A pair of new studies examining rare homicides committed under the influence of psychedelic drugs indicates that sudden breaks with reality and false beliefs about imminent danger frequently drive these fatal attacks. One study, published in Journal of Police and Criminal Psychology, analyzed statistical and thematic patterns across more than 600 cases.
A companion study in Psychiatry, Psychology and Law profiled 22 individual offenders in detail, indicating along with the broader analysis that delusion-driven killings often target loved ones without prior hostility and result in starkly different prison sentences across countries.
Psychedelics, psychosis, and criminal responsibility
Classic psychedelics are mind-altering drugs, including lysergic acid diethylamide (LSD), psilocybin from “magic mushrooms,” mescaline, and dimethyltryptamine (DMT, the main active ingredient in the plant brew ayahuasca), that temporarily alter perception, mood, and thought. Clinical research into their mental health benefits and recreational use have both grown rapidly, and severe emergencies remain rare: a large international survey of magic mushroom users found that only 0.2 percent of past-year users sought emergency medical treatment, mostly for short-lived panic or anxiety that resolved within 24 hours.
Population-level data have even linked certain psychedelics to lower crime rates, as a national survey analysis of U.S. adults found that lifetime psilocybin use was associated with lower statistical odds of ever having been arrested, while LSD showed no statistical association with arrests. Even so, clinical researchers have raised concerns that enthusiasm over psychedelic therapy has outpaced research into rare severe harms, with almost no medical studies examining psychedelic-related lethal violence since the 1970s.
Don’t miss the next big psychology finding
Every morning, PsyPost sends a short briefing on the most interesting new peer-reviewed research in psychology and neuroscience.
The primary psychiatric state linking certain drugs to sudden violence is psychosis, a condition in which a person loses contact with external reality through hallucinations (seeing or hearing things that are not there) or delusions (firmly held false beliefs that persist despite obvious contrary evidence). That pathway is well established for stimulants: a 2014 longitudinal study of methamphetamine users in Australia found that violent behavior rose substantially during periods of use, largely in connection with paranoid psychotic symptoms.
Psychedelics trigger psychosis far less often than methamphetamine, yet acute intoxication can occasionally bring on temporary delusions or longer-lasting breaks with reality, termed “substance-associated psychosis” when clinicians cannot tell whether a drug caused a brief reaction or unmasked an illness like schizophrenia. As the authors of a 2008 legal and psychiatric analysis argued, such cases strain criminal law because courts traditionally allow insanity defenses for internal mental illnesses while barring them for voluntary intoxication.
Seeking to document these rare tragedies without taking a stance on legalization, Jesse J. Norris, an associate professor of criminal justice in the Social Work and Criminal Justice Department at the State University of New York at Fredonia, led two complementary studies using a shared database.
“Psychedelic drugs such as LSD, psilocybin and ayahuasca are rapidly increasing in popularity, but there is limited research about their adverse effects,” Norris told PsyPost. “A small percentage of psychedelic users experience psychotic episodes, but little is known about the consequences of psychedelic-induced psychotic symptoms for people’s behavior.”
Patterns of delusions across 626 homicides
For the broader descriptive, exploratory study, Norris searched court archives, academic journals, global news databases, and web engines across 31 languages for unlawful killings committed between 1962 and July 2025.
Cases were included only when court rulings, police records, academic reports, or local news articles explicitly stated the offender was on a psychedelic during the killing. Non-classic drugs like MDMA (ecstasy) and ketamine were excluded, and the dataset was made available upon request.
The search yielded 626 homicide events involving 660 perpetrators across 33 countries, as some killings had multiple offenders, with roughly 86 percent of cases occurring in the United States or Canada. Norris tested seven statistical hypotheses comparing subgroups of offenders and thematically coded the 129 cases in which sources detailed the content and motives of the offenders’ delusions.
About 88 percent of the homicides involved LSD, 11 percent involved psilocybin, and 58 percent involved concurrent use of non-psychedelic drugs, usually alcohol or cannabis. Delusions appeared in 28.3 percent of cases (177 homicides), rising from an average of two reported cases per year between 1966 and 2010 to 5.9 per year between 2011 and 2025, which Norris suggests may reflect wider use, larger doses, or fuller reporting.
Among the 129 cases with specified delusions, religious themes were unusually common, appearing in roughly 45 percent of delusional offenders across several combined categories. Specifically, about 23 percent involved perceived threats from demons or the devil, roughly 9 percent involved beliefs about communicating with God, and 6 percent involved offenders claiming to be God or an angel.
“I was surprised at how many psychedelic-intoxicated homicide offenders were experiencing vivid supernatural delusions that motivated their offenses,” Norris said. “Compared to other forms of psychosis, supernatural delusions might be more common in psychedelic-associated psychosis.”
Outside of religious beliefs, another 20 percent of specified cases centered on secular persecutory delusions that other people were plotting against or attacking the user. Regarding motive, countering an imagined human, animal, or supernatural threat drove 52 percent of specified cases, while smaller shares involved obeying hallucinated commands to kill (9 percent) or attempting to “wake up” into normal reality.
Offenders with delusions were more than twice as likely as non-delusional offenders to kill an intimate partner and nearly twice as likely to kill a friend or a parent, with parents accounting for 8.4 percent of all victims versus roughly 2 percent of homicides overall. Psilocybin accounted for a larger share of homicides in recent years, and psilocybin-intoxicated offenders were roughly twice as likely to exhibit delusions as users of other psychedelics, a pattern Norris proposes may reflect a cultural trend toward very large mushroom doses. Perpetrators who committed robbery or sexual assault alongside the killing, by contrast, were much less likely to report delusions.
Contrary to Norris’s prediction, offenders who mixed psychedelics with alcohol, cannabis, or other drugs were roughly one-third less likely to exhibit delusions than those taking psychedelics alone, perhaps because solo users take higher doses while mixed-drug killings more often involve alcohol-fueled impulsivity. About 14 percent of cases involved “overkill” (massive injuries far beyond what is fatal) and 15 percent of offenders reported amnesia. However, the link between delusions and overkill was not statistically significant, and the amnesia difference between mixed-drug and psychedelic-only offenders was not statistically significant, suggesting psychedelics themselves can cause blackouts.
Inside 22 cases and their courtroom outcomes
Because statistical coding cannot establish whether a psychedelic drove an individual killing or show how courts weigh blame, Norris partnered with Dr. Brian Holoyda, chief psychiatrist at Contra Costa County Detention Health Services in Martinez, California, and an adjunct assistant professor at the Medical College of Wisconsin, to profile specific cases from the same database.
The authors first narrowed the 626 cases to 71 well-documented homicides in which psychedelics likely played a direct role. From that group, they selected 22 diverse offenders across the United States, Canada, Australia, and the United Kingdom to examine clinical features and sentencing, providing the non-academic sources for those 22 profiles in a supplementary file.
Selected for variety rather than statistical representativeness, the 22 profiles echoed the full database, where only about 5 percent of offenders had a delusion-related psychiatric diagnosis (almost always assigned after arrest): just one of the 22 offenders had a prior schizophrenia diagnosis. Several offenders had used LSD or psilocybin peacefully many times before an abrupt psychotic crisis, including a Canadian teen experienced with large psilocybin doses who took mushrooms and hallucinated his father turning into a werewolf. At the same time, at least four and possibly six of the 22 offenders had experienced prior episodes of psychedelic-associated psychosis yet took psychedelics again.
“Although other explanations for these crimes are possible, such as effects from other drugs or pre-existing psychosis, the circumstances of many of these cases strongly suggest that psychedelic use was what triggered the offenders’ behavior,” Norris explained.
The scarcity of prior psychosis diagnoses and the lower delusion rate among drug-mixing offenders stand in tension with a 2017 review of clinical and published reports on ayahuasca and DMT, which found psychotic episodes outside clinical settings were more common among people with personal or family histories of psychosis or bipolar disorder and those mixing DMT with other drugs. That contrast likely reflects measurement differences, as the new studies examined legal and news records across many psychedelics (which often omit confidential medical or family histories) rather than clinical reports of prolonged psychosis from ayahuasca and DMT.
The profiles also illustrated “delusions of unreality,” the fixed belief that reality is a dream, movie, or video game, documented in three offenders. Unlike derealization, a common dissociative feeling on psychedelics in which surroundings seem artificial while the user retains doubt, these offenders lost reality testing completely, including a British teenager on a synthetic psychedelic who fatally stabbed his friend, whom he perceived as a skull, believing that doing so would wake him.
“Some of these offenders seemed genuinely surprised when they learned that their murders were real,” Norris noted.
Six profiled cases involving alcohol or other non-psychedelic drugs showed how sleep deprivation or stimulants complicate causal judgments, while accessible knives, guns, or car keys heightened the danger across nearly all 22 cases. Several of the killings also unfolded at remote campsites or hiking trails where outside intervention was unlikely.
“I was also surprised at how several of the murders occurred in isolated locations such as camping sites and hiking trails,” he said. “I think this is because groups of friends often use psychedelics while on nature outings.”
Sentences across the 22 cases ranged from two years of juvenile probation to life imprisonment and one California death sentence, averaging 26.4 years in the United States versus 13 years in Australia and 6.9 years in Canada. These comparisons come from a small, non-representative sample and are complicated by procedural differences across jurisdictions, including juvenile court sentencing, plea-bargaining discounts, mandatory minimums, and parole eligibility rules. Even so, the gaps align with differing substantive legal standards: most U.S. states bar voluntary intoxication as an insanity defense, whereas Canada permits acquittals for unforeseeable drug-induced automatism (involuntary action without conscious control) and parts of Australia allow leniency for unforeseeable psychosis. Offenders across the sample with prior episodes of psychedelic-associated psychosis averaged longer sentences (25.4 years versus 19 years).
Caveats, harm reduction, and next steps
There are a few things to keep in mind when considering these findings. Because both studies examined only homicides involving psychedelic intoxication, without a control group or data on total users, they cannot estimate the overall risk of psychedelic-related violence or compare it with alcohol or other drugs. Six hundred twenty-six cases across six decades and 33 countries represent a tiny fraction of total psychedelic trips, and psychedelics may still reduce offending among many users while triggering violent delusions in rare cases. Relying on court and news records also left gaps in medical charts, family psychiatric histories, dosages, and drug lab tests, with roughly nine in ten cases coming from English-speaking countries.
“I would like to clarify that we don’t have any particular agenda that we’re trying to promote through this research,” Norris added. “We just think these cases are important to document so that people have a more complete understanding of the effects of psychedelics on people’s behavior. Even if some homicides can be attributed to psychedelic use, that does not necessarily mean that the authorities should maintain the illegal status of these drugs or more strictly enforce laws against them. Policymakers and the general public should be aware of these homicides and keep them in mind in their decision-making, but other issues—such as the possibility that psychedelics prevent some crime through ameliorating certain psychological conditions—are also worth considering.”
Beyond broader drug policy, the authors suggest that acknowledging these rare tragedies can inform practical harm reduction, such as reconsidering isolated wilderness trips where bystanders cannot safely intervene and maintaining psychiatric follow-up if delusions persist after a drug wears off. Norris emphasized that users and those around them can take straightforward steps to lower the chance of a crisis turning violent.
“While these murders are not particularly common, they are a real risk that users should seek to reduce through increased caution about dosages, the use of other drugs at the same time, and the setting of psychedelic use,” Norris said. “Users should not have access to weapons or vehicles while on these drugs. If someone develops psychotic symptoms, immediate psychiatric attention may be important for preventing violence.”
“Moreover, people with any history of delusions, whether drug-related or not, should avoid psychedelic use, since they are likely to experience delusions again and put others at risk,” he added. “We found several examples of people who had previously had drug-induced psychosis and yet used psychedelics again, leading to tragic results.”
Future studies incorporating full medical records, emergency-room data, and large user surveys could help determine who is most vulnerable to violent delusions and how bystanders, clinicians, and courts can respond more effectively. Norris, whose published research also examines substance use and criminal offending more broadly, said his work on psychedelic-related crime is expanding beyond homicides.
“We are in the midst of additional projects documenting additional negative behavioral outcomes of psychedelic use, such as different types of crimes committed by people under the influence of psychedelics,” he said.
The study, “The Role of Delusions in Homicides by Psychedelic-Intoxicated Offenders: A Mixed-Method Analysis,” was authored by Jesse J. Norris. The study, “Homicides associated with psychedelic use: Psychiatric features, risk factors and sentencing outcomes,” was authored by Jesse J. Norris and Brian Holoyda.