Marijuana might not be the culprit in adolescent IQ decline

Although the two are linked, there seems to be another cause underlying both.

by Cathleen O'Grady

flickr user: Frank

The widely mocked stoner sloth anti-drug campaign typifies a long-held stereotype about marijuana—it makes you slow, stupid, dull. There’s been some research backing up the stereotype, suggesting that marijuana usage can be linked to declines in intellectual function.

But it’s hard to establish that marijuana is definitely the culprit. It’s not like conducting a clinical trial with marijuana would be feasible given the drug's status. So it’s necessary to look in other places for a ready-made control group that can help give a clearer picture of what’s causing the decline.

A paper in this week’s PNAS does just that by looking at marijuana use in twins. The research found that while there was a link between marijuana and lower scores on IQ tests, there seems to be another mystery factor underlying both the drug use and the cognitive decline.

Does marijuana change the brain?

Marijuana use during adolescence has been singled out as a concern because the adolescent brain is still undergoing important development. Because so many teens smoke weed, if it does have any negative effects, they are likely to be pretty large when taken across society as a whole.

But past research on the topic has had mixed results. Some studies suggest that teenagers who use marijuana have decreased intelligence, poorer attention, and reduced memory. The problem with these studies is that they look at a snapshot in time, making it difficult to figure out whether the marijuana use comes before or after the cognitive effects.

Studies that look at marijuana use over time also have their problems. Some of them do find evidence of decline following marijuana use, but there’s a problem of confounding factors. It’s possible that people with certain cognitive patterns are more likely to use marijuana. It could also be that a third cause is driving both marijuana use and cognitive performance.

“Part of this disagreement in the literature speaks to the complexity of trying to infer causal mechanisms from correlational data,” write the authors of the PNAS study. They identify three things you could expect to see in the data if marijuana is causing cognitive problems. First, you’d expect to see marijuana use happening before the cognitive decline. Secondly, you’d see a greater decline for heavier users. And finally, you’d see the effects of marijuana regardless of someone’s genetics or family background.

Stoner twins

To test whether these things are happening, the authors pulled data from two different twin studies, gathering information from slightly more than 3,000 individual twins. Twin studies are an important way to get better insight into confusing causal pictures because they can control for two of the major confounding factors: family background and genetics.

Because identical twins share nearly all their DNA and non-identical twins share only around half their DNA, seeing greater similarity between identical twins tells us how much difference can be explained by genetics.

To establish a baseline, the researchers used a number of IQ sub-tests to assess their participants’ cognitive function “before marijuana involvement” between the ages of 9 and 12. Next, they conducted another set of tests when the subjects were 17 to 20 years old. At each stage, they asked the participants about their marijuana usage and usage of other drugs (including alcohol) using standard diagnostic interviews designed to ascertain substance abuse.

This is potentially a weakness in the research. It’s intuitive to think that teenagers would be likely to lie about their use of drugs, and there’s some research backing up this intuition. Others have found that surveys like these have fair but not flawless accuracy. It’s also possibly a little naive to take it as certain that 12-year-olds have never smoked weed.

Crystallized intelligence takes a hit

The researchers found that overall, marijuana users had lower intelligence test scores in their adult IQ tests. However, there were other interesting effects in the data that painted a more complicated picture than a simple neurotoxicity effect of marijuana.

There are two kinds of intelligence measured in a standard intelligence test: fluid and crystallized intelligence.

Fluid intelligence refers to people’s skill at solving problems that don’t require them to have acquired certain knowledge, and it's assessed through various kinds of logical puzzles. Crystallized intelligence refers to people’s ability to use their acquired knowledge to solve problems. There are a few tests that assess this knowledge, including vocabulary tests.

When it came to vocabulary tests, the group from one of the two twin studies showed marijuana users having lower scores in childhood before (in all likelihood) they started using marijuana. The same didn’t hold for the other twin study or for most of the other IQ sub-tests, so it’s not clear if this is just a fluke.

Vocabulary tests also showed a difference following marijuana use, with smokers showing declining scores and non-smokers seeing an increase. The effects were small, with smokers around four IQ points lower than non-smokers—a small but statistically significant result. It suggests that crystallized intelligence is seeing an effect, but fluid intelligence isn’t.

What the research didn’t find is any evidence of a dose effect—smoking more marijuana didn't have a more severe impact. They also found that in twin pairs where one twin smoked and the other didn’t, they both showed a similar decline. This was true for identical and non-identical twins.

These results indicate that while something is causing a decline in both twins, it doesn’t seem to be marijuana. It also doesn’t seem to be genetics. The obvious remaining culprit is family background. Of course, this was a very small section of the overall sample, with only 47 twin pairs fitting this description, so it will definitely be necessary for other researchers to try to replicate the results.

Family culture

These results are difficult to interpret, and they could mean different things. The authors suggest that it’s not marijuana causing the decline but rather something specific to certain families that causes both cognitive decline and marijuana use. This could be the result of poverty, but it could also be family culture differences that exist regardless of poverty, according to the authors. Some examples might be how much parents monitor their children or how much they emphasize the importance of education.

The study comes after some recent news that previous research on the innocuousness of marijuana might not be exactly flawless. Yesterday, the watchdog blog Retraction Watch posted a note drawing attention to problems with a study released last August. This paper initially found no ill health effects from marijuana usage, but it has subsequently been updated with slightly different results—a re-analysis of the data found increased risk of psychotic disorders.

Although the recent PNAS paper is based on entirely different research, it’s worth looking for parallels between the two to watch out for similar pitfalls. In the case of the corrected study, the results were controversial partly because they contradicted large bodies of previous research. That’s always worth being wary of because while it could suggest a needed overhaul of current thinking, it’s also possible—even likely—that it’s just the result of an error.

The PNAS paper doesn’t fly in the face of existing studies; it adds incrementally to a slowly building picture, contradicting some studies and tying in with others. Nonetheless, it’s still just one study with its own set of limitations. In particular, the results often show differences between the data from the two different twin studies, which means that it can contribute to our understanding without definitively clarifying things.

“We cannot rule out a neurotoxic effect of long-term marijuana use,” the authors caution. Because the participants were all young adults at their latest assessment, it’s possible that ongoing use could eventually do permanent damage in adulthood.

Even though this isn’t a definitive answer on the question of marijuana and neurological development, it does open up some very interesting new questions for researchers to answer in future. Looking at family behaviors that increase the chances of both marijuana use and cognitive decline could turn up some important information.