Hypothetical Cannabis Purchase Tasks Accurately Predict Real Smoking Behavior
Hypothetical cannabis purchase decisions closely matched actual purchasing and smoking behavior in a lab setting, validating the widely used Marijuana Purchase Task as a research tool.
Quick Facts
What This Study Found
Hypothetical and actual cannabis purchase task performance showed high correlations (r=0.45-0.81), though participants were willing to pay more and showed less price sensitivity when purchases were real; self-reported anticipated consumption predicted 66% of actual smoking variance.
Key Numbers
92 participants (81 with non-zero trials); price-level correlations r=0.45-0.81; index correlations r=0.46-0.81; anticipated vs actual consumption R²=0.66 (p<0.001); actual MPT showed higher Omax/Pmax (d=0.47-0.51).
How They Did This
Laboratory cannabis self-administration study with 92 twice-weekly cannabis users completing both hypothetical and actual Marijuana Purchase Tasks, with one random trial actualized for smoking during a 1-hour session.
Why This Research Matters
The Marijuana Purchase Task is used extensively in cannabis research — confirming that hypothetical responses predict real behavior validates years of prior findings built on this tool.
The Bigger Picture
Validating research tools matters — if hypothetical measures didn't predict real behavior, a large body of cannabis demand research would be unreliable. This study confirms the field's methods are sound.
What This Study Doesn't Tell Us
Laboratory setting differs from real-world purchasing; participants were regular users (≥2x/week) so may not generalize to occasional users; single session doesn't capture longitudinal demand patterns.
Questions This Raises
- ?Why are people willing to pay slightly more when purchases are real?
- ?Could modified purchase tasks better account for the hypothetical-real gap?
Trust & Context
- Key Stat:
- Evidence Grade:
- Well-designed laboratory validation study with both hypothetical and actualized conditions, though limited to regular users in a controlled setting.
- Study Age:
- Published in 2026, providing critical methodological validation for a widely used cannabis research instrument.
- Original Title:
- Is toke cheap? Correspondence between cannabis demand and purchase in the laboratory.
- Published In:
- Addiction (Abingdon, England) (2026)
- Authors:
- Aston, Elizabeth R(9), Berey, Benjamin L(3), Amlung, Michael(5), Swift, Robert, MacKillop, James, Metrik, Jane
- Database ID:
- RTHC-08092
Evidence Hierarchy
Frequently Asked Questions
Can researchers trust hypothetical cannabis surveys?
Yes — this study showed that hypothetical cannabis purchase decisions were highly correlated (r=0.45-0.81) with actual purchasing and smoking behavior in the laboratory.
Do people accurately predict how much cannabis they'll use?
Self-reported anticipated consumption predicted 66% of actual cannabis smoked in the lab, suggesting people are fairly accurate about their own use patterns.
Read More on RethinkTHC
Cite This Study
https://rethinkthc.com/research/RTHC-08092APA
Aston, Elizabeth R; Berey, Benjamin L; Amlung, Michael; Swift, Robert; MacKillop, James; Metrik, Jane. (2026). Is toke cheap? Correspondence between cannabis demand and purchase in the laboratory.. Addiction (Abingdon, England). https://doi.org/10.1111/add.70282
MLA
Aston, Elizabeth R, et al. "Is toke cheap? Correspondence between cannabis demand and purchase in the laboratory.." Addiction (Abingdon, 2026. https://doi.org/10.1111/add.70282
RethinkTHC
RethinkTHC Research Database. "Is toke cheap? Correspondence between cannabis demand and pu..." RTHC-08092. Retrieved from https://rethinkthc.com/research/aston-2026-is-toke-cheap-correspondence
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.