Reversing the Atypical Valuation of Drug and Nondrug Rewards in Smokers Using Multimodal Neuroimaging

Abstract

Background: Chronic substance use can disrupt the reward function of the anterior cingulate cortex (ACC), biasing the ACC to favor goal-directed behaviors that converge on drug use. Here we used multimodal neuroimaging methods to ask whether modulating reward-related signaling in the ACC can reverse the atypical valuation of nondrug and drug rewards in abstinent smokers.

Methods: We first recorded functional magnetic resonance imaging data from 20 moderately dependent cigarette smokers (mean age = 25 years; no history of neuropsychiatric disorders), following an overnight period of abstinence, to identify regions of the left dorsal lateral prefrontal cortex associated with the anticipation of drug-related rewards (cigarette puff). Next, we recorded the reward positivity-an electrophysiological signal believed to index sensitivity of the ACC to rewards-while participants engaged in two feedback tasks to gain either monetary or cigarette rewards. Lastly, guided by functional magnetic resonance imaging data, a robotic arm positioned a repetitive transcranial magnetic stimulation coil over a subject-specific dorsal lateral prefrontal cortex target, and 50 repetitive transcranial magnetic stimulation pulses were delivered at 10 Hz (excitatory stimulation) immediately before each block of 10 trials of the money condition and at 1 Hz (inhibitory stimulation) before each block of 10 trials of the cigarette condition.

Results: Our findings show that abstained smokers exhibited a heightened reward positivity to cigarette rewards relative to monetary rewards, and by applying excitatory or inhibitory repetitive transcranial magnetic stimulation to a subject-specific frontal-cingulate reward pathway, this pattern of results was reversed.

Conclusions: By modulating how the brain links value to drug and nondrug rewards, novel brain-based treatments may finally be on the horizon.

Keywords: Addiction; Dopamine; Reinforcement learning; Reward positivity; Reward prediction error signals; fMRI; rTMS.