantic associations, and (c) a sentinel role. Moreover, this process model for the neural optimization of complex behavior in the DMN offers parsimonious explanations for recent experimental findings in animals and humans.

The surprising role of the default mode network in naturalistic perception

“…In this case, surprise may lead to switching between unexpected incoming information (external mode), memory of previous events and our internal model (internal modes), as we integrate across all 3. A similar switching role has been recently hypothesized particularly for the DMN node corresponding to the right AG 31 , which is most prominent in its link to surprise in the current study. Altogether, this suggests that coactivation of DMN and subcortical regions support integration across external information and internal representations, as we experience the mental state of surprise.…”

Preferential activation of the posterior Default-Mode Network with sequentially predictable task switches

“…In concert, they are particularly well positioned to monitor and predict environmental changes. There, the PCC/PrCC has been highlighted as sub-serving a change-detection role in an internally-driven manner [5,20,47,58]. The current empirical results are compatible with that perspective, insofar as the participants monitor, learn and predict the sequential structure of switch events.…”

Deep learning identifies partially overlapping subnetworks in the human social brain

“…Neural activity responses in the NAC are not usually thought to encode differences in intentions of the interaction partner per se 26 , 29 , 30 . Instead, perspective-taking processes are typically attributed to a set of higher-level cortical regions with prominent involvement of both the TPJ and mPFC.…”