A moving target, not a verdict
Ask what cannabis does to the brain and you'll get three answers. A daily user gives you one. A weekend smoker gives you another. A researcher who has to hold both at once gives you a third, and that one is the messiest and the most honest. We are a team of neuroscientists, and across three recent studies we tried to pin down exactly where cannabis bites into cognition and psychology, and just as carefully where it doesn't.
Start with scale, because the numbers frame everything. The United Nations Office on Drugs and Crime reported that in 2018 roughly 192 million people worldwide aged 15 to 64 used cannabis recreationally. That's not a fringe habit. It's one of the most popular drugs on earth, used by humans for thousands of years, legal to prescribe or possess in several countries, and wrapped in effects people genuinely like: a lift in mood, a sense of relaxation.
The wrinkle is who uses it. About 35% of people between 18 and 25 have used cannabis, against only 10% of those over 26. The people reaching for it are adolescents and young adults, the same people whose brains are still under construction. That overlap is the whole story in miniature. The heaviest use lands right on the development window where the organ is most plastic, which means most vulnerable to lasting change.
The receptor map behind the effects
To understand what changes, you need to know where the drug lands. Tetrahydrocannabinol, THC, is the main psychoactive compound in cannabis. It works on the brain's endocannabinoid system, a network of receptors that respond to cannabis's chemical components. Those receptors sit thickest in the prefrontal and limbic regions, the areas that run reward and motivation. They regulate the signalling of three chemical messengers that matter a great deal: dopamine, gamma-aminobutyric acid (GABA), and glutamate.
That map tells you what to expect before you run a single experiment. Dopamine is the workhorse of motivation, reward and learning. GABA and glutamate feed cognitive processes, including learning and memory. So when a psychoactive compound floods exactly these circuits, the sensible prediction is that motivation and memory are where you'll see fingerprints. Our work largely confirmed it.
Memory and flexibility under cannabis-use disorder
Cognition is where the effects are clearest, and they are clearest in people who have crossed into cannabis-use disorder. That diagnosis means a persistent desire to use the drug alongside real disruption to daily life, work slipping, education derailed. Roughly 10% of cannabis users meet the criteria, so we're not talking about a scare story applied to everyone. We're talking about a defined subset.
In one study we tested the cognition of 39 people with the disorder, asked to be clean on the testing day, and compared them with 20 people who had never used cannabis or used it only rarely. The disorder group performed significantly worse on memory tests drawn from the Cambridge Neuropsychological Test Automated Battery, the CANTAB. They also showed weaker executive function, the mental machinery behind things like flexible thinking.
Here's the detail we keep coming back to. The degree of impairment tracked the age at which a person first started using. The younger the onset, the worse the executive functioning. That's a sobering dose-response relationship, and it lines up precisely with the vulnerability argument above. Milder users weren't immune either. They tended to make riskier decisions than non-users and showed more trouble with planning.
Men, women, and a real sex gap
Most of the research in this field has been done on men, which is its own kind of problem. When we looked carefully, the effects split by sex, and not in a way that cancels out. Male cannabis users had poorer memory for visually recognising things. Female users instead showed more problems with attention and executive function. Two different profiles, from what people assume is one drug doing one thing.
We didn't let those differences hide behind confounds. They held up even after we controlled for age, IQ, alcohol and nicotine use, mood and anxiety symptoms, emotional stability, and impulsive behaviour. Whatever we're measuring, it isn't a side effect of someone being more anxious or drinking more.
The reward system: a quieter result than expected
If cognition is where cannabis shouts, reward is where it whispers, at least at moderate use. Earlier work had suggested that the brain's reward and motivation circuits get disrupted by cannabis, which could explain why some users feel less driven and less gratified by success at school or work. That's a serious clinical claim, so we tested it directly with brain imaging.
We put participants in a scanner and showed them orange or blue squares. An orange square could lead to a monetary reward, after a delay, if the participant responded in time. This let us watch the brain react to reward itself. We focused on the ventral striatum, a central hub of the reward system. The result was underwhelming in the best scientific sense: we found no direct effect of cannabis in the ventral striatum. The effects were subtle.
But read that finding with the right caveat. Our participants were moderate users. The blunting may only become pronounced in severe, chronic use, the kind you see in full cannabis-use disorder. A null result at low dose is not the same thing as no risk at high dose, and we'd be careless to imply it is.
Anhedonia, and the lockdown years
Cannabis can reach the part of mental health concerned with pleasure itself. We found it linked to higher anhedonia, the reduced ability to feel pleasure, in adolescents. What sharpened that signal was timing. The effect stood out particularly during the COVID-19 pandemic lockdowns, when isolation was already dragging at young people's mood. A drug blunting reward inside an environment already short on reward is a bad combination.
The psychosis question, and why predisposition matters enormously
Then there is the claim people argue about loudest, and the one where the data deserve more nuance than either side gives it. Adolescent cannabis use has been reported as a risk factor for psychotic experiences and for schizophrenia. The honest version isn't "cannabis causes psychosis." It's that the size of the risk swings on who you are.
One study of 2,437 adolescents and young adults, aged 14 to 24, captured that swing perfectly. Among cannabis users with no predisposition to psychosis, the risk of psychotic symptoms rose by about six percentage points, from 15% to 21%. Modest. Among users who did carry a predisposition, the kind flagged by scoring high on a checklist of paranoid ideas and psychoticism, the jump was 26 points, from 25% to 51%. The same drug, two utterly different outcomes, because the starting point was different. We don't fully know the mechanism, though dopamine and glutamate are the leading suspects in the neurobiology. It's also the central question in our broader work on whether cannabis psychosis may represent a distinct subtype rather than ordinary schizophrenia with a trigger bolted on.
There's a structural clue too. A separate study of 780 teenagers tied the link between cannabis use and psychotic experiences to a small brain region called the uncus. The uncus sits at the junction of the parahippocampus, involved in memory, and the olfactory bulb, involved in smell, and it is thick with cannabinoid receptors. It has been implicated in schizophrenia before. A plausible anatomical thread, but a thread is not yet a proof.
Temporary, permanent, or still an open question
So where does that leave us? Effects depend, in part, on dosage, meaning frequency, duration and strength, layered on top of sex, genetic vulnerability, and the age at which someone starts. None of these factors acts alone. The hardest unknown is reversibility, and we should say the quiet part plainly: we don't yet know which changes stick. A summary of many studies suggests that with mild use, effects may weaken after periods of abstinence, the territory covered by the literature on cognitive and executive impairment that persists after the high fades. At the heavier end, where daily use reshapes the brain's baseline, the picture looks less forgiving.
Even granting the uncertainty, the weight of the evidence points one way for one group in particular. Prolonged, problematic use in young people whose brains are still developing carries real cognitive and psychological costs. The moderate weekend user is not the same risk profile. But the 18-year-old using daily is exactly the case we'd want them, and their doctor, to take seriously. The honest scientific answer is neither the panic nor the all-clear. It's "it depends, and here's exactly what it depends on."
This article draws on research by Barbara Jacquelyn Sahakian, Christelle Langley, Martine Skumlien and Tianye Jia, originally published via Neuroscience News.
