Prediction error correlates in the striosome-dopamine circuit emerge from information gain.

Resting-state scans are a central component of large-scale, international neuroimaging efforts aimed at mapping the human functional connectome in vivo, in both healthy and patient populations. This prominence of resting-state in cognitive and clinical neuroscience research is motivated by its experimental simplicity, short scanning time (which translates into cost effectiveness), richness of information, and low demands on participants (typical instructions might be “. Yet, despite its wide adoption, we still lack an understanding of the nature and biological purpose of the neural processes that give rise to the observed synchronized patterns of spontaneous activity during rest. This knowledge gap severely limits both our ability to interpret results and to develop reliable brain-based biomarkers of disease. Neural mechanisms previously suggested as explanatory phenomena for resting-state functional connectivity patterns (rsFC) include memory consolidation, learning, fluctuations in wakefulness, homeostatic processes needed to maintain the brain’s functional integrity, the replay and update of predictive models of the environment, processing of interoceptive signals from the rest of the body, and those associated with different aspects of the subjective experience participants undergo while being scanned. Of these, the role of the in-scanner subjective experience is perhaps the least established. This is because researchers rarely collect information about what participants think and feel inside the scanner. A few instances where such data is available suggest that some aspects of in-scanner experience can be related to resting-state signals. For example, reported levels of imagery correlate with the strength of resting-state fluctuations in perilingual cingulate cortex and the face fusiform area, and reported levels of comfort with that of somatosensory cortex; a higher propensity to focus in-scanner thoughts on current concerns (as opposed to future planning) has been linked to stronger betweenness centrality for the left middle frontal gyrus (a key component of the salience network); increased activity in the dorsal attention network to subjective reports of attentional control; increase activity in lateral fronto-parietal cortices to increases in external awareness; stronger connectivity between the default mode network and the control networks has been linked to reduced variability in the semantic content of thoughts; and the qualia of inner voices (i.e., inner hearing vs. inner speaking) was associated with differential patterns of spontaneous activity during rest in language related areas. There are also studies linking a limited number of brain states (i.e., recurrent resting-state whole brain activity patterns) to the amount of time subjects report solving problems about the future or having intrusive thoughts about the past; as well as to individual variability in FC between the visual and sensorimotor systems. Despite these initial reports, it is not yet established how strongly systematic differences in in-scanner experience influence rsFC patterns. This matters because unmodeled variance based on these differences could hide aberrant patterns of inter-regional communication linked to clinical symptoms and hinder efforts aimed at developing brain-based biomarkers of disease progression and therapeutic efficacy. In fact, some have argued for the use of naturalistic experiment designs in detriment of resting-state, due to the many interpretational challenges stemming from resting-state lacking a well-defined relationship with cognitive processes. Conversely, others have argued that variations in in-scanner experience might have a minimal or no modulatory effect at all on rsFC. To shed light on these debates, which are key for forecasting the true clinical potential of rsfMRI and understanding the role that in-scanner experience might play in common research applications of rsfMRI, we seek to answer three specific questions.