2010年4月27日 星期二

Study: Brain Exercises Don't Improve Cognition

Study: Brain Exercises Don't Improve Cognition

You've probably heard it before: the brain is a muscle that can be strengthened. It's an assumption that has spawned a multimillion-dollar computer-game industry of electronic brainteasers and memory games. But in the largest study of these games to date, a team of British researchers has found that healthy adults who undertake computer-based "brain training" do not improve their mental fitness in any significant way.

The study, published online on Tuesday by the journal Nature, tracked 11,430 participants through a six-week online study. The participants were divided into three groups: the first group undertook basic reasoning, planning and problem-solving activities (like choosing the "odd one out" of a group of four objects); the second completed more complex exercises of memory, attention, math and visual-spatial processing that were designed to mimic popular brain-training computer games and programs; and the control group was asked to use the Internet to research answers to trivia questions. (See different workouts for your brain.)

All participants were given a battery of unrelated, benchmark cognitive-assessment tests before and after the six-week study. These tests, designed to measure overall mental fitness, were adapted from reasoning and memory tests that are commonly used to gauge brain function in patients with brain injury or dementia. All three study groups showed marginal — and identical — improvement on these benchmark exams.

But the improvement had nothing to do with the interim brain-training, says study co-author Jessica Grahn of the Cognition and Brain Sciences Unit in Cambridge, Mass. Grahn says the results confirm what she and other neuroscientists have long suspected: people who practice a certain mental task — for instance, remembering a series of numbers in sequence, a popular brainteaser used by many video games — improve dramatically on that task, but the improvement does not carry over to cognitive function in general. (Indeed, all the study participants improved in the tasks they were given; even the control group got better at looking up answers to obscure questions.) The "practice makes perfect" phenomenon probably explains why the study participants improved on the benchmark exams, says Grahn — they had all taken them once before. "People who practiced a certain test improved at that test, but improvement does not translate beyond anything other than that specific test," she says. (See a Q&A on the future of brain enhancement.)

The authors believe the study, which was run in conjunction with the BBC television program Bang Goes the Theory, undermines the sometimes outlandish claims of brain-boosting websites and digital games. On TIME.com in 2009, Anita Hamilton wrote about HAPPYneuron, a brain-training website that invites visitors to "give the gift of brain fitness" and claims that its users see "16%+ improvement" through exercises such as learning to associate a bird's song with its species and shooting basketballs through virtual hoops. Hamilton also took note of Nintendo's best-selling game Brain Age, which promises to "give your brain the workout it needs" through exercises like solving math problems and playing Rock, Paper, Scissors on the handheld DS. However, "the widely held belief that commercially available computerized brain-training programs improve general cognitive function in the wider population lacks empirical support," the Nature paper concludes.

Not all neuroscientists agree. In 2005, Torkel Klingberg, a professor of cognitive neuroscience at the Karolinska Institute in Sweden, used brain imaging to show that brain-training can alter the number of dopamine receptors in the brain — dopamine is a neurotransmitter involved in learning and other important cognitive functions. Other studies have suggested that brain-training can help improve cognitive function in elderly patients and those in the early stages of Alzheimer's disease, but the literature is contradictory. (See a potential test to predict the presence of Alzheimer's disease.)

Klingberg has developed a brain-training program called Cogmed Working Memory Training, and owns shares in the company that distributes it. He tells TIME that the Nature study "draws a large conclusion from a single negative finding" and that it is "incorrect to generalize from one specific training study to cognitive training in general." He also criticizes the design of the study and points to two factors that may have skewed the results.

On average, the study volunteers completed 24 training sessions, each about 10 minutes long — for a total of three hours spent on different tasks over six weeks. "The amount of training was low," says Klingberg. "Ours and others' research suggests that 8 to 12 hours of training on one specific test is needed to get a [general improvement in cognition]."

Second, he notes that the participants were asked to complete their training by logging onto the BBC Lab UK website from home. "There was no quality control. Asking subjects to sit at home and do tests online, perhaps with the TV on or other distractions around, is likely to result in bad quality of the training and unreliable outcome measures. Noisy data often gives negative findings," Klingberg says.

Brain-training research has received generous funding in recent years — and not just from computer-game companies — as a result of the proven effect of neuroplasticity, the brain's ability to remodel its nerve connections after experience. The stakes are high. If humans could control that process and bolster cognition, it could have a transformative effect on society, says Nick Bostrom of Oxford University's Future of Humanity Institute. "Even a small enhancement in human cognition could have a profound effect," he says. "There are approximately 10 million scientists in the world. If you could improve their cognition by 1%, the gain would hardly be noticeable in a single individual. But it could be equivalent to instantly creating 100,000 new scientists."

For now, there is no nifty computer game that will turn you into Einstein, Grahn says. But there are other proven ways to improve cognition, albeit by small margins. Consistently getting a good night's sleep, exercising vigorously, eating right and maintaining healthy social activity have all been shown to help maximize a brain's potential over the long term.

What's more, says Grahn, neuroscientists and psychologists have yet to even agree on what constitutes high mental aptitude. Some experts say physical skill, which stems from neural pathways, should be considered a form of intelligence — in that case, masterful ballet dancers and basketball players would be considered geniuses.

Jason Allaire, co-director of the Games through Gaming lab at North Carolina State University says the Nature study makes sense; rather than finding a silver bullet for brain enhancement, he says, "it's really time for researchers to think about a broad or holistic approach that exercises or trains the mind in general in order to start to improve cognition more broadly."

Or, as Grahn puts it, when it comes to mental fitness, "there are no shortcuts."

— With reporting by Tara Kelly / London