Answers are grounded in this paper and cite their sources.
Rationale Cannabidiol (CBD) modulates anxiety and fear in rodents; however, the dose-dependent characteristics of these effects remain unclear. Objectives & Methods Regression-based analyses assessed how consistent dose-effect inferences are across diverse studies and determined whether linear or nonlinear models best describe the relationship between CBD dose and behavioral outcomes. Hedges’ g effect sizes from 312 experiments in mice and rats were analyzed separately and in pooled datasets, using both raw and allometrically normalized doses. We also evaluated how biological and procedural variables moderated dose-effect relationships. Results Intermediate doses of systemic CBD administration were more often associated with larger antiaversive effect sizes; however, this trend did not hold across species or behavioral domains. Some analyses showed nonlinear trends, primarily associated with high-dose observations. After allometric scaling and pooling of anxiety and conditioned fear datasets, linear models fit better than nonlinear models. Analyses of pooled and anxiety datasets revealed no significant moderators of the dose-effect relationship. In contrast, analyses limited to conditioned fear identified species as a significant moderator, while sex, stress history, treatment regimen, age, and treatment-testing interval were not significant. Conclusions These findings indicate that the preclinical literature does not support a stable or generalizable CBD dose-effect function. The observed relationships likely reflect substantial methodological heterogeneity and may also reflect CBD’s pharmacological complexity. Factorial studies with balanced dose distributions and controlled experimental conditions are necessary to improve reproducibility and mechanistic understanding of CBD’s antiversive effects.
No takes yet. Share an insight, caveat, or question.
Lewandowski et al. (2026) studied this question.
Synapse has enriched 10 related papers. Consider them for comparative context: