Does weed kill brain cells? There is no good evidence that cannabis kills brain cells in people at the amounts adults use. The idea took hold after a small 1980 study in monkeys, and a larger US government monkey programme a decade later did not consistently find brain damage. That does not make cannabis harmless for the brain. A New Zealand study that followed 1,037 people from birth linked persistent heavy use, especially use that began in the teens, to a fall in IQ, though twin studies suggest family background explains much of that. Most thinking problems in young users fade after a few days without cannabis.
Last updated:
Key takeaways
- The “kills brain cells” claim traces back largely to a 1980 study of 21 rhesus monkeys that reported changes at the synapse after months of heavy exposure.
- A larger programme at the US National Center for Toxicological Research exposed 64 monkeys to cannabis smoke daily for a year and, in its own review, did not consistently find neurotoxicity.
- In the Dunedin birth cohort, people dependent on cannabis at three or more assessments lost about 6 IQ points between age 13 and 38. Those who started in adolescence lost about 8.
- Two US twin studies found that cannabis-using twins did not show a bigger IQ drop than their non-using siblings, pointing to family factors rather than cannabis alone.
- A 2018 meta-analysis of 69 studies found only a small thinking deficit in young frequent users, and no significant deficit in studies that required more than 72 hours without cannabis.
Does weed kill brain cells, or is that a myth?
As a literal claim, it is not supported. No human study has shown that using cannabis destroys brain cells, and the human research on this question does not count neurons at all. It measures memory, attention and IQ over time. What the evidence does support is a more careful statement: heavy, long-term use, particularly from a young age, is linked to poorer thinking, and some of that link may not be caused by cannabis itself.
The question matters because the answer changes how people should think about risk. If weed simply killed brain cells, every use would chip away at the brain. The research points somewhere else. Occasional adult use is not linked to lasting damage in the better studies, while the people whose results look worst are those who used heavily for many years, often starting as teenagers. That is where the real caution belongs.
Where did the brain cell myth come from?
Mostly from animal research in the 1970s and 1980s, and one monkey study in particular. Robert Heath and colleagues at Tulane University published it in Biological Psychiatry in 1980. It was small, used high exposures and reported changes in brain structure, and its conclusion that THC could “produce permanent alterations” travelled far beyond the lab.
Heath’s team studied 21 rhesus monkeys. Some breathed smoke from active cannabis, some received THC by injection, and controls breathed smoke from cannabis with the cannabinoids removed. After two to three months, the heavily exposed animals showed changes in electrical recordings from deep brain areas, including the hippocampus, which is central to memory. Under the electron microscope, their brains showed changes at the synapse and inside nerve cells. The control animals did not.

A much larger programme followed. At the US National Center for Toxicological Research in Arkansas, William Slikker, Andrew Scallet and colleagues trained 64 young male rhesus monkeys and exposed them to the smoke of one cannabis cigarette a day, either daily or only at weekends, for a full year. Placebo and sham groups were exposed to extracted smoke or none. Seven months after the last exposure, the team found no changes in brain chemical messengers in the regions measured, and cannabinoid receptors did not appear to be permanently altered. Reduced motivation seen during exposure disappeared within two to three months of stopping.
The picture in rats was less reassuring. In a 1987 study, rats given high oral doses of THC every day for 90 days showed smaller nerve cells and 44% fewer synapses in part of the hippocampus, months after the last dose. Scallet’s 1991 review of the field set this out plainly: in rats, damage appeared only after at least three months of dosing, which he estimated would compare with about seven to ten years in humans, while monkey studies of up to 12 months “have not consistently reported neurotoxicity”. Longer exposures had not been studied. Our guide to cannabis pharmacology explains how THC acts on the receptors involved.
Does long-term cannabis use lower IQ?
In one well-known study, persistent heavy use was linked to a drop of about 6 IQ points by midlife, and about 8 points in people who started in adolescence. That study, from Dunedin in New Zealand, is the strongest evidence that cannabis might harm thinking over the long term. Its causal reading is disputed.
Madeline Meier, Terrie Moffitt, Avshalom Caspi and colleagues reported the findings in the Proceedings of the National Academy of Sciences in 2012. The Dunedin Study has followed 1,037 people born in 1972 and 1973 from birth. Their IQ was tested at age 13, before most had tried cannabis, and again at 38. Cannabis use was recorded in interviews at ages 18, 21, 26, 32 and 38.

People who never used cannabis gained slightly in IQ. Those who met the criteria for cannabis dependence at three or more of the five assessments lost 0.38 standard deviations, which the authors translated as about 6 IQ points. Among those whose dependence began before 18, the decline was about 8 points. Stopping use did not fully restore functioning in the adolescent-onset group. The link remained after the authors accounted for education, other drug and alcohol dependence and recent cannabis use, and friends and family also reported more memory and attention problems in persistent users.
Even if you read Dunedin at face value, it does not show that weed kills brain cells, and two points keep it in proportion. The declines were concentrated in a small group: 38 people with dependence at three or more waves. And the study was about dependence over decades, not occasional use. Our page on medical cannabis tolerance explains how regular use changes the body’s response over time.
What did the twin studies find?
They found that cannabis-using twins did not lose more IQ than the twin who did not use. Twins share family background and many or all of their genes, so this design strips out a lot of the confounding that affects ordinary comparisons. The result suggests that something about the family, not cannabis alone, explains much of the IQ gap.
Nicholas Jackson and colleagues published the analysis in the same journal in 2016, using two US studies of adolescent twins with 789 and 2,277 participants. IQ was tested at ages 9 to 12, before any cannabis use, and again at 17 to 20. As in Dunedin, users scored lower and showed a decline in crystallised intelligence, the knowledge-based part of IQ. But there was no dose-response pattern, meaning heavier use did not bring a bigger drop, and the using twin did not decline more than the abstinent sibling.
An economist, Ole Rogeberg, had raised a similar challenge to Dunedin in 2013. He showed by simulation that differences in socioeconomic background could reproduce the reported pattern, and wrote that the true effect “could be zero”. The Dunedin authors replied that their data did not support that explanation. Neither side claimed to have settled it.
| Study | Who was studied | Main finding |
|---|---|---|
| Heath and colleagues, 1980 | 21 rhesus monkeys, months of heavy exposure | Changes at synapses and in deep brain recordings |
| NCTR programme, 1991 to 1992 | 64 rhesus monkeys, daily or weekend smoke for 1 year | No consistent neurotoxicity; motivation effects faded within months |
| Meier and colleagues, 2012 | 1,037 people in New Zealand, age 13 to 38 | About 6 IQ points lower with persistent dependence, about 8 with teenage onset |
| Jackson and colleagues, 2016 | 3,066 US twins, age 9 to 20 | Using twins did not decline more than non-using siblings |
| Scott and colleagues, 2018 | 69 studies, 2,152 users and 6,575 comparison participants | Small deficit overall; not significant after more than 72 hours abstinent |
Summary of the studies discussed on this page. Animal findings may not apply to people, and the human studies are observational, so none of them alone can prove or rule out a causal effect.
Does your thinking recover if you stop?
For most young users, largely yes, according to the best pooled evidence. A 2018 meta-analysis found that the thinking deficits linked to frequent cannabis use were small, and that they were no longer statistically significant in studies where participants had gone more than 72 hours without cannabis.
J. Cobb Scott, Ruben Gur and colleagues at the University of Pennsylvania combined 69 studies in JAMA Psychiatry, covering 2,152 frequent or heavy users with an average age of about 21 and 6,575 people with little or no cannabis exposure. Overall, users performed slightly worse, a small effect. Age and the age users started did not change the size of the effect. In the 15 studies that required more than 72 hours of abstinence, the gap was close to zero.
The authors concluded that earlier research may have overstated how large and how lasting the deficits are, and that many reported problems may reflect the after-effects of recent use or withdrawal. That fits a simple practical point: THC impairs memory and attention while it is active, and it can stay in the body for days. Our guide to how long dagga stays in your system explains the timelines. The analysis did not look at other outcomes such as psychosis, which our page on the side effects of weed covers.
If you are weighing cannabis for a health reason and are worried about memory or concentration, a registered doctor can talk through the risks for you personally. Create your Professor Green account when you are ready to start the prescriber-led route.
Before you start
In South Africa, medical cannabis containing THC is prescription-only through a registered doctor, usually via a SAHPRA Section 21 authorisation. This is general information for adults 18 and over, not medical or legal advice. The research that raises the most concern involves people who used heavily from their teens, while the brain is still developing. Do not drive or do safety-sensitive work while affected, avoid cannabis if you are pregnant or breastfeeding, and tell your prescriber about any mental health history and any other medicines you take. If someone is very confused, unusually drowsy, has chest pain or cannot be woken, call an ambulance on 10177, or 112 from a mobile.
Who is most at risk?
The research points to people who start young and use heavily for years. In Dunedin, the IQ decline was concentrated in adolescent-onset users with repeated dependence, and did not fully reverse after they stopped. Occasional adult use has not been tied to lasting harm in the stronger studies.
Several factors seem to matter, based on the studies above:
- Age at first use. Adolescent onset was linked to the largest IQ decline in Dunedin, although the twin studies suggest family factors play a big part.
- Frequency and duration. The Dunedin declines tracked the number of assessments at which people met the criteria for dependence, not simply whether they had ever used.
- Recent use. Thinking problems measured within a few days of use shrink sharply after more than 72 hours of abstinence.
- Product strength. The older studies did not measure THC dose. Today’s products can be far stronger, as our guide to THC percentage explains, so dose and potency deserve attention with a prescriber.
What does South African law say?
Adults may use and grow cannabis privately, but buying from an unlicensed seller is dealing, and THC medicines are prescription-only. The Constitutional Court’s 2018 judgment in Minister of Justice and Constitutional Development v Prince protected private use and cultivation by adults. It did not legalise trade.
The Cannabis for Private Purposes Act 7 of 2024 defines “deal in” to include selling, offering for sale and buying, and section 4 makes dealing an offence. Our guide to the Cannabis for Private Purposes Act explains what private use allows. THC is listed in Schedule 6, the entry most recently amended by Government Notice R5181 in Government Gazette 51171 of 6 September 2024. Where no registered medicine suits a patient, a registered prescriber may apply to SAHPRA under section 21 of the Medicines and Related Substances Act to use an unregistered product, which a licensed pharmacy then dispenses.
How Professor Green fits into the process
Professor Green is a South African platform that connects adults to the registered prescriber route for medical cannabis. Registering creates your account and opens the member store-access area so you can begin that journey. Registration is the first step and nothing more: it is not an assessment, not an approval and not a guarantee that any particular product will be suitable or available to you.
The regulated work sits with other parties. A registered healthcare practitioner assesses you and decides whether to prescribe, and what product and strength suit you, taking account of concerns such as memory, mental health and age. Where the product is unregistered, that practitioner applies to SAHPRA for section 21 authorisation. A licensed pharmacy dispenses it, and a courier delivers it where that is permitted. Professor Green does not prescribe, does not dispense and does not decide clinical suitability. Our medical cannabis products page shows the formats a prescriber works with.
The lungs raise a similar myth-versus-evidence question. Our guide to weed smokers lungs after 5 years sets out what the long-term lung function and lung cancer studies found, measured in joint-years.
The same gap between headlines and data shows up in fertility research, as our guide to weed effects on sperm explains study by study.
Cannabis and the brain: frequently asked questions
Does weed kill brain cells?
There is no good evidence that cannabis kills brain cells in people. The idea came largely from a small 1980 monkey study, and a larger year-long monkey programme did not consistently find neurotoxicity. Heavy long-term use, especially from the teens, has been linked to lower IQ.
How many IQ points does weed take away?
In the Dunedin study, people dependent on cannabis at three or more assessments lost about 6 IQ points between age 13 and 38, and adolescent-onset users about 8. Twin studies suggest family factors explain much of that gap, so the true effect of cannabis is uncertain.
Does smoking weed once damage your brain?
There is no evidence that a single use causes lasting brain damage. THC affects memory and attention while it is active, and the research concerns about lasting effects involve heavy use over many years.
Can your brain recover after you stop using weed?
For most young users, the evidence is encouraging. A 2018 meta-analysis of 69 studies found thinking deficits were small and not significant in studies requiring more than 72 hours without cannabis. In Dunedin, stopping did not fully restore IQ in adolescent-onset users.
Is weed worse for the teenage brain?
The strongest concern involves teenage onset. In Dunedin, people whose dependence began before 18 showed the largest IQ decline. Twin studies found no extra decline in using twins versus their siblings, so the cause is debated, but caution before adulthood is sensible.
Can a doctor in South Africa consider memory concerns before prescribing cannabis?
Yes. A registered prescriber assesses your history, including memory, mental health and other medicines, before deciding whether a cannabis medicine suits you. Where the product is unregistered, they apply to SAHPRA under section 21, and a licensed pharmacy dispenses it.
The short answer: does weed kill brain cells?
Does weed kill brain cells? No good evidence shows that it does in people. The myth grew from a small monkey study, and a larger year-long monkey programme did not consistently confirm it, although high-dose rat studies found real changes. In humans, persistent heavy use, especially from the teens, is linked to a lower IQ by midlife, but twin studies suggest family background explains much of that, and most thinking problems in young users fade after a few days without cannabis. The sensible conclusion is to take age, frequency and potency seriously. In South Africa, buying from an unlicensed seller is dealing, and the lawful route to a cannabis medicine runs through a registered prescriber and a licensed pharmacy.
Get legal, doctor-led medical cannabis access in South Africa
If you have been asking does weed kill brain cells because you are worried about your memory or focus, the right next step is a registered prescriber who can weigh that against your health needs. Professor Green connects you with that route so you can pursue medical cannabis access legally and safely.
Sources and 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): 657-690.
- Slikker W, Paule MG, Ali SF, Scallet AC, Bailey JR, “Chronic marijuana smoke exposure in the rhesus monkey. I”, Fundamental and Applied Toxicology, 1991, 17(2): 321-334; Ali SF and colleagues, “… IV: Neurochemical effects”, Pharmacology Biochemistry and Behavior, 1991, 40(3): 677-682; Paule MG and colleagues, “… II: Effects on progressive ratio and conditioned position responding”, Journal of Pharmacology and Experimental Therapeutics, 1992, 260(1): 210-222.
- Westlake TM and colleagues, “Chronic exposure to delta 9-tetrahydrocannabinol fails to irreversibly alter brain cannabinoid receptors”, Brain Research, 1991, 544(1): 145-149.
- Scallet AC and colleagues, “Morphometric studies of the rat hippocampus following chronic delta-9-tetrahydrocannabinol”, Brain Research, 1987, 436(1): 193-198; Scallet AC, “Neurotoxicology of cannabis and THC: a review of chronic exposure studies in animals”, Pharmacology Biochemistry and Behavior, 1991, 40(3): 671-676.
- Meier MH, Caspi A, Ambler A and colleagues, “Persistent cannabis users show neuropsychological decline from childhood to midlife”, Proceedings of the National Academy of Sciences, 2012, 109(40): E2657-E2664.
- Rogeberg O, “Correlations between cannabis use and IQ change in the Dunedin cohort are consistent with confounding from socioeconomic status”, Proceedings of the National Academy of Sciences, 2013, 110(11): 4251-4254.
- Jackson NJ, Isen JD, Khoddam R and colleagues, “Impact of adolescent marijuana use on intelligence: results from two longitudinal twin studies”, Proceedings of the National Academy of Sciences, 2016, 113(5): E500-E508.
- Scott JC, Slomiak ST, Jones JD and colleagues, “Association of cannabis with cognitive functioning in adolescents and young adults: a systematic review and meta-analysis”, JAMA Psychiatry, 2018, 75(6): 585-595.
- Parliament of South Africa, Cannabis for Private Purposes Act 7 of 2024, sections 1 and 4.
- SAHPRA, Amendment to the Schedules, GN R5181, Government Gazette 51171, 6 September 2024.
- Republic of South Africa, Medicines and Related Substances Act 101 of 1965, section 21.
- Constitutional Court of South Africa, Minister of Justice and Constitutional Development and Others v Prince and Others [2018] ZACC 30.

