Theory of Mind, often abbreviated as ToM, is the cognitive machinery that allows humans to attribute mental states to others and to predict behavior on that basis. First described in the landmark 1978 question of whether chimpanzees possess a theory of mind, the capacity depends on a distributed network of cortical regions, particularly the prefrontal, insular and cingulate areas. Those same regions are among the most consistently altered in people with substance use disorder, whether through reduced cortical volume, disrupted connectivity or changes in neurochemistry. This anatomical overlap is one reason researchers have long suspected that psychoactive substances could compromise the ability to understand other people.

Previous syntheses had already documented ToM problems among people with alcohol use disorder, with effect sizes ranging from moderate to very large. Meta-analyses of facial emotion recognition, a related social-cognitive skill, also pointed to deficits among alcohol and substance users. But evidence on substances other than alcohol remained scattered. The only earlier systematic review covering the broader landscape, published in 2016, found relatively weak evidence for ToM impairments linked to methamphetamine and no impairments among recreational cocaine users. The new study, led by José Vitor de Lima Neto and Flavia de Lima Osório of the University of São Paulo, set out to resolve that uncertainty with a formally registered, PRISMA-guided search of eight databases including Embase, Scopus, PsycINFO, Web of Science, PubMed, PsycNet, LILACS and SciELO.

The review included original quantitative studies comparing Theory of Mind performance between adult psychoactive substance users and controls. After screening, 37 studies met the criteria, and 30 of them—covering 35 distinct samples—provided data suitable for meta-analysis. Most of the research originated in Europe, with 70 percent of studies conducted there, and the field’s first study dates only to 2006, making this a young but rapidly growing literature. Alcohol was the predominant substance studied, accounting for roughly half of the samples, followed by stimulants, opioids, cannabis and a single study involving the hallucinogen MDMA. Participants were mostly men, with a mean age of about 37 years, and were largely recruited from inpatient detoxification clinics.

The pooled result was striking. Across the meta-analyzed samples, substance users showed significantly poorer Theory of Mind performance than controls, with a standardized mean difference of −0.80, a large effect by conventional statistical standards. The 95 percent confidence interval ran from −1.13 to −0.47, well clear of zero. Heterogeneity was high, at roughly 93 percent, reflecting the diversity of instruments, populations and study designs, but sensitivity analyses removing statistical outliers left the overall estimate essentially unchanged. Publication bias tests, including Begg’s rank correlation test and funnel plot inspection, found no evidence that negative findings were being suppressed.

Subgroup analyses revealed something perhaps even more consequential: the impairment did not appear to depend on which substance was used. Samples of alcohol users produced an effect size of −0.92, opioid users −1.12 and stimulant users −1.29. A formal between-group test found no statistically significant differences among these classes, with overlapping confidence intervals pointing to a common magnitude of deficit. The authors also tested whether the choice of assessment tool mattered, given ongoing debate about the Reading the Mind in the Eyes Test, the most frequently used instrument in this literature. A mixed-effects meta-regression comparing studies that used the Eyes Test with those using other measures found no significant moderating effect, suggesting the deficits are not an artifact of one particular task.

Two theoretical explanations dominate the interpretation of these findings, and they are not mutually exclusive. The first is neurotoxicity: chronic exposure to alcohol, opioids or stimulants damages or dysregulates precisely the prefrontal, insular and cingulate circuits that Theory of Mind depends upon, and decades of neuroimaging work document structural and functional changes in these regions among people with substance use disorder. The second is vulnerability: weak social cognition may predate substance use, increasing susceptibility to social stress, isolation and exclusion, and leading individuals to use substances as a way of coping with relational difficulties or of conforming to perceived social norms. Supporting the vulnerability hypothesis, a 2022 meta-analysis found an inverse relationship between alcohol-related problems and ToM performance even in non-clinical samples, where use never reached diagnostic thresholds.

The new review itself is cross-sectional in its evidence base, with 36 of the 37 included studies using cross-sectional designs, so causality cannot be established. The authors are explicit that their pooled estimate should be read as evidence of association rather than cause. They also note that the two mechanisms likely interact in a self-reinforcing loop: a primary ToM weakness predisposes someone to use substances, chronic use then intensifies the deficit through neurotoxic effects, and the deepened impairment in turn hinders therapeutic adherence and relapse control. Interestingly, one study in the review found that long-term MDMA users actually performed better on Theory of Mind tasks than non-users, a finding consistent with evidence that the pursuit of empathetic social experience motivates MDMA use—though it rests on a single study.

Important gaps remain. Evidence on cannabis, hallucinogens and nicotine is too sparse for meta-analysis; two cannabis studies found no significant differences, though the tasks used assessed affective rather than cognitive ToM, a subdomain that may be selectively vulnerable given cannabis-linked frontal reductions. Methodological quality across the field was only moderate, most samples came from high-income countries, inpatient recruitment may have inflated severity, and polysubstance use and psychiatric comorbidities were rarely controlled. Nevertheless, the authors argue the practical message is clear: sociocognitive skills deserve formal assessment in both prevention programs and addiction treatment, because deficits in reading minds can raise vulnerability to initiating use, undermine therapy and complicate recovery long after detoxification ends.

Subject of Research: Theory of Mind deficits among psychoactive substance users

Article Title: Theory of Mind in Psychoactive Substance Users: A Systematic Literature Review and Meta-analysis

Article References: Neto, J. V. D. L., & Osório, F. D. L. (2026). Theory of Mind in Psychoactive Substance Users: A Systematic Literature Review and Meta-analysis. International Journal of Mental Health and Addiction. https://doi.org/10.1007/s11469-026-01717-0

Image Credits: AI Generated

DOI: 10.1007/s11469-026-01717-0

Keywords: Theory of Mind, psychoactive substances, substance use disorder, social cognition, meta-analysis, alcohol, opioids, stimulants, addiction, prefrontal cortex, mental state attribution, systematic review