A researcher reported that cannabis caused lasting damage to monkeys’ brains. The claim reached the public in 1974, became part of US politics, and was simplified into “cannabis kills brain cells.” Nobody ever replicated it, and two larger monkey studies contradicted it. During the same years, US-funded anthropologists studied heavy long-term users in Jamaica and found little obvious harm. Cannabis advocates have quoted that result ever since. The two stories mirror each other. Together they are the clearest case study we know of how an old result gets frozen into public belief and stays there after the science has moved on.
This is a case study in research methods and media literacy. It covers US-funded research from the 1970s and 1980s. It does not claim that cannabis is harmless, and nothing here is medical advice. Current evidence on risks is discussed in the Science and Research section (topic 273). Practical dosing questions belong in topic 1312.
1. The Study
Robert Galbraith Heath, a psychiatrist at Tulane University, reported in the 1970s that cannabis caused lasting changes in the brains of rhesus monkeys. In a December 1974 magazine interview he put it like this: “Alcohol is a simple drug with a temporary effect. Marijuana is complex with a persisting effect.” [2]
A 1978 newspaper report described monkeys trained to smoke the equivalent of one cannabis cigarette a day, five days a week, for six months [3]. The 1974 interview also described a study funded by the National Institutes of Health that used 13 monkeys in three groups: heavy exposure (about three joints a day), moderate exposure (one a day) and a control group that smoked inactive cannabis [2].
The dates matter. The claim became public in 1974. That year the brain-damage case against cannabis was presented at US Senate hearings chaired by Senator James Eastland [4], and the press and politicians took it up. Heath’s main peer-reviewed paper appeared later, in 1980 (Heath et al., Biological Psychiatry) [1]. It reported two studies in 21 monkeys. The authors described abnormal recordings from deep brain areas and changes at the synapses (the connections between nerve cells) seen under the electron microscope.
The claim that reached the public was blunter than anything in the papers: cannabis kills brain cells. The 1980 paper reported changes in structure, not cell death.
2. What Happened to It
Nobody ever replicated it.
Two larger monkey studies followed. At the US National Center for Toxicological Research (NCTR), a team led by Dr William Slikker used about 64 monkeys. Half were exposed to marijuana smoke for a year, and the first results were published in 1991 [5]. At SRI International, Charles Rebert and Gordon Pryor exposed more than 30 monkeys for 6–12 months and reported in 1993 [6]. California NORML, a group that campaigns for cannabis reform, summarised both studies in 1992. Neither found structural or neurochemical changes in the monkeys’ brains when they were examined a few months after exposure ended [7]. The BBC’s “On This Day” page on the 1974 claim says that Heath’s findings of permanent brain damage “have been dismissed by similar, independently conducted studies”. It adds that other scientists argue these animal methods are inconclusive in the first place [8].
That is the whole scientific story: a striking result, no replication, and contradiction by larger studies.
3. The Methodological Charge, Handled Honestly
Cannabis-advocacy writing makes a much stronger accusation. It says the monkeys were not smoking in any normal sense but were being asphyxiated: pumped with large amounts of smoke through a mask over a few minutes without enough oxygen. On this account, the damage came from carbon monoxide poisoning and lack of oxygen, not from cannabinoids.
The charge is usually traced to 1980. That year details of Heath’s exposure protocol were released and reported by Playboy magazine. Advocacy accounts say the release followed Freedom of Information requests. We treat the charge as contested, not established. It rests on a reading of the protocol, not on a published replication or a retraction. We have also not been able to check the original Playboy report or the released documents ourselves. The record we can verify says the study collapsed because it could not be replicated. It does not name a specific flaw that produced a false result.
This distinction matters, and it is a reason to be sceptical of the whole story. You do not need the asphyxiation charge. Failure to replicate, plus contradiction by two larger studies, is enough, and both are independently documented. Reaching for the more dramatic version and stating it as fact is the same error this article describes, committed by the other side.
4. Why It Persisted Anyway
Four mechanisms, all still at work.
Political amplification. The finding entered the political rhetoric of the time. Many later accounts say this included Ronald Reagan, then Governor of California. Anti-drug education repeated it for years. Once a claim enters a political programme, it has supporters regardless of whether the evidence holds up.
Unequal coverage. “Cannabis kills brain cells” is a headline. “A 1970s monkey study was not replicated” is not. A failed replication gets only a small share of the attention the original claim received. You can still measure this pattern today. A 2022 meta-analysis of randomised trials of cannabinoids for pain (Gedin et al., JAMA Network Open) found that media attention was high and mostly positive, whatever the trials found. Overall, pain relief with cannabinoids did not differ significantly from placebo [9].
Institutional inertia. Once educational material is printed and handed out, it keeps circulating. Teachers who learned it teach it.
It confirmed what people already believed. The claims that last longest are the ones an audience is ready to accept. That works in both directions, which brings us to the other side’s favourite old study.
5. The Mirror Image: Ganja in Jamaica and Cannabis in Costa Rica
The Jamaica study is Rubin and Comitas, Ganja in Jamaica: A Medical Anthropological Study of Chronic Marihuana Use (1975) [10]. The Research Institute for the Study of Man carried it out with funding from the US National Institute of Mental Health. The fieldwork took place in the early 1970s (roughly 1970–72), and 1975 is the year the book came out. The Costa Rica study was a separate project. A team led by anthropologist William E. Carter of the University of Florida published it as Cannabis in Costa Rica: A Study of Chronic Marihuana Use (1980) [12].
Both studies used a design that is still their main strength. Instead of bringing volunteers into a laboratory, the researchers studied long-term, heavy users for whom cannabis was part of local culture, in their own communities. They combined medical examinations with ethnographic fieldwork, meaning long-term observation of daily life. That gave them real-world exposure, which a single-dose laboratory study cannot provide. Their main finding was reassuring: they found no gross medical or psychiatric harm caused by long-term use. Advocacy writing often adds broader claims, such as no gateway effect, or cannabis replacing rum in Costa Rica. Those claims are less secure.
The qualifications that almost never come with them:
- Small samples, no randomisation. The Jamaican clinical comparison involved 30 long-term users and 30 matched controls [11]. The Costa Rican clinical work also compared a small number of matched pairs of users and non-users. Observational field studies can describe a population. They cannot show what would have happened to the same people without cannabis.
- Healthy-user and survivor effects. The people available to study were the ones still working, still living in their communities and still willing to take part. Anyone whose life had gone badly was less likely to be in the sample. The same filter makes personal testimonials unreliable.
- The product was not today’s product. Jamaican ganja around 1970 is not 2026’s high-potency cannabis. The EU-GEI case-control study compared people with and without psychotic disorder. Daily cannabis use was linked to roughly three times the odds of psychotic disorder compared with never using it. Daily use of high-potency cannabis (THC ≥10%) was linked to roughly five times the odds [15].
- Cognitive testing was cruder. The tests available around 1970 were less sensitive than those used today.
- They were criticised. G. G. Nahas published a formal critique of the Jamaica study in the UN Bulletin on Narcotics in 1985. It focused on the study’s methods and small number of subjects [13]. Citing the findings without mentioning that debate is not a neutral omission.
What has been learned since. The evidence has grown and become more nuanced rather than more favourable. A 1996 study of long-term users in Costa Rica found that older long-term users did worse than non-users on short-term memory and attention tests [14]. Earlier field methods were not designed to detect that kind of specific deficit. A 2020 meta-analysis found cannabis withdrawal syndrome in about 47% of regular or dependent users [16]. The potency–psychosis link above is another example. On the gateway question, a major 2017 review by the US National Academies found moderate evidence that cannabis use is statistically associated with later dependence on other substances. It did not show that cannabis causes this [17]. Researchers also still debate the link between cannabis and crime. How much of it comes from the drug, and how much from prohibition and shared social risk factors, remains unresolved.
So the honest summary is neither “the Jamaica study proved cannabis is safe” nor “the Jamaica study has been refuted.” It is that careful observational studies of specific populations in a specific era found no gross harms, and later research with better tools found more specific harms in more specific groups. Both statements are true, and they do not contradict each other.
What still stands. The design was right. Studying real exposure in real settings is a legitimate method that researchers use too little. The studies did not find the catastrophic harm that the political rhetoric of the time predicted. That is not the same as showing that heavy daily use is safe. The substitution question, often attributed to the Costa Rica work, asks whether people use cannabis in place of alcohol. It has some survey support but has never been tested in a randomised trial, so it remains a hypothesis.
The two errors are the same error. Citing 1975 findings as if nothing had been published since is the mirror image of keeping the dead-monkey claim alive. Applying findings from adult men in rural Jamaica around 1970 to adolescents in 2026 is a similar mistake. The Jamaican men had used low-potency cannabis all their lives under community rules. The adolescents use high-potency products in an unregulated market. The two groups differ on exactly the factors later research showed to matter. Neither side gets to pick a decade and stop.
6. What This Should Change In How You Read
Five questions. They apply to every dramatic finding, including the ones you want to be true.
| Ask | Why |
|---|---|
| Has it been replicated? | A single study, however striking, is a hypothesis. Replication is what turns it into knowledge |
| By whom? | Independent groups, not the original lab |
| What was the actual exposure? | Dose, route and duration decide whether the model resembles anything real |
| What is the sample size? | Small samples can raise a question, but they rarely settle one alone |
| What happened next? | The most useful thing to know about a 1970s claim is what the field did with it over the next fifty years |
The last question is the most useful and the least used. The epilepsy story (topic 1309/31) went the opposite way. A small 1980 trial of cannabidiol [18] was eventually followed through to a licensed medicine. Comparing the two cases is the clearest way to see the difference between a real result and a false one.
7. The Balanced Position
Heath’s specific claim does not stand. No one has shown that cannabis kills brain cells in the way he described.
That does not mean cannabis has no effects on the brain. Thinking and memory are clearly impaired while someone is intoxicated [17]. In the Costa Rica follow-up, older long-term users scored worse on memory and attention [14]. Frequent use of high-potency cannabis is linked to higher odds of psychotic disorder [15]. All of this rests on larger and better evidence than Heath ever had.
The Jamaica findings do not support the opposite conclusion either. They describe one population, one era and one product.
The right conclusion from both episodes is not that cannabis is safe, or that it is dangerous. It is that one study is not proof, whichever direction it points. Decades of argument over an old animal study or an old field study are no substitute for larger, better-designed human research, which can actually answer the question.
8. Why This Still Matters in Thailand and Southeast Asia
Both stories are American, but the way they frame the issue outlived them. Versions of the “cannabis kills brain cells” message still appear in drug-prevention and public-information material in many countries, possibly including parts of this region. One likely reason is that prevention messages favour simple, memorable warnings, and simple claims outlast the evidence behind them. We make no claim about any specific national curriculum. If you have seen this claim in local materials, say where.
Thailand also has its own, more relevant evidence. Cannabis was removed from the narcotics list in 2022. Thai researchers have since published data on the period of wide availability, including nationally representative surveys of adult cannabis use [19]. Under current law, cannabis flower is a controlled herb under the Notification of the Ministry of Public Health Re: Controlled Herbs (Cannabis) B.E. 2568, issued 26 June 2025. It may be dispensed only against a prescription on form PT 33 from an authorised practitioner. Each prescription covers no more than 30 days and cannot be refilled, and any flower sold must come from a GACP-certified cultivation site. The Cannabis Control Act is still a draft. Cabinet approved it on 22 September 2026 and sent it to the House.
Do you know of Thai or regional research on cannabis use and outcomes since 2022? Were you taught the dead-monkey claim, or another cannabis “fact” you later found to be wrong? Post it in the Science and Research section (topic 273). International studies cannot answer questions about this population, but local research can. Working out where a piece of received knowledge came from is a skill, and you can use it on every other health claim you meet.
References
- Heath RG, Fitzjarrell AT, Fontana CJ, Garey RE. Cannabis sativa: effects on brain function and ultrastructure in rhesus monkeys. Biological Psychiatry. 1980;15(5).
- Chandler D. “Pot Is Safe, Right? Wrong, Says a Doctor: It Can Cause Brain Damage.” People, 9 December 1974. https://people.com/archive/pot-is-safe-right-wrong-says-a-doctor-it-can-cause-brain-damage-vol-2-no-24/
- Rensberger B. “Marijuana Tied to Brain Change in Monkey Tests.” The New York Times, 13 January 1978. https://www.nytimes.com/1978/01/13/archives/marijuana-tied-to-brain-change-in-monkey-tests.html
- US Senate, Committee on the Judiciary, Subcommittee to Investigate the Administration of the Internal Security Act. Marihuana-Hashish Epidemic and Its Impact on United States Security. Hearings, 93rd Congress, 2nd session, May–June 1974. https://catalog.hathitrust.org/Record/001278569
- Slikker W Jr, et al. Chronic marijuana smoke exposure in the rhesus monkey. Fundamental and Applied Toxicology. 1991;17:321–332.
- Rebert CS, Pryor GT. Chronic inhalation of marijuana smoke and brain electrophysiology of rhesus monkeys. International Journal of Psychophysiology. 1993;14:144 (abstract).
- California NORML Reports, April 1992 (reprinted by Erowid). https://www.erowid.org/plants/cannabis/cannabis_myth16.shtml
- BBC News, On This Day, 2 October: “1974: Cannabis ‘causes brain damage’.” BBC ON THIS DAY | 2 | 1974: Cannabis 'causes brain damage'
- Gedin F, Blomé S, Pontén M, et al. Placebo response and media attention in randomized clinical trials assessing cannabis-based therapies for pain: a systematic review and meta-analysis. JAMA Network Open. 2022;5(11):e2243848. pubmed.ncbi.nlm.nih.gov
- Rubin V, Comitas L. Ganja in Jamaica: A Medical Anthropological Study of Chronic Marihuana Use. The Hague: Mouton; 1975.
- Beaubrun MH, Knight F. Psychiatric assessment of 30 chronic users of cannabis and 30 matched controls. American Journal of Psychiatry. 1973;130(3):309. https://doi.org/10.1176/ajp.130.3.309
- Carter WE, ed. Cannabis in Costa Rica: A Study of Chronic Marihuana Use. Philadelphia: Institute for the Study of Human Issues; 1980.
- Nahas GG. Critique of a study on ganja in Jamaica. Bulletin on Narcotics. 1985;37(4). UNODC - Bulletin on Narcotics - 1985 Issue 4 - 003
- Fletcher JM, Page JB, Francis DJ, et al. Cognitive correlates of long-term cannabis use in Costa Rican men. Archives of General Psychiatry. 1996;53(11):1051–1057. https://scholarworks.gsu.edu/psych_facpub/169
- Di Forti M, et al. The contribution of cannabis use to variation in the incidence of psychotic disorder across Europe (EU-GEI). Lancet Psychiatry. 2019;6(5):427–436.
- Bahji A, et al. Prevalence of cannabis withdrawal symptoms among people with regular or dependent use of cannabinoids: a systematic review and meta-analysis. JAMA Network Open. 2020;3(4):e202370.
- National Academies of Sciences, Engineering, and Medicine. The Health Effects of Cannabis and Cannabinoids. Washington, DC: National Academies Press; 2017.
- Cunha JM, et al. Chronic administration of cannabidiol to healthy volunteers and epileptic patients. Pharmacology. 1980;21(3):175–185.
- Wichaidit W, et al. Density of cannabis outlets vs. cannabis use behaviors and prevalent cannabis use disorder: findings from a nationally-representative survey. PeerJ. 2024;12:e17317. Density of cannabis outlets vs. cannabis use behaviors and prevalent cannabis use disorder: findings from a nationally-representative survey - PMC
General educational information on evaluating scientific claims. Not medical advice.