Cannabidiol shows preclinical promise for anxiety, depression and cognition

By Vijay Kumar MalesuReviewed by Susha Cheriyedath, M.Sc.Jul 1 2026

A new Molecular Psychiatry meta-analysis suggests CBD may improve emotional behavior and cognition in animal models, but researchers caution that clinical benefits still need to be proven in humans.

Review: Effects of cannabidiol on anxiety- and depressive-like behaviors and cognition: a systematic review and meta-analysis of preclinical studies. Image Credit: Creativan / Shutterstock

In a recent study published in the journal Molecular Psychiatry, a group of researchers systematically evaluated the effects of cannabidiol (CBD) on anxiety-like behaviors, depressive-like behaviors, and cognitive function in preclinical animal studies using a systematic review and meta-analysis.

Background

Can the non-intoxicating compound from Cannabis sativa help with emotional behavior and cognition without producing the intoxicating effects commonly associated with the cannabis plant? CBD is already approved for treating specific seizure disorders, while experimental research has suggested benefits in neurological diseases, mood disorders, inflammation, and cognitive impairment. CBD acts through multiple biological pathways, including the endocannabinoid and serotonin systems, which influence emotion and memory. However, findings from preclinical studies have been inconsistent, underscoring the need for a quantitative synthesis to evaluate CBD's efficacy.

About the study

The researchers conducted a systematic review and meta-analysis after registering the study protocol in the International Prospective Register of Systematic Reviews (PROSPERO) and following the Systematic Review Center for Laboratory Animal Experimentation (SYRCLE) guidelines and the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) recommendations.

Researchers searched PubMed, Scopus, and Web of Science for English-language studies published from January 1995 to November 2024. Studies that used an isolated form of CBD with rodents were considered in this systematic review if they measured validated behavioral tests to quantify anxiety behavior, depression-like behavior, or cognitive function. Studies that included human participants, cell culture studies, CBD-rich extract studies, studies using combination therapy, and/or had no direct relationship to CBD and behavior were excluded from this review.

Reviewers screened the articles and extracted specific details, including animal species, sex, age, treatment duration, disease model, behavioral outcomes, and statistical measures. Study quality and risk of bias were independently assessed using the Collaborative Approach to Meta-Analysis and Review of Animal Data from Experimental Studies (CAMARADES) criteria and the SYRCLE Risk of Bias tool. The collected data were analyzed using various statistical methods, including random-effects meta-analysis, standardized mean difference (SMD), heterogeneity tests, publication bias testing, subgroup analysis, meta-regression, and sensitivity analysis to assess the strength and consistency of the evidence.

Study results

In the systematic review, 2,333 records were identified, of which 123 preclinical studies were deemed eligible after duplicate removal and detailed screening. Most of these 123 studies were rated moderate or high quality by CAMARADES and SYRCLE assessment tools and involved research with rats, mice, and hamsters. Overall, 67 studies were rated as high quality and 56 as moderate quality, although sample-size calculations, randomization, and blinding were often underreported.

Analysis of anxiety-like behaviors included 200 outcomes from 75 studies. CBD produced a moderate overall reduction in anxiety-like behaviors with an SMD of −0.50 after outlier removal. Moderate benefits were observed in the elevated plus maze (SMD −0.55) and novelty-suppressed feeding test (SMD −0.68), while smaller but significant improvements were found in the open field test (SMD −0.38) and light-dark box test (SMD −0.47). CBD also reduced anxiety-like behaviors in animals exposed to stress models. With respect to different pathologies, the greatest effects were noted in maternal alcohol consumption (SMD −1.33), type 1 diabetes (SMD −1.12), stress (SMD −0.66), and pain (SMD −0.59). While adjustment for publication bias did attenuate the estimated effect size, CBD retained a statistically significant anxiolytic effect.

Depressive-like behaviors were evaluated using 160 outcomes from 46 studies. After removing outliers, CBD maintained a moderate antidepressant effect (SMD −0.71) with low heterogeneity. Consistent benefits were observed in the forced swim test (SMD −0.74), tail suspension test (SMD −0.73), and sucrose preference test (SMD −0.56). CBD improved depressive-like behaviors in animals without an induced disease model and in several disease models, including cerebral ischemia (SMD −1.07), Parkinson's disease (SMD −0.91), pain (SMD −0.78), type 1 diabetes (SMD −0.72), depression (SMD −0.67), lipopolysaccharide-induced inflammation (SMD −0.61), and stress (SMD −0.55). The antidepressant effect remained significant after adjustment for publication bias.

Memory and cognitive function were evaluated using 102 results from 50 studies. After removing outliers, CBD was associated with a significant overall improvement in cognitive performance (SMD −0.99). Preclinical behavioral paradigms used to evaluate cognitive functioning included Y-maze (SMD −1.22), inhibitory avoidance (SMD −1.14), novel object recognition (SMD −1.07), object location (SMD −0.96), and social recognition (SMD −0.55). However, the results from both the Barnes maze and Morris water maze, measures of hippocampal-dependent spatial memory, were variable and did not reach statistical significance. Cognitive functioning was also improved in several models of disease; specifically, lipopolysaccharide-induced inflammatory models exhibited the greatest effect size (SMD −2.06), schizophrenia (SMD −1.40), depression (SMD −1.08), Alzheimer’s disease (SMD −0.86), cerebral ischemia (SMD −0.76), and maternal alcohol consumption (SMD −0.77). However, cognition-related findings were more heterogeneous than emotional-behavior outcomes, and publication bias adjustment reduced the pooled estimate, which remained statistically significant.

The authors also analyzed fear-related memory separately because reduced freezing in these paradigms can reflect altered aversive memory rather than a straightforward improvement in cognition. After outlier removal, CBD showed a small, significant effect, mainly in contextual fear-memory tasks, while cued fear-memory findings were not significant.

Conclusions

The researchers concluded that CBD generally reduced anxiety-like and depressive-like behaviors while improving cognitive performance across diverse preclinical animal models. Evidence was most consistent for emotional behaviors, whereas cognitive effects were larger in some analyses but varied across specific behavioral tasks, disease models, doses, and assessment timing. Multiple biological pathways, including serotonergic, endocannabinoid, anti-inflammatory, neurotrophic, mitochondrial, and synaptic pathways, may contribute to these outcomes. While the experimental data provide substantial preclinical support for the potential therapeutic application of CBD, human data supporting the therapeutic use of CBD remain limited and inconclusive. The researchers emphasized the need for additional well-designed human studies to confirm these benefits.

Want to read later? Download your PDF copy by clicking here.