In Cocaine Dependence, Neural Prediction Errors During Loss Avoidance Are Increased With Cocaine Deprivation and Predict Drug Use

• DOI: 10.1016/j.bpsc.2018.07.009
• Publication Date: 2018-08-02T22:06:34Z
• Authors: John M. Wang; Lusha Zhu; Vanessa M. Brown; Richard De La Garza; Thomas Newton; Brooks King-Casas; Pearl H. Chiu
• Keywords: Adult; Male; Brain Mapping; Cross-Over Studies; Brain; Middle Aged; Models, Psychological; 16. Peace & justice; Magnetic Resonance Imaging; 3. Good health; Cocaine-Related Disorders; Avoidance Learning; Humans; Learning; Reinforcement, Psychology

In substance-dependent individuals, drug deprivation and use trigger divergent behavioral responses to environmental cues. These are consonant with data showing that short- long-term adaptations in dopamine signaling similarly sensitive state of use. The literature suggests a state-dependent role learning maintaining substance use; evidence linking both reinforcement addiction provides framework test this possibility.In randomized crossover design, 22 participants current cocaine disorder completed probabilistic loss-learning task during functional magnetic resonance imaging while on off (44 sessions). Another 54 without Axis I psychopathology served as secondary reference group. Within-drug paired-subjects' effects were assessed computational model-derived individual parameters. Model-based neuroimaging analyses evaluated neural signals. Relationships among rates (α+, α-), prediction error signals (δ+, δ-), use, desire assessed.During deprivation, cocaine-dependent individuals exhibited heightened positive (α+), (δ+) responses, association α+ δ+ responses. deprivation-enhanced specific successful loss avoidance, comparable psychiatric conditions, mediated relationship between chronicity cocaine.Neurocomputational status suggest by avoidance negative outcomes may contribute seeking deprivation. More generally, attention is important for delineating substrates addiction.