Anatomical dissociation of intracerebral signals for reward and punishment prediction errors in humans

Abstract

Whether maximizing rewards and minimizing punishments rely on distinct brain systems remains debated, given inconsistent results coming from human neuroimaging and animal electrophysiology studies. Bridging the gap across techniques, we recorded intracerebral activity from twenty participants while they performed an instrumental learning task. We found that both reward and punishment prediction errors (PE), estimated from computational modeling of choice behavior, correlate positively with broadband gamma activity (BGA) in several brain regions. In all cases, BGA scaled positively with the outcome (reward or punishment versus nothing) and negatively with the expectation (predictability of reward or punishment). However, reward PE were better signaled in some regions (such as the ventromedial prefrontal and lateral orbitofrontal cortex), and punishment PE in other regions (such as the anterior insula and dorsolateral prefrontal cortex). These regions might therefore belong to brain systems that differentially contribute to the repetition of rewarded choices and the avoidance of punished choices.

Integrated Amygdala, Orbitofrontal and Hippocampal Contributions to Reward and Loss Coding Revealed with Human Intracranial EEG

“…Intracranial recordings can fill an important niche in neuroscientific studies of reward as they have high spatial and temporal resolution, are a direct measure of neural activity and record from many of the regions important for reward processing without signal distortion either by the skull/meninges or air-filled chambers within the skull. Yet only a handful of studies have exploited these advantages for studying reward processing in humans with the majority looking at each region piecemeal (Gueguen, et al, 2021;Li et al, 2016;Ramayya et al, 2015;Saez et al, 2018;Vanni-Mercier et al, 2009) rather than in tandem (Jenison, 2014;Lopez-Persem et al, 2020;Zheng et al, 2017;2019). To date, intracranial data directly comparing human amygdala and OFC function and their connectivity in reward and motivation are limited (Jenison, 2014).…”

Group-level inference of information-based measures for the analyses of cognitive brain networks from neurophysiological data

“…We used simulated data to compare the group-level approaches and complementary methods for correcting for multiple comparisons. We then tested our approach on human data to reproduce and extend existing results based on MEG data acquired during a visuomotor task (Brovelli et al, 2017, 2015) and spatially sparse intracranial data for investigating neural correlates of probabilistic learning (Gueguen et al, 2021). Finally, we illustrated how this framework can be applied at the level of local neural activity, on pairwise functional connectivity links or on measures summarizing network properties.…”

“…For simplicity, the description above about the statistical framework is considering the participant as the fixed or random variables. The same description still holds if sessions are used instead or, in the context of intracranial data, the recording contacts across participants (Gueguen et al, 2021).…”

“…We then investigated outcome-related effects in intracranial HGA, previously observed using general linear models (GLM). Previous analyses revealed differently modulated activity during the reward and punishment conditions in the anterior insula (aINS) (Gueguen et al, 2021). The dataset was composed of twenty patients with medically-intractable epilepsy, implanted with five to seventeen stereotactic electroencephalography (sEEG) electrodes.…”

“…We tested the workflow on simulated data to compare fixed-versus random-effect group inference with test- or cluster-wise p-values correction. We then re-analyzed cortical high-gamma activity (HGA) data from different recording modalities: a MEG experiment (Brovelli et al, 2015), where we successfully retrieved the visuomotor network using information metrics from both IT and ML and an intracranial experiment (Gueguen et al, 2021), revealing the outcome-related activity of the anterior insula during probabilistic learning. Finally and as a contribution toward reproducibility, we also provide to the community the neuroinformatics tools wrapped in an open-source Python toolbox called Frites 2 ( Framework of Information Theoretical analysis of Electrophysiological data and Statistics ).…”