How many times have you seen headlines like this?
Fox News writes: Pot Does Lower IQ, Study Finds
On, the other side of the spectrum, the Washington Post opines: No, Marijuana Use Doesn’t Lower Your IQ
So, who’s right?
“Yes, Cannabis Lowers IQ.”
Most of the reports that cannabis lowers IQ rely on one study: the seminal Dunedin Study led by Madeline Meier. In all fairness, it’s one of the best studies we have to date. Most research of this kind is retrospective (or captures a small period of time); the Dunedin Study, on the other hand, is a prospective cohort study.
In retrospective studies, individual outcomes of the participants are known from the outset, and the investigators look back in time to determine how various factors influenced the outcome. Prospective studies, on the other hand, don’t know the individual outcomes from the outset; they follow a group of people over time, identifying predicted outcomes and determining how various factors influence the outcomes.
In the Dunedin Study, researchers followed 1,037 people born between 1972 and 1973 in Dunedin, New Zealand, from birth to age 38. In fact, their study pool represented 91% of all eligible births. The study—like all studies—had limitations. But, unlike far too many studies, the authors were candid in what their limitations were, and tried to minimize their influence by excluding many of the most common confounding factors that could provide alternative explanations for an IQ decline, including:
- Acute or residual cannabis intoxication
- Tobacco dependence
- Hard-drug dependence (e.g., heroin, cocaine, amphetamines)
- Alcohol dependence
There are other potentially unaccounted factors that could influence outcomes. For example, could it be people with unknown (or undefined) shared characteristics are more likely to misuse cannabis at a young age, and these shared characteristics are associated with a decline in IQ? We know nutrition, industrial toxins, stress, and exposure to trauma are a just few factors that can influence IQ (and possibly the risk of addiction).
However, it’s impractical to control for every possible confounding factor. What we expect is that the authors are forthcoming in their study’s limitations, and that they make their best effort to control what they can. The Dunedin team clearly did that. It was well designed, and the authors were nuanced in their conclusions.
The authors examined six hypotheses:
- Cognitive decline: Persistent cannabis consumers demonstrate greater decline in test performance from childhood to adulthood than nonusers.
- Specificity: Are impairments confined to specific neuropsychological domains or are they present across each of the five specific domains? The authors hypothesized that impairments aren’t limited to specific cognitive domains.
- Education: Some evidence suggests staying in school can boost one’s intelligence. Could persistent cannabis consumers experience neuropsychological decline simply because they abandoned academics in favor of other opportunities?
- Everyday Cognition: Does cannabis-induced neuropsychological impairment translate into functional problems in daily life? The authors posit that it would.
- Developmental Vulnerability: Cannabis has heightened toxic effects on the developing brain, thus adolescents are particularly vulnerable to the effects of persistent cannabis use.
- Recovery Hypothesis: Former persistent users who quit or reduce their cannabis use may be able to restore their neuropsychological health.
What were their conclusions?
- Early initiation to cannabis correlates to a decline in IQ as adults, with more persistent use associated with a greater decline in IQ.
- Predictably to some, surprisingly to others, they found no decline in IQ among individuals who waited until they were adults before using cannabis. This finding was consistent not only among casual users, but also among those with a use disorder.
How did abstainers compare to users?
This is where it gets interesting.
Abstainers: Of the 1,037 study participants, 242 reported never having used cannabis. Those participants experienced a modest uptick in their IQ—from 99.84 to 100.64.
Casual Users: 479 participants reported prior cannabis use, but were never diagnosed with a use disorder. This group experienced a modest decline in IQ—from 102.32 to 101.25. Roughly one point.
Problematic Users: Among the 38 participants who met the criteria for cannabis use disorder at three out of five follow-up assessments (ages 18, 21, 26, 32, and 38), the IQ decline was far more profound. They experienced a decline nearly six points—from 99.68 to 93.93.
What was particularly interesting is the IQ difference between those reporting being lifelong abstainers versus those who reported prior cannabis (but had never met the criteria for a use disorder). The average IQ for the latter was nearly 2.5 points higher than the abstinent group prior to induction into cannabis. At age 38, due to modest upticks in the “never used” group and the modest decline in the “used, never diagnosed” group, the IQ gap narrowed. But the “used, never diagnosed” group managed to maintain a 0.61 point IQ advantage. Perhaps that finding warrants a headline: “Study: Cannabis Abstainers Are Less Intelligent Than Casual Users.”
“No, Cannabis Doesn’t Lower IQ.”
Contrary to many assertions, the Dunedin Study was well designed and the authors attempted to control many of the most common confounding factors that could have tainted the results. Interestingly, many of those who initially criticized this study as biased later praised the authors’ follow-up study (analyzing the same populations) that found the only negative physical health effect associated with long-term cannabis use was an increased risk of gum disease.
Nonetheless, several notable critiques emerged:
Declines could be explained by socioeconomic factors. Norwegian research economist Ole Røgeberg suggested socioeconomic factors could account for the decline. The basis of his argument is that cannabis use is more common among those of lower socioeconomic status, and that education temporarily boosts the IQ of these children, creating a false impression that IQ declines later in life. However, even applying these potential factors, Meier’s findings (which are consistent with other studies) have found that socioeconomic background is a poor predictor of cannabis use. Cannabis is popular across all backgrounds. Likewise, Meier didn’t find any differences in adolescent IQ versus adult IQ among those who came from lower socioeconomic backgrounds.
Personality traits explain the decline. British behavioral scientist Dr. Michael Daly suggested personality traits could explain positive non-causal associations between cannabis use and cognitive functioning. High levels of openness to experience could lead people to seek out activities—including cannabis experimentation—that promote cognitive functioning. Daly analyzed data from 6,401 individuals who participated in the British study. His post hoc analysis found openness positively predicted cannabis use and an increase in neuropsychological functioning.
Likewise, he challenged Meier’s findings, suggesting conscientiousness (one of the five defined personality traits) may explain the cannabis decline. However, Daly didn’t assess personality—his primary predictor—until well after cannabis initiation, at age 50. Early initiation to cannabis was a central feature of Meier’s study. Moreover, Meier applied the hypothesis to her data, and found that it didn’t alter the outcomes to a statistical significance.
Sample size is too small. Others have also weighed in. Columbia University’s Dr. Carl Hart noted that Meier’s study had a very small sample of heavy users—only 38 people. The small sample size, he claims, limits how generalizable the results could be. He’s right: 38 people is a small sample size. But it should be noted that Meier’s study included three other subsets of early onset users, all of which experienced declines proportional to how persistent their use was. However, there are still likely limits to how generalizable the studies are. Why, for example, do those with one use disorder diagnosis start with the lowest IQ (lower than even those diagnosed three times), while those diagnosed twice have the second highest starting IQ (102.14)—second only to the “tried, never diagnosed” group—of all groups?
Twin Studies on Cannabis and IQ
As I mentioned previously, it’s virtually impossible to eliminate every potential confounding variable. However, one way to eliminate most of them is by conducting a study on twins. And, at least one such study did just that. A multi-institutional study, “Impact of adolescent marijuana use on intelligence: Results from two longitudinal twin studies,” compared IQ changes in twin siblings who either used or abstained from cannabis during a 10 year period. After accounting for potentially confounding familial factors, the scientists found no link between cannabis use and a decline in IQ. In twin sets where one twin used cannabis and the other didn’t, they found comparable declines in IQ, potentially validating their hypothesis—there were other familial confounds (such as genetics or home environment) that could account for a decline in IQ scores.
This particular study provided the most convincing evidence to counter the Dunedin Study. However, it’s not without notable limitations. For one, the study contained little detail on frequency of use. Moreover, the study looked at two populations (Minnesota and Los Angeles), but used different surveys for each. The survey for Los Angeles posed questions with far less specificity than Minnesota. For example, if a youngster pulled a Bill Clinton and tried it but “didn’t inhale,” he or she would still be categorized as a “user.”
The Final Verdict
For adults, cannabis use doesn’t seem to impact IQ. However, chronic use is associated with adverse cognitive effects that seem to be reversible. For teens, the jury may be still out on how much of an impact cannabis use has. But we do know that the brain is particularly vulnerable while it still developing.
I suspect there’s truth to both sides of the debate. One, that early (and frequent) cannabis use can have persistent (and possibly irreversible) consequences. Two, it’s also likely that cannabis use is not the only factor contributing to a decline in IQ. If a kid is engaging in frequent cannabis use, chances are there are other things going on at home or in their life that may influence their addiction risk and potentially affect their IQ. However, in reviewing most of the research out there, Meier’s study thus far puts forth the most persuasive evidence that early onset cannabis use likely has a negative impact on IQ.
The effect of cannabis exclusive of other factors may not be as profound as presented in the Dunedin Study. But the findings shouldn’t be discounted. How much does it really matter if cannabis causes a two point or six point decline? Both suggest adverse outcomes. Perception of risk—not the legality or illegality of cannabis—influences a teen’s decision to use or not. And, as the stigma around cannabis diminishes, it’s important that teens understand cannabis is not completely harmless. And that the use of it, particularly frequent use, carries risks.