In this project, the investigators examine behavior and associated brain activity during explore-exploit decision-making tasks performed pre- and post-modulation of affective state using autobiographical memory recall. The investigators hypothesize that a positive memory recall will reduce negative affective state, reduce explore-exploit biases and normalize the associated brain activity. The investigators propose a randomized double-blind, sham-controlled trial of positive autobiographical memory recall with 80 adults (n=40 per arm) with methamphetamine use disorder (MUD) currently involved in abstinence only treatment centers.
The growing epidemic of methamphetamine use disorder (MUD) is a significant burden on public health with surging overdose deaths, high likelihood of relapse, and current lack of approved medication to treat the disorder. When it comes to decision-making, individuals with MUD often prioritize drug over non-drug rewards despite negative life consequences; in addition, they may not sufficiently "explore" all available choices to "exploit" the best one (in other words, make the optimal choice leading to positive consequences). Therefore, the "explore-exploit" trade-off is often dysfunctional in MUD. Decision-making imbalances in the explore-exploit trade-off may extend well into abstinence, a period marked by a negative affective state (low mood, high depression and anxiety, withdrawal), which in turn triggers heightened craving and subsequent drug use urges. The insula, anterior cingulate cortex and striatum are crucial brain regions involved in explore-exploit behaviors and affective state signaling that have also been linked to drug reward processing in MUD. We propose that reducing negative affective state (improving mood) could help normalize explore-exploit behaviors and the response of these brain areas in individuals currently abstinent from methamphetamine and other drugs. This project will use a non-drug-related autobiographical memory recall to improve the mood of individuals with MUD and measure whether it normalizes non-drug decision-making, using a functional magnetic resonance imaging-based 3-arm bandit task and a behavioral contextual reinforcement learning task. A mixed experimental design in n=80 (72 completers, assuming 10% attrition) allows the identification of a between-subjects effect of positive (n=40, 36 completers) vs. neutral (n=40, 36 completers) mood modulation and assess the within-subject impact on explore-exploit behaviors pre- versus post-mood modulation. Mood groups will be compared on positive and negative affect, and behavioral/brain responses to reward valuation, outcomes and learning rates. The overarching goal is to establish that improving mood in individuals with MUD can reduce their negative affective state, normalize outcome sensitivity in key brain regions and associated learning, and reduce the influence of drug rewards on the valuation of non-drug rewards. This approach of this proposal embodies the goals of the NIH RDoC Initiative and the NeuroMAP Center by identifying an actionable disease-modifying target (mood) and studying its effect on the cognitive and neural dysfunction underlying a specific cognitive process (explore-exploit behaviors) relevant to MUD, and possibly other related neuropsychiatric disorders. By targeting the intertwined mechanisms between negative affect and explore-exploit biases, innovative, effective intervention strategies for MUD may be unveiled, addressing a critical public health challenge.
Study Type
INTERVENTIONAL
Allocation
RANDOMIZED
Purpose
BASIC_SCIENCE
Masking
DOUBLE
Enrollment
80
Autobiographical memory recall designed to modulate mood and affective state by reminiscing about personal life events
Laureate Institute for Brain Research
Tulsa, Oklahoma, United States
RECRUITINGCurrent affective state
Use of PANAS-PA and PANAS-NA questionnaires to measure positive and negative affect respectively (scale range: 0=low; 60=high; low mood is defined as low positive and/or high negative affect)
Time frame: Change from pre-intervention (baseline) to post-intervention (90min later) on study day 1
Learning rates
Computational modeling is used to compute how fast (learning rate) participants update their choice strategy following appetitive and aversive outcomes during 3-arm bandit task
Time frame: Change from pre-intervention (baseline) to post-intervention (30min later) on study day 1
Response to punishment in insular cortex
BOLD signal is used to measure the activity in the insula as a predefined region of interest (ROI), using Brainnetome atlas, following the delivery of aversive outcomes during 3-arm bandit task
Time frame: Change from pre-intervention (baseline) to post-intervention (30min later) on study day 1
Correct choice rate of mid-value option in contexts 1 (M1) and 2 (M2) during the testing phase
Proportion of trials for which the mid-value option is correctly chosen when paired against a high- or low-value option during contextual RL task
Time frame: Change from pre-intervention (baseline) to post-intervention (60min later) on study day 1
∆BIC for absolute vs. relative value coding
Difference in model-fitting between the absolute and intrinsically enhanced or range adaptive (relative) models as measured by Bayesian Information Criterion (∆BIC) during contextual RL task
Time frame: Change from pre-intervention (baseline) to post-intervention (60min later) on study day 1
Response to punishment in anterior cingulate cortex (ACC)
BOLD signal is used to measure the activity in the ACC as a predefined ROI, using Brainnetome atlas, following the delivery of aversive outcomes during 3-arm bandit task
Time frame: Change from pre-intervention (baseline) to post-intervention (30min later) on study day 1
Response to punishment in striatum
BOLD signal is used to measure the activity in the striatum as a predefined ROI, using Brainnetome atlas, following the delivery of aversive outcomes during 3-arm bandit task
Time frame: Change from pre-intervention (baseline) to post-intervention (30min later) on study day 1
Influence of abstinence duration on value coding
Across participants, correlation of number of days since last use of amphetamine with ∆BIC during contextual RL task
Time frame: Change from pre-intervention (baseline) to post-intervention (60min later) on study day 1
Influence of craving on value coding
Across participants, correlation of intensity of craving for amphetamine with ∆BIC during contextual RL task
Time frame: Change from pre-intervention (baseline) to post-intervention (60min later) on study day 1
Influence of withdrawal on value coding
Across participants, correlation of intensity of withdrawal for amphetamine with ∆BIC during contextual RL task
Time frame: Change from pre-intervention (baseline) to post-intervention (60min later) on study day 1
This platform is for informational purposes only and does not constitute medical advice. Always consult a qualified healthcare professional.