While programs to improve students’ working memory are among the hottest new education interventions, new studies are calling into question whether exercises to improve this foundational skill can actually translate into greater intelligence, problem-solving ability, or academic achievement.
Working memory is the system the mind uses to hold information during decisionmaking and analysis. As much as half of the variation in individual intelligence can be explained by differences in working-memory capacity, research shows. Working memory has come to be considered by researchers and educators as a key leverage point in boosting brainpower overall—and programs designed to strengthen it are already finding their way into some schools and homes.
But a systematic review of 23 studies on working-memory training programs, published online last month by the journal Developmental Psychology, found such training produced few long-term benefits to working-memory skills and no improvements to other cognitive skills like verbal ability, attention, word decoding, or arithmetic.
And a randomized, controlled study to be published online next week in the Journal of Experimental Psychology: General, found that improving on a key task used in working-memory training did not lead to improvements on any of a battery of 17 cognitive-ability measures, including problem-solving intelligence, multitasking, and perception speed.
“I’m a little torn,” said Thomas S. Redick, an assistant psychology professor and working-memory researcher at Indiana University-Purdue University Columbus and the lead author of the randomized trial. “I’m happy to see an application of basic research being used in the classroom, but I’m not sure it’s ready for prime time. For it to already be implemented in classrooms, particularly in a time of budget cuts, makes me wonder whether it might not be the most effective use of resources.”
Big Brain Business
Brain training, and especially working-memory training, has become big business thanks to the confluence of increased research in neuroplasticity—the brain’s ability to change physically in response to experience—and rising interest in brain boosting from aging Baby Boomers and achievement-focused parents.
The San Francisco-based market-research firm SharpBrains estimates that between 2005 and 2009, the worldwide market for “brain-fitness software,” including working-memory training programs, rose 31 percent, to $295 million, with $148 million of those purchases coming from the United States. There are already more than a half-dozen widely used programs intended to train working memory alone, including Pearson’s Cogmed Working Memory Training, Memosyne Ltd.'s JungleMemory, and Mind Sparke’s Brain Fitness Pro.
“The brain-fitness software industry is only in its infancy,” wrote SharpBrains Chief Executive Officer Alvaro Fernandez in an online commentary on the market. “It is an emerging and largely unregulated market where many products have limited clinical validation and often present confusing claims that make it difficult for consumers to separate wheat from chaff.”
Research by Nelson Cowan, the director of the Working Memory Lab at the University of Missouri-Columbia, has shown that a child’s working-memory capacity naturally increases as he or she ages. While 7-year-olds typically can handle three to five “chunks” of unrelated information at a time, by age 12, children can handle about seven, as many as adults.
Neuroplasticity research has also shifted the field’s conception of how the brain works from a computer hard-wired with skills to a muscle that can be built up with exercise.
“These neuroplasticity models are kind of similar in terms of driving changes to the brain at a physiological level to what you need to do to build muscles: You need to do lots of reps; it needs to be frequent and intense,” explained Paula A. Tallal, a co-director of the Center for Molecular and Behavioral Neuroscience at Rutgers University in New Brunswick, N.J.
“It needs to be individually adaptive; I can do this over and over again, but I’ll do better if I start with a weight, and I need a weight that’s right for me,” she added, “and you need to give timely rewards based on the child’s response.”
‘Fire and Wire’
“The neurons that fire together wire together, but only if they get reinforced by rewards,” Ms. Tallal said at a presentation earlier this year before the American Educational Research Association.
That’s how most working-memory training programs operate. A student plays one or a battery of concentration games that adapt to each response and become increasingly difficult. The student might have to remember the position of a shape on a grid, coupled with the color or type of shape, and recall that shape-type or position after seeing several other grids in a row—an exercise commonly called an “N-back” test. He might have to remember words flashed on the screen at the same time numbers are heard through headphones or recall the direction of a series of arrows that make a pattern.
Researchers are pretty much in agreement on the brain-as-muscle metaphor, and many studies have confirmed that if someone practices a task, be it a cabbie driving the streets of London or a student in Chicago recalling the direction of arrows, that person will improve in that task and build the neural pathways associated with doing it. But researchers don’t agree on whether doing a specific exercise will build a student’s ability to do other tasks or enhance general brain capacity—in the same way that climbing stairs both builds leg muscles and strengthens the heart.
“No one doubts if you train yourself on a task, you get better at that task. The real test is … does it increase your brain power generally?” Mr. Cowan said. “I’m almost convinced myself it doesn’t generally, but that’s the controversy.”
Torkel Klingberg, a professor of cognitive neuroscience at the Stockholm Brain Institute in Sweden, is among those who believe training in working-memory tasks can beef up the brain as a whole. Mr. Klingberg developed the exercises used by Cogmed and said studies of the program have shown a rigorous regimen of the mental exercises—30 minutes a day, five times a week, for at least five weeks—has “robust effects” on improving working memory. “We know working memory is correlated to many other aspects such as academic performance, so we have indirect evidence.” For example, he said, “training of working memory improves the ability to remember instructions; it would be very easy to see that would be beneficial in everyday life.”
Charles Hulme, a psychology professor at University College London, in the United Kingdom, is a skeptic, calling benefits of working-memory training “a lot of hype.”
Mr. Hulme, a co-author of the meta-analysis in Developmental Psychology, examined 23 studies of Cogmed, JungleMemory, and other working-memory training programs on groups, both typical and with cognitive and attention disabilities. He and special education researcher Monica Melby-Lervåg of the University of Oslo, Norway, found that the programs improved visual and verbal working memory in the short-term, but nine months later, there was no evidence for better verbal memory and only limited evidence that the improvements to visual-spatial memory remained. There was no evidence that any of these skills led to better attention, word decoding, or arithmetic skills.
“The sorts of tasks that are assessed here are probably pushing up against some clearly identified limitations on human performance,” Mr. Hulme said. “The idea that you could train someone for eight to 12 weeks and have them improve memory and performance is unlikely to say the least.”
Mr. Klingberg called that analysis “fairly crude,” noting, as was pointed out in the study, that it grouped together children, adults, and those of various cognitive abilities, so that a beneficial effect for one group of learners might be canceled out by another.
Moreover, Mr. Klingberg argued that a single intervention should not be expected to sustain a permanent brain change if it is not continued.
“If you did a reading intervention, you would not expect the effect to last forever, but it often leads to the child reading more, and that sustains the change,” he said.
“Similarly, I would expect for working-memory training—and what we hear though we cannot measure—is children who take working-memory training, they pay attention more in class, do more homework, they do these cognitively demanding tasks, and that can set up a feedback loop.”
Mr. Redick said that his team researchers found improvements in the adaptive N-back tasks he used in the randomized trial, but after 20 training sessions, participating college-age students on three campuses performed no better on 17 cognitive abilities tested than a control group. However, those who took the training did report that they felt as though their memory and cognitive skills had improved, Mr. Redick said.
A version of this article appeared in the June 13, 2012 edition of Education Week as Brain Training Draws Questions About Benefits