Changes in Model-Based and Model-Free Control Predict Future Drinking Trajectories in Young Adults

Poster No:

584 

Submission Type:

Abstract Submission 

Authors:

Hao Chen1, Sören Kuitunen-Paul2, Maria Garbusow3, Quentin Huys4, Andreas Heinz3, Michael Rapp5, Michael Smolka1

Institutions:

1Department of Psychiatry and Psychotherapy, Technische Universität Dresden, Dresden, Germany, 2Chair of Clinical Psychology and Psychotherapy, Technische Universität Chemnitz, Chemnitz, Germany, 3Department of Psychiatry and Psychotherapy, Charité – Universitätsmedizin Berlin, Berlin, Germany, 4Division of Psychiatry; Max Planck UCL Centre for Computational Psychiatry and Ageing Research, UCL, London, UK, 5Area of Excellence Cognitive Sciences, University of Potsdam, Postdam, Germany

First Author:

Hao Chen  
Department of Psychiatry and Psychotherapy, Technische Universität Dresden
Dresden, Germany

Co-Author(s):

Sören Kuitunen-Paul  
Chair of Clinical Psychology and Psychotherapy, Technische Universität Chemnitz
Chemnitz, Germany
Maria Garbusow  
Department of Psychiatry and Psychotherapy, Charité – Universitätsmedizin Berlin
Berlin, Germany
Quentin Huys  
Division of Psychiatry; Max Planck UCL Centre for Computational Psychiatry and Ageing Research, UCL
London, UK
Andreas Heinz  
Department of Psychiatry and Psychotherapy, Charité – Universitätsmedizin Berlin
Berlin, Germany
Michael Rapp  
Area of Excellence Cognitive Sciences, University of Potsdam
Postdam, Germany
Michael Smolka  
Department of Psychiatry and Psychotherapy, Technische Universität Dresden
Dresden, Germany

Introduction:

Building on our previous findings on the link between imbalanced model-based/model-free (MB/MF) control and the development of drinking behaviour between ages 18 and 21, this study investigates whether changes in MB/MF control over these years are consequences of alcohol consumption or whether they consistently act as predisposing factors, potentially predicting drinking patterns from ages 21 to 24.

Methods:

In this six-year longitudinal study, we tracked a community sample of young men, starting at age 18. Participants underwent functional magnetic resonance imaging at ages 18 and 21 while performing a two-step sequential decision-making task to evaluate imbalances in MB and MF control. Drinking behaviours were annually assessed through an interview-based binge drinking score, measured in grams of alcohol per occasion, and biannually through a self-reported consumption score using the Alcohol Use Disorders Identification Test. To assess the effects of drinking on MB/MF control, we analysed total alcohol consumption between ages 18 and 21 using area under the curve calculations and examined its correlation with changes in the two-step decision-making parameters. To investigate how MB/MF control predicts future drinking behaviour, latent growth curve models were employed to explore the influence of these parameters on drinking trajectories from ages 21 to 24.

Results:

We found no associations between the total amount of alcohol consumption from ages 18 to 21 and any change in the parameters of the two-step task, indicating that drinking behaviour does not affect the imbalance in MB/MF control. The MB behavioural score exhibited a protective effect against the development of the binge drinking score: a higher MB behavioural score was associated with diminished progression in the binge drinking score. In addition, an increased MF behavioural score was linked to a higher binge drinking score at age 21, but it did not predict further development of binge drinking. In terms of the consumption score, as expected, we observed that an enhanced MF reward prediction error (RPE) in the ventral striatum (VS) was associated with escalated consumption score development. However, the MF RPE signal in the ventromedial prefrontal cortex (vmPFC) showed an inverse association with the development of the consumption score. This unexpected association was elucidated by further mediation analysis, which indicated that the MF RPE signal in the vmPFC partially mediated the relationship between the MF RPE in the VS and the development of the consumption score.
Supporting Image: LGCM_graph_peroccasion_behavioral_change.png
   ·A higher MB behavioural score was associated with diminished progression in the binge drinking score, while an increased MF behavioural score was linked to a higher binge drinking score at age 21.
 

Conclusions:

Our findings suggest that changes in MB behavioural control and MF RPE signals are not consequences of earlier drinking, but rather they predispose individuals to the development of drinking behaviours from ages 21 to 24. These findings suggest potential avenues for developing targeted prevention strategies that modulate MB and MF controls to mitigate the risk of future drinking behaviours.

Disorders of the Nervous System:

Psychiatric (eg. Depression, Anxiety, Schizophrenia) 1

Higher Cognitive Functions:

Decision Making 2

Learning and Memory:

Learning and Memory Other

Modeling and Analysis Methods:

Activation (eg. BOLD task-fMRI)

Keywords:

Addictions
FUNCTIONAL MRI
Learning
Other - Reinforcement Learning; Risky Drinking; Longitudinal

1|2Indicates the priority used for review

Provide references using author date format

Chen, H., Mojtahedzadeh, N., Belanger, M. J., Nebe, S., Kuitunen-Paul, S., Sebold, M., Garbusow, M., Huys, Q. J. M., Heinz, A., Rapp, M. A., & Smolka, M. N. (2021). Model-Based and Model-Free Control Predicts Alcohol Consumption Developmental Trajectory in Young Adults: A 3-Year Prospective Study. Biological psychiatry, 89(10), 980–989. https://doi.org/10.1016/j.biopsych.2021.01.009