Ketamine disrupts naturalistic coding of working memory in primate lateral prefrontal cortex networks

Ketamine is a dissociative anesthetic drug, which has more recently emerged as a rapid-acting antidepressant. When acutely administered at subanesthetic doses, ketamine causes cognitive deficits like those observed in patients with schizophrenia, including impaired working memory. Although these effects have been linked to ketamine’s action as an N-methyl-D-aspartate receptor antagonist, it is unclear how synaptic alterations translate into changes in brain microcircuit function that ultimately influence cognition. Here, we administered ketamine to rhesus monkeys during a spatial working memory task set in a naturalistic virtual environment. Ketamine induced transient working memory deficits while sparing perceptual and motor skills. Working memory deficits were accompanied by decreased responses of fast spiking inhibitory interneurons and increased responses of broad spiking excitatory neurons in the lateral prefrontal cortex. This translated into a decrease in neuronal tuning and information encoded by neuronal populations about remembered locations. Our results demonstrate that ketamine differentially affects neuronal types in the neocortex; thus, it perturbs the excitation inhibition balance within prefrontal microcircuits and ultimately leads to selective working memory deficits.We thank registered veterinary technicians Kim Thomaes and Rhonda Kersten from the University of Western Ontario for their assistance in surgery and animal care; Guillaume Doucet from the University of Ottawa for technical assistance related to Unreal Development Kit; Maryam Nouri Kadijani from the University of Western Ontario for assisting with initial data exploration; Kevin Barker from Neuronitek for engineering equipment for our experiments; Jonathan C. Lau from the Division of Neurosurgery, University Hospital for providing advice regarding surgery and surgical planning; Matthew Leavitt, AI Resident at Facebook for access to MATLAB code related to polynomial plane fitting and advice on electrophysiological analysis. This work was supported by Canadian Institute of Health Research Project Grant; Natural Sciences and Engineering Research Council of Canada (NSERC); Ontario Graduate Scholarship; Jonathan & Joshua Memorial Graduate Scholarship in Mental Health Research. Chrysalis Foundation (London, Ontario). LP acknowledges salary support from the Tanna Schulich Endowment Chair for Neuroscience and Mental Health. RMB acknowledges support from MINECO (Spain; BFU2017-85936-P), the Howard Hughes Medical Institute (HHMI, ref 55008742), and the ICREA Academia (2016)