stock photo / Gergerger7
Bloomberg News

Teens may lose IQ points later in life if they smoke marijuana before age 18, according to a study that comes on the heels of a survey showing that the drug's use has risen in this age group for four straight years.

The research, reported Monday in the Proceedings of the National Academy of Sciences, found an average decline of eight points on IQ, or intelligence quotient, tests done at ages 13 and 38 among those who began using marijuana as a teenager. That compared with no decrease in those who used pot later in life, and a slight increase in those who never used it.

Because marijuana is the most frequently used illegal drug in the U.S., looking into on how it changes the brain is important, said study author Madeline Meier, a post-doctoral researcher at Duke University in Durham, North Carolina, who led the study. Daily use among high school seniors is at a 30-year peak, according to a 2011 survey by the National Institute on Drug Abuse in Bethesda, Md.

"What this says is not don't ever do it, but if you do it during this critical period of development, you'll get these long-term negative changes," said Staci Gruber, an assistant professor of psychology at Harvard University in Cambridge, Mass., who wasn't involved in the study. "In almost every case, the subjects who started as adults don't have declines. Those who started as teenagers do."

Tests for intelligence quotient, or IQ, measure a person's capacity to learn, apply knowledge, and use abstract reasoning. The average score is 100.

Although a drop of eight points may not seem like much, going from a score of 100 to 92 drops someone from being in the 50th percentile in intelligence to the 29th, Duke's Meier said.

The 2011 report by the institute on drug abuse found that about 6 percent of 12th graders smoked pot 20 or more times in the 30 days before the survey; 46 percent of high school seniors had tried it at least once, according to the U.S. agency's annual Monitoring the Future survey of U.S. High Schools.

Today's study used a cohort of more than 1,000 New Zealanders in the Dunedin Study, who have been followed since their births in 1972 through 1973. Previous work, including a 2010 finding by Harvard's Gruber that early marijuana use may change the brain structure, couldn't rule out other underlying factors that may have contributing to the brain changes.

The Dunedin cohort may help settle those questions. The group got IQ tests at 13, before any of them began smoking weed. They were tested again at 38. Friend and family, who were routinely interviewed as part of the study, said these people had attention and memory problems.

The decline in IQ couldn't be explained by alcohol or other drug use, or education gaps between those who smoked pot as adolescents and those who didn't, the study found. The greatest impairments were in processing speed and executive functioning, a name for the mental processes involved in planning, organizing, and detail work.

"Most other studies had data at one time point, and they couldn't rule out that cognitive deficits were already in play," Meier said. "We ruled that out."

The brain makes a lot of changes in adolescence, Meier and Gruber said. Marijuana may interfere with those changes, making it more difficult for those who used as adolescents to think. It's not clear from today's data what dose might be harmful, Meier said.