Investigating value-based learning and decision making and associated neural BOLD dynamics during esketamine treatment in patients with difficult to treat depression.

Abstract

This project builds on the hypothesis that esketamine treatment elicits its clinical effect in patients with difficult to treat depression (DTD), at least in part, via modulation of the monoaminergic, especially dopaminergic and serotonergic systems by upstream modulation of the lateral habenula. The lateral habenula has an (indirect) inhibitory connection to the substantia nigra pars compacta and ventral tegmental area (SNc-VTA) complex — the primary source of dopamine in the human brain — and the dorsal raphé, the primary source of serotonin. Dopaminergic and serotonergic signaling within fronto-limbic circuits — including the striatum, amygdala, medial prefrontal cortex (mPFC), and anterior insula — are involved in various aspects of value-based learning and decision making relevant to DTD. Specifically, these include reward and punishment outcome processing, prediction error coding, and value updating, all of which contribute to behavioral control. Depression has been associated with hypersensitivity to punishment, hyposensitivity to reward, blunted reward prediction error, lower learning rates and altered effects of valence on behavioral inhibition and approach-avoidance behaviors. Human fMRI studies employing value-based decision-making tasks in DTD patients have identified associated neural patterns, specifically aberrant prediction error coding in the habenula and (possibly related) blunted signal in the ventral striatum as well as amygdala and insula hyperreactivity to aversive stimuli, altered value signals in the mPFC, and reduced VTA-striatal connectivity during outcome presentation. Here, we will employ a multi-phase reinforcement learning and decision paradigm including instrumental learning, Pavlovian conditioning, and Pavlovian-to-Instrumental Transfer (PIT) phases, in a functional magnetic resonance imaging (fMRI) context to assess the effect of esketamine treatment in patients with DTD on value-based learning and decision making and associated neural BOLD dynamics. Our primary neural analyses will examine whether habenula response to the presentation of shock-related conditioned stimuli (CS) predicts the effect of four weeks of esketamine treatment on depression symptom severity. Beyond neural responses, behavioral indices of PIT may provide an accessible read-out of habenula-dependent dopaminergic dysfunction. In particular, action-specific PIT — where aversive CS (relative to appetitive or neutral stimuli) bias instrumental responding — has been shown to depend on dopaminergic processing. Given the habenula’s central role in modulating dopaminergic signaling, aberrant action-specific PIT may serve as a behavioral marker of habenula-related dysfunction in difficult-to-treat depression. Accordingly, our primary behavioral analyses will assess whether action-specific PIT, measured prior to treatment initiation, predicts changes in depression symptom severity over four weeks of esketamine treatment. Secondarily, the study has a broader aim of linking fMRI-derived neurocomputational phenotypes to esketamine treatment outcomes in DTD.

Topics & conditions

Detected automatically by Psychedelic Commons from the title and abstract, not labels from the source. They may be incomplete.

Keywords