Related Experiment Video
Updated: Jan 9, 2026
06:19
Integration of Animal Behavioral Assessment and Convolutional Neural Network to Study Wasabi-Alcohol Taste-Smell Interaction
Published on: August 16, 2024
802
Artificial Intelligence Approximates Human Affect Ratings of Cannabis Images
Jacob T Borodovsky1,2, Richard J Macatee3, Sarah M Preum1,4
1Center for Technology and Behavioral Health, Dartmouth Geisel School of Medicine, Lebanon, NH, USA.
Medrxiv : the Preprint Server for Health Sciences
|December 3, 2025
Summary
Multimodal generative artificial intelligence (MGAI) can approximate human emotional responses to cannabis images, showing potential for public health research and regulation of online marketing.
Area of Science:
- Computational Social Science
- Cannabis Regulatory Science
- Artificial Intelligence in Health
Background:
- Online cannabis imagery is increasing, potentially impacting public health.
- Tools are needed to assess the influence of this imagery on population health.
- Human affective responses to cannabis cues are well-documented.
Purpose of the Study:
- To test if multimodal generative artificial intelligence (MGAI) can replicate human affect ratings of cannabis images.
- To evaluate MGAI's potential as a scalable tool for analyzing cannabis marketing impacts.
Main Methods:
- Four MGAI agents were created, each simulating a human subgroup based on cannabis administration method (bong, bowl, joint/blunt, vaporizer).
- Agents used zero-shot prompting to rate a standardized image set on valence, arousal, and urge.
- MGAI ratings were compared to human ratings from a previous study using equivalence tests and Spearman correlations.
Main Results:
- MGAI ratings closely approximated human mean ratings (Mean difference = -0.31, SD = 1.23).
- Moderate to high correlations were observed between MGAI and human ratings for valence (rs=0.55), arousal (rs=0.34), and urge (rs=0.56).
- MGAI reproduced the observed relationship between rating means and standard deviations in human data.
Conclusions:
- MGAI shows promise in approximating human cannabis cue-reactivity patterns.
- Further refinement of MGAI could lead to a valuable tool for Cannabis Regulatory Science.
- MGAI may aid in the regulatory oversight of online cannabis marketing and its public health implications.
Related Concept Videos
01:27Hallucinogens and Psychedelics
570
Hallucinogens are psychoactive substances that profoundly alter perceptual experiences, generating unreal visual and sensory images. Often referred to as psychedelic drugs — a term derived from the Greek words "psyche" (mind) and "delos" (revealing) — these substances include marijuana and lysergic acid diethylamide (LSD), among others. These drugs vary in intensity and effects.
Marijuana, derived from the dried leaves and flowers of the hemp plant, contains...
Marijuana, derived from the dried leaves and flowers of the hemp plant, contains...
570
02:16Stereotype Content Model
15.3K
The Stereotype Content Model (SCM) was first proposed by Susan Fiske and her colleagues (Fiske, Cuddy, Glick & Xu, 2002; see also Fiske, 2012 and Fiske, 2017). The SCM specifies that when someone encounters a new group, they will stereotype them based on two metrics: warmth—or that group’s perceived intent, and how likely they are to provide help or inflict harm—and competence—or their ability to carry out that objective. Depending on the warmth-competence...
15.3K