Taste matters: lessons learned in translating a conditioned drug cue reactivity paradigm from the rodent to the human laboratory

Abstract

Bidirectional translation between human and animal models of drug-predictive cue reactivity may accelerate biomarker and treatment development for substance use disorders. We report findings from a pilot study in which de novo cue conditioning procedures from nonhuman animal models of drug (here: alcohol) cue reactivity were adapted for neuroimaging and psychophysiology experiments with human subjects. Participants were 10 healthy young adults (60% female, age 21-23 yr) who could undergo electroencephalogram (EEG) recordings and functional magnetic resonance imaging (fMRI) and endorsed binge drinking at least once in the past 3 months. Prior to the experiment, participants were informed that they would be consuming sips of a strong, sweet, mixed drink (10-15% ethanol v/v in 15% w/v sucrose). We hypothesized that prior real-world experience with alcohol, including anticipation of pleasurable post-ingestive psychoactive effects, might shape a person's hedonic and incentive responses to a new visual cue for receipt of an unfamiliar alcohol beverage. Consequently, we expected to observe conditioned liking of the alcohol-paired cue, motivated attention as indexed by the amplitude of the late positive potential (LPP) in cue-locked EEG segments, and reward-network engagement (fMRI). Participants also underwent visual cue pairings with sips of sugar water (15% w/v sucrose). Most participants showed conditioned disliking of the alcohol-paired cue. While they exhibited more motivated attention to the alcohol-paired than -unpaired control cue, only the left orbitofrontal cortex was activated more strongly to the alcohol-paired than -unpaired cue. In contrast, most participants showed conditioning liking of the sugar-paired cue and more motivated attention to it than its unpaired control cue as well as broader engagement of the brain's reward network (e.g., bilateral anterior cingulate, anterior insula, and orbitofrontal cortex). These findings demonstrated the feasibility of developing human laboratory experimental paradigms that closely mirror non-human animal models of alcohol and other drug cue conditioning. Yet the pilot study's unexpected results suggest that for orally administered drugs like alcohol, taste reactivity may matter as much for de novo cue conditioning in human subjects as in non-human animal models, even when those human subjects endorse prior experience with the drug or positive drug expectancies.