Brain scans predicted with 100% accuracy when daily cannabis smokers chose to buy cannabis

A machine learning classifier identified distinct brain activation patterns that perfectly distinguished decisions to purchase cannabis from decisions to decline in daily smokers.

Bedi, Gillinder et al.·Neuropsychopharmacology : official publication of the American College of Neuropsychopharmacology·2015·Preliminary EvidenceRandomized Controlled Trial
RTHC-00913Randomized Controlled TrialPreliminary Evidence2015RETHINKTHC RESEARCH DATABASErethinkthc.com/research

Quick Facts

Study Type
Randomized Controlled Trial
Evidence
Preliminary Evidence
Sample
N=17

What This Study Found

Researchers combined brain imaging with a real-world-like purchasing task where daily cannabis smokers made repeated decisions to buy or decline puffs of cannabis at various prices. One randomly selected decision was actually carried out: if they had chosen to buy, they paid and smoked in the lab.

Using machine learning, a classifier achieved 100% accuracy distinguishing brain patterns during purchase decisions from decline decisions at the individual level. Key brain regions contributing to this neural signature included the dorsal striatum, insula, posterior parietal regions, anterior and posterior cingulate, and dorsolateral prefrontal cortex.

Purchasing behavior followed expected patterns: participants bought more puffs at lower prices and of active cannabis compared to placebo. The findings provide a brain-based framework for understanding drug-related decision making.

Key Numbers

21 daily cannabis smokers participated. 17 purchased cannabis and were included in fMRI analysis. Machine learning classifier achieved 100% accuracy at the individual level. 6 brain regions reliably contributed to the decision signature.

How They Did This

Functional MRI study with a within-subject choice task. Twenty-one daily cannabis smokers made repeated decisions to purchase or decline 1-12 cannabis puffs at prices from $0.25 to $5. One decision was randomly implemented. Machine learning with leave-one-subject-out cross-validation identified discriminating neural patterns in 17 participants who purchased cannabis.

Why This Research Matters

Understanding the neural basis of decisions to use drugs could enable measurement of how treatments change decision-making processes in the brain, potentially improving addiction treatment development and evaluation.

The Bigger Picture

Addiction is increasingly understood as a disorder of decision-making. This study advances that framework by identifying specific brain signatures associated with the actual choice to use a drug, not just responses to drug cues. This could eventually help test whether behavioral or pharmacological treatments change these decision patterns.

What This Study Doesn't Tell Us

Very small sample (17 in fMRI analysis). Participants were non-treatment-seeking daily users, so results may not generalize to those trying to quit. The laboratory environment may not fully capture real-world decision contexts. 100% accuracy in a small sample may reflect overfitting.

Questions This Raises

  • ?Would this neural signature hold up in larger samples?
  • ?Can treatment interventions measurably change these brain decision patterns?
  • ?Do people who successfully quit cannabis show different neural signatures during cannabis-related decisions?

Trust & Context

Key Stat:
100% classifier accuracy distinguishing buy vs. decline decisions
Evidence Grade:
Small fMRI study with innovative methodology but limited sample size. Results need replication.
Study Age:
Published in 2015. This was early work establishing neural signatures of drug-related decision making.
Original Title:
An fMRI-Based Neural Signature of Decisions to Smoke Cannabis.
Published In:
Neuropsychopharmacology : official publication of the American College of Neuropsychopharmacology, 40(12), 2657-65 (2015)
Database ID:
RTHC-00913

Evidence Hierarchy

Meta-Analysis / Systematic Review
Randomized Controlled TrialGold standard for testing treatments
This study
Cohort / Case-Control
Cross-Sectional / Observational
Case Report / Animal Study

Participants are randomly assigned to treatment or placebo groups to test cause and effect.

What do these levels mean? →

Frequently Asked Questions

Can brain scans tell if someone will use cannabis?

In this small study, brain activity patterns during a purchasing task perfectly distinguished buy from decline decisions. But this was in a controlled lab setting with daily users and has not been validated for predicting real-world behavior.

What brain regions are involved in cannabis purchasing decisions?

The dorsal striatum, insula, cingulate cortex, posterior parietal regions, and dorsolateral prefrontal cortex all contributed to the neural signature distinguishing buy from decline decisions.

Read More on RethinkTHC

Cite This Study

RTHC-00913·https://rethinkthc.com/research/RTHC-00913

APA

Bedi, Gillinder; Lindquist, Martin A; Haney, Margaret. (2015). An fMRI-Based Neural Signature of Decisions to Smoke Cannabis.. Neuropsychopharmacology : official publication of the American College of Neuropsychopharmacology, 40(12), 2657-65. https://doi.org/10.1038/npp.2015.135

MLA

Bedi, Gillinder, et al. "An fMRI-Based Neural Signature of Decisions to Smoke Cannabis.." Neuropsychopharmacology : official publication of the American College of Neuropsychopharmacology, 2015. https://doi.org/10.1038/npp.2015.135

RethinkTHC

RethinkTHC Research Database. "An fMRI-Based Neural Signature of Decisions to Smoke Cannabi..." RTHC-00913. Retrieved from https://rethinkthc.com/research/bedi-2015-an-fmribased-neural-signature

Access the Original Study

Study data sourced from PubMed, a service of the U.S. National Library of Medicine, National Institutes of Health.

This study breakdown was produced by the RethinkTHC research team. We analyze and report published research findings without making health recommendations. All interpretations are based solely on the published abstract and study data.