Psychotic-like experiences and neurobehavioral reward processing in adolescents who use cannabis

Importance: Cannabis use (CU) in adolescence is linked to increased risk of developing psychotic-like experiences (PLEs) that can lead to lifelong impairment. A candidate mechanism may be disruptions in reward processing supported by dopamine but this association has not yet been tested in adolescents with CU and PLEs. The objective of the study was to assess the interrelation between CU, PLEs, and neurobehavioral markers of reward processing (behavioral reward motivation, tissue iron signal [1/nT2*], an indicator of dopamine-related neurophysiology).

Methods: Participants were adolescents (aged 14 - 17) with either >11 lifetime cannabis episodes or <5 lifetime cannabis episodes (controls). CU quantity was assessed with the Daily Sessions, Frequency, Age of Onset, and Quantity of Cannabis Use Inventory. PLEs were assessed with the Achenbach Youth Self-Report Thought Problems Scale. Reward motivation (proportion of hard choices) was assessed using a behavioral task. 1/nT2* was derived from functional magnetic resonance imaging. Hypotheses, formulated during data collection as recognition grew of the risk factors of developing PLEs with CU, were: PLEs will magnify negative associations between CU and reward motivation; the positive association between greater CU and PLEs will depend on subcortical 1/nT2*.

Results: 81 participants included 47 control and 34 cannabis group

Participants: mean [SD] age, 15.53 [1.12]; 52 female (64%). There was a significant interaction between CU and PLEs on reward motivation (OR, 2.41; 95% CI, 1.10, 5.27, p=0.028), depending on reward amount and effort required. At high PLEs, the cannabis group, compared to controls, were less motivated to work for a reward when effort required was low and medium. Associations between CU quantity and PLEs depended on 1/nT2* (at high 1/nT2*, CU positively predicted PLEs). Daily concentrate hits interacted with 1/nT2* in the left putamen (β=51.58, 95% CI, 12.07, 91.09, p=0.01) and right thalamus (β=70.93, 95% CI, 12.32, 129.54, p =0.02), daily cannabis grams in the left putamen (β=73.34, 95% CI, 12.32, 129.54, p=0.02), right putamen (β=101.69, 95% CI, 41.72, 161.67, p=0.001), and right thalamus (β=102.38, 95% CI, 28.00, 176.77, p=0.01). Conclusions and Relevance: Adolescent CU was associated with alterations in dopamine-relevant neural and behavioral indices as a function of PLEs. Behaviorally, this manifested as less reward motivation and strategic action. Consistent with prior research on psychosis-spectrum vulnerability and dopamine-related neurophysiology, elevated 1/nT2* was observed to moderate the link between CU and PLEs in key subcortical brain regions. Longitudinal studies are needed to understand the timing and mechanism of neurobehavioral reward processing alterations.