Does Cannabis Affect Memory and Cognitive Function?

Regular cannabis use is linked with small cognitive deficits, which mostly shrink after a few days without use, but decades of use starting in adolescence is linked with lasting IQ decline. A 2018 meta-analysis found small effects in young people, while the Dunedin birth cohort found long-term users lost about 5.5 IQ points by midlife. All of this evidence is observational. This page is not medical advice.

Not medical advice. Cannabis laws vary. If cannabis use is causing problems for you, speak to a doctor.

What the meta-analyses show

The 2018 review concluded that the associations were “small and may be of questionable clinical importance for most individuals.” It suggested earlier studies may have overstated how large and lasting the deficits are, with some reflecting recent use or withdrawal.

The 2020 adult meta-analysis looked at people using cannabis at least 4 days a week for at least 2 years, tested after at least 12 hours without use. Attention, working memory and information processing differences were small and not significant. Longer abstinence (25 days or more) did not change the results, except that executive function differences were no longer significant.

Short-term effects: during and after use

A 2016 systematic review of about a decade of studies found that verbal learning and memory and attention are the abilities most consistently impaired, both during intoxication and in long-term users. Psychomotor skills, such as coordination and reaction time, are most affected while intoxicated. The review found little evidence that regular users become tolerant to these acute effects, and said problems with verbal memory, attention and some executive functions may last through long periods without use, though recovery is under-researched.

Many products are now sold with added CBD on the claim that it softens THC’s effects. A 2023 randomized crossover trial in 46 healthy, infrequent users tested cannabis with four different CBD:THC ratios. THC impaired delayed verbal recall, and none of the CBD doses significantly changed that. The authors concluded that “at CBD:THC ratios most common in medicinal and recreational cannabis products, we found no evidence that CBD protects against the acute adverse effects of cannabis.”

Does memory recover after stopping?

The meta-analysis finding that deficits fade after about three days fits with a 2018 randomized study of 88 regular users aged 16-25. Participants were assigned to 4 weeks without cannabis, confirmed by urine tests, or to monitoring with no requirement to stop. The abstinence group had better verbal memory, an improvement that came mostly in the first week. Attention improved equally in both groups, which the authors put down to practice on the tests. This shows that part of the memory gap in regular users reflects recent use, at least in young people over a few weeks.

The long-term cohort evidence

The Dunedin Study followed 1,037 people born in New Zealand in 1972-1973. Testing at age 13, before cannabis use, and again at 38 found that persistent use was linked with decline across many areas of thinking. Decline was concentrated in people who started in adolescence, and stopping did not fully restore functioning in that group.

By age 45, the same cohort showed that long-term users had IQ decline from childhood (mean -5.5 points), poorer learning and processing speed, and smaller hippocampal volume. These deficits were not explained by tobacco, alcohol, other drug use, childhood socioeconomic status or low childhood self-control, and were smaller or absent in people who had quit. They were also smaller or absent in long-term tobacco users, long-term alcohol users and people who only used cannabis recreationally in midlife. Friends and family of long-term users also reported more memory and attention problems in daily life, so the gaps were noticeable outside the test room. The study kept 94% of its members to age 45, which is unusually high and reduces the risk that dropouts skewed the results.

Is cannabis the cause?

Not everyone agrees the long-term findings are caused by cannabis. Two longitudinal twin studies (n = 789 and n = 2,277) tested IQ at ages 9-12, before any use, and again at 17-20. Users scored lower and showed a decline in crystallized intelligence, but there was no dose-response relationship, and users did not decline more than their twins who did not use cannabis. The authors concluded that the declines may come from family factors that lead to both cannabis use and lower intellectual attainment.

The twin studies followed people only to about age 20, while Dunedin followed them for decades. The two are not directly comparable: the twin results argue against a large effect of adolescent use over a few years, while Dunedin’s strongest findings involve persistent use into midlife.

How to read the evidence

The meta-analyses mostly compare users with non-users at one point, so they cannot separate cause and effect. The Dunedin study measured people before they started, which is stronger, but it is still one cohort. Overall, occasional adult use seems linked with small, mostly short-lived effects, while heavy long-term use from adolescence has the most concerning findings.

The studies do not all agree on age of onset. In the 2020 adult meta-analysis, how long people had used cannabis and how young they started did not change the results, while Dunedin found decline concentrated in people who started as teenagers. Different study designs and different definitions of heavy use probably explain part of the gap. Products have also changed since the long cohorts began, so the findings may not map neatly onto what is sold today.

Andrew Huberman discusses these risks on his cannabis episode and with researcher Matthew Hill.

Practical points

• Expect a lag: In the 2018 analysis, deficits diminished after more than 72 hours without use, so recent use may affect study or work.
• Adolescence matters most: The strongest long-term findings involve early, persistent use.
• Don’t count on CBD: Added CBD did not reduce THC’s effect on memory in the 2023 trial.
• Plan around important tasks: Verbal memory and attention are the most affected abilities, so exams, learning and driving are where effects matter most.
• Look after sleep: Sleep matters for memory; see the sleep guide.

For related topics, see alcohol and the brain, nicotine and focus and the focus and cognition guide.