Many people think of IQ as a genetic trait, like brown eyes or short legs: You're born with it and you're stuck with it. Now, a growing body of research is showing that a person's IQ can rise – and even fall – over the years.

Scores can change gradually or quickly, after as little as a few weeks of cognitive training, research shows. The increases are usually so incremental that they're not immediately perceptible to individuals, and the intelligence-boosting effects of cognitive training can fade after a few months.

In the latest study, 33 British students were given IQ tests and brain scans at ages 12 to 16 and again about four years later by researchers at the Wellcome Trust Centre for Neuroimaging at University College London; 9% of the students showed a significant change of 15 points or more in IQ scores.

On a scale where 90 to 110 is considered average, one student's IQ rose 21 points to 128 from 107, lifting the student from the 68th percentile to the 97th compared with others the same age, says Cathy Price, professor of cognitive neuroscience at the center and co-author of the study, published last month in Nature. Another student's score skidded out of the "high average" category, to 96 from 114.

Swings in individual IQ scores are often written off as the product of measurement error or a test subject having a bad day. But MRIs in this study showed changes in gray matter in areas corresponding to fluctuations in the kids' skills, Dr. Price says. Although the sample size is small, the study drew wide attention because it is among the first to show how changes in IQ scores may be reflected in actual shifts in brain structure.

"There are many myths about IQ, such as the notion that IQ is a fixed number or that it is a crystal ball for future performance," says Eric Rossen, director of professional development and standards for the National Association of School Psychologists in Bethesda, Md.

The first reliable tests of intelligence in the U.S. were published in the early 1900s, says Alan S. Kaufman, clinical professor of psychology at Yale University and co-author of several IQ tests. Scores compare people to others of the same age based on a wide range of cognitive questions and tasks, from processing information and analyzing patterns, to solving age-appropriate math problems and recalling facts or vocabulary. A score in the 90 to 110 range is considered average. A "genius" may score 140 and above, he says.

IQ tests have been a target of ongoing criticism. Their use led to the misclassification of many children as "intellectually disabled" in the 1970s and 1980s. Similar cognitive tests used by employers to screen recruits have been attacked as discriminatory against African-American and Hispanic job candidates.

Today in schools, individual IQ-type tests are limited mainly to helping plan instruction for some children with specific learning disabilities and helping identify students for gifted programs. Kathleen Lundquist, president of APTMetrics, a Darien, Conn., human-resources consulting firm, says cognitive tests in the workplace today are often revised to eliminate adverse effects on minorities and are most often used as a screening tool for entry-level jobs.

There are practical steps people can take to see longer-term IQ changes. A 30-year study at the National Institute of Mental Health found that people whose work involves complex relationships, setting up elaborate systems or dealing with people or difficult problems, tend to perform better over time on cognitive tests. Test scores of people whose jobs are simple and require little thought actually tend to decline, according to the research, published in 1999 in Psychology and Aging.

New tasks stimulate the brain most. When researchers at the University of Hamburg subjected 20 young adults to one month of intense training in juggling, they found an increase in the corresponding gray matter in the brain as early as seven days after the training began. The added gray matter receded when the training was stopped, although the participants were still able to juggle, says the study, published in 2008 in PLoS One.

IQ tests don't measure such abilities as creativity, common sense or social sensitivity. They do assess many kinds of knowledge and abilities, including abstract reasoning skills. Rising scores in abstract reasoning are the main reason average IQ scores have been increasing by about three points every decade since the 1930s, based on studies by James Flynn, a professor emeritus of political studies at the University of Otago in New Zealand. That may be partly because children spend nearly twice as many years in school, on average, than children decades ago, says Wendy M. Williams, a professor in the department of human development at Cornell University.

Click here to read more from The Wall Street Journal.