Storm on predictive brain: A neurocomputational account of ketamine antidepressant effect

Hugo Bottemanne, Lucie Berkovitch, Christophe Gauld, Alexander Balcerac, Liane Schmidt, Stephane Mouchabac, Philippe Fossati
Neuroscience & Biobehavioral Reviews. 2023-11-01; 154: 105410
DOI: 10.1016/j.neubiorev.2023.105410

PubMed

Lire sur PubMed

Bottemanne H(1), Berkovitch L(2), Gauld C(3), Balcerac A(4), Schmidt L(5),
Mouchabac S(6), Fossati P(7).

Author information:
(1)Paris Brain Institute – Institut du Cerveau (ICM), UMR 7225 / UMRS 1127,
Sorbonne University / CNRS / INSERM, Paris, France; Sorbonne University,
Department of Philosophy, Science Norm Democracy Research Unit, UMR, 8011,
Paris, France; Sorbonne University, Department of Psychiatry, Pitié-Salpêtrière
Hospital, Assistance Publique-Hôpitaux de Paris (AP-HP), Paris, France.
Electronic address: .
(2)Saclay CEA Centre, Neurospin, Gif-Sur-Yvette Cedex, France; Department of
Psychiatry, GHU Paris Psychiatrie et Neurosciences, Service
Hospitalo-Universitaire, Paris, France.
(3)Department of Child Psychiatry, CHU de Lyon, F-69000 Lyon, France; Institut
des Sciences Cognitives Marc Jeannerod, UMR 5229 CNRS & Université Claude
Bernard Lyon 1, F-69000 Lyon, France.
(4)Paris Brain Institute – Institut du Cerveau (ICM), UMR 7225 / UMRS 1127,
Sorbonne University / CNRS / INSERM, Paris, France; Sorbonne University,
Department of Neurology, Pitié-Salpêtrière Hospital, Assistance
Publique-Hôpitaux de Paris (AP-HP), Paris, France.
(5)Paris Brain Institute – Institut du Cerveau (ICM), UMR 7225 / UMRS 1127,
Sorbonne University / CNRS / INSERM, Paris, France.
(6)Paris Brain Institute – Institut du Cerveau (ICM), UMR 7225 / UMRS 1127,
Sorbonne University / CNRS / INSERM, Paris, France; Sorbonne University,
Department of Psychiatry, Saint-Antoine Hospital, Assistance Publique-Hôpitaux
de Paris (AP-HP), Paris, France.
(7)Paris Brain Institute – Institut du Cerveau (ICM), UMR 7225 / UMRS 1127,
Sorbonne University / CNRS / INSERM, Paris, France; Sorbonne University,
Department of Philosophy, Science Norm Democracy Research Unit, UMR, 8011,
Paris, France.

For the past decade, ketamine, an N-methyl-D-aspartate receptor (NMDAr)
antagonist, has been considered a promising treatment for major depressive
disorder (MDD). Unlike the delayed effect of monoaminergic treatment, ketamine
may produce fast-acting antidepressant effects hours after a single
administration at subanesthetic dose. Along with these antidepressant effects,
it may also induce transient dissociative (disturbing of the sense of self and
reality) symptoms during acute administration which resolve within hours. To
understand ketamine’s rapid-acting antidepressant effect, several biological
hypotheses have been explored, but despite these promising avenues, there is a
lack of model to understand the timeframe of antidepressant and dissociative
effects of ketamine. In this article, we propose a neurocomputational account of
ketamine’s antidepressant and dissociative effects based on the Predictive
Processing (PP) theory, a framework for cognitive and sensory processing. PP
theory suggests that the brain produces top-down predictions to process incoming
sensory signals, and generates bottom-up prediction errors (PEs) which are then
used to update predictions. This iterative dynamic neural process would relies
on N-methyl-D-aspartate (NMDAr) and
α-amino-3-hydroxy-5-methyl-4-isoxazole-propionic receptors (AMPAr), two major
component of the glutamatergic signaling. Furthermore, it has been suggested
that MDD is characterized by over-rigid predictions which cannot be updated by
the PEs, leading to miscalibration of hierarchical inference and
self-reinforcing negative feedback loops. Based on former empirical studies
using behavioral paradigms, neurophysiological recordings, and computational
modeling, we suggest that ketamine impairs top-down predictions by blocking NMDA
receptors, and enhances presynaptic glutamate release and PEs, producing
transient dissociative symptoms and fast-acting antidepressant effect in hours
following acute administration. Moreover, we present data showing that ketamine
may enhance a delayed neural plasticity pathways through AMPAr potentiation,
triggering a prolonged antidepressant effect up to seven days for unique
administration. Taken together, the two sides of antidepressant effects with
distinct timeframe could constitute the keystone of antidepressant properties of
ketamine. These PP disturbances may also participate to a ketamine-induced time
window of mental flexibility, which can be used to improve the psychotherapeutic
process. Finally, these proposals could be used as a theoretical framework for
future research into fast-acting antidepressants, and combination with existing
antidepressant and psychotherapy.