How a Memory Game Can Multiply Your Brain’s Computing Power

By Guest Blogger | July 19, 2013 2:05 pm

by Ben Thomas

n-back memory game

When’s the last time you forgot your cell phone? What about your anniversary? We’ve all wished for a better memory at some point. And in those moments, what we’re generally referring to are our fact-based memory systems—the ones that store experiences and learned knowledge, the ones that can be strengthened by flash cards and mnemonic devices.

But there’s another type of memory that’s equally important, though less talked about: Working memory. Working memory holds items in the current moment, like digits in a phone number or lyrics to a song. It’s the mental capability that lets you keep multiple things in mind—say, competing goals for a project at work—and navigate a solution.

And it turns out working memory is fundamental to some of our most important cognitive abilities. People with a spacious working memory are, on average, better test-takers and better students, and they have more executive control and better problem-solving skills. A strong working memory is even correlated with greater lifetime earning potential.

Until the late 1990s, many scientists thought working memory was a static quality like IQ. But in recent years, neuroscience has revealed that this core aspect of cognition can actually be bolstered with behavioral training—and the means might be as simple as an exercise in counting cards.

Measuring memory capacity

The idea of working memory got its scientific pedigree in 1956, when Princeton psychologist George Miller demonstrated that a person’s immediate memory capacity was about seven items, plus or minus two. Miller’s paper became one of the top-cited publications in psychology history, and it took decades before anyone seriously challenged his “magical number seven.”

Those challenges began to arrive in the 1990s, when researchers showed that a person’s working memory capacity depends heavily on the type of items being remembered, and on whether those items can be sub-grouped into related “chunks,” such as scenes in a story. Then in 2004, cognitive neuroscience professor Torkel Klingberg combined a battery of psychological tests with brain-scanning technology to demonstrate that individuals can increase their working memory capacity with training. What’s more, this training was correlated with increased neural activity in frontal and parietal brain areas, which play central roles in complex cognition and attention.

Klingberg’s work opened the floodgates. Researchers soon discovered correlations between working memory training and the integrity of a person’s white matter—the connective tissue that speeds neural signals from one processing center to another. Others found that similar training could increase the density of information-processing gray matter. Still other research reported that training increased the number of dopamine receptor sites in frontal and parietal brain areas—a change that could support enhanced attention and sharper decision-making, as many studies have found close linkages between frontoparietal dopamine activity and sustained attention.

How memory expands

Still, until recently, no one knew exactly what a human brain actually does, in functional terms, as it’s expanding its working memory capacity. But Bornali Kundu, a neuroscientist at the University of Wisconsin, Madison, has found a new piece of that puzzle.

Kundu’s work relied on a test called an “n-back” test. This is a common psychological test that asks people to identify if an object is one they’ve seen before. For a one-back test, say, of a series of colored dots, you would press the button if you saw two reds in a row.

The game gets exponentially more challenging when there are intervening objects, however. On a two-back test, you push the button if you saw the red dot exactly two screens ago. On a four-back test, your brain has to store five items at any one time.

Using a free piece of brain-training game software called Brain Workshop, Kundu and her research team set up precisely these kinds of tests. Players watched patterns of colored dots flash on a computer screen one-by-one and, depending on their skill level, played a two-back or four-back version. A few even made it to a five-back level. Not exactly high-stakes blackjack, but still challenging enough to keep players engaged.

As the players trained on their n-back tasks for an average of one hour a day, five days a week for five weeks, Kundu and her team monitored their brain activity with a hybrid technique called TMS/EEG. Researchers were specifically looking for brain areas that activated together on very short time scales. They wanted to analyze not only which areas of the players’ brains were responding to the tasks, but also how those areas were cooperating with one another.

Memory and more

As Kundu expected, the participants who trained on n-back tasks displayed measurably improved visual working memory capacity as the study neared its close. The control group, who trained by playing Tetris rather than n-back tasks, displayed increased attention but unimproved working memory capacity—at the end of the trial they scored higher at Tetris but not on any n-back tasks. But participants who’d trained on n-back tasks could remember, on average, one more visual item than when they’d started.

That alone might be underwhelming. However the effects extended far beyond memory: The trainees’ scores improved on a variety of different attention and focus tasks. For instance, their Tetris scores improved along with controls’, due to the fact that visual working memory training strengthens many of the same neural circuits that our brains use to maintain visual attention. What’s more, the working-memory trainees’ visual reaction times also sped up, and these players scored higher on the n-back tests than they had before their training. These improvements indicated that they weren’t just improving their skills at a single test but rather were more generally training their working memory to hold more items at a time—with the side benefit of increased focus in the face of noise.

And most interestingly, the experiments established for the first time a clear causal link between certain brain connectivity changes and working memory expansion. After training on the n-back test for five weeks, the subjects showed increased connectivity between frontal and parietal brain areas compared to controls; indicating that working memory training was the cause of these connectivity changes. “We’re showing that not only do these frontal and parietal brain areas light up in response to working memory tasks,” Kundu says, “but that there’s a causal influence of training that increases connectivity between these brain areas—and that’s never been shown before.”

Though this particular result is new, it supports years of neuroimaging studies that have found clear correlations between heightened frontoparietal connectivity and heightened intelligence. The implication, then, is that a simple memory game—not unlike the icebreaker games kids play to learn each others’ names on the first day of camp—can significantly change memory’s performance and the physical structure of the brain itself.

Training your memory

So how broad are the applications of Kundu’s training? That depends, she says, on the types of items you’re aiming to store in working memory. If you train your visual working memory, as  subjects did in this study, you’re likely to see improvements on other tasks that require intense visual attention, like searching for a specific photo in a folder—while training your working memory for, say, musical notes can improve your skill at recognizing a tune. The bottom line, Kundu says, is that any transfer effect will be limited to aspects of the skill that depend on the same brain circuitry you’re training.

Even so, several recent studies have reported that a high verbal working memory capacity is a better predictor of a person’s long-term success than an IQ score is. “Fluid intelligence is highly related to learning, and working memory ability correlates with real-world measures like reading comprehension, performance on standardized tests and lifetime earning potential,” Kundu says. The reason is simple: The more items you can hold in working memory at once—and the better you are at filtering distractions—the more complex your calculations can be.

Kundu’s study demonstrates that with steady practice, even the most absentminded among us can fight off the errors that a distraction or a memory lapse can produce—but her ambitions don’t stop there. Over the next few years, she hopes to use her discoveries about brain connectivity to help pinpoint some of the core neural mechanisms that enable us to learn and remember. “Studying connectivity is one of the most important things we can do right now,” she says, “because everything the brain does is supported by a whole network of areas. This is all really new stuff; it’s all really exciting.”

Ben Thomas is an author, journalist, inventor and independent researcher who studies consciousness and the brain. A lifelong lover of all things mysterious and unexplained, he weaves tales from the frontiers of science into videos, podcasts and unique multimedia events. Lots more of his work is available at

Image courtesy R.Ashrafov / Shutterstock


CATEGORIZED UNDER: Mind & Brain, Top Posts
MORE ABOUT: brain, memory, neuroscience
  • Shannon Lee-Hong Dong

    This article fails to mention a growing body of research that fails to replicate these benefits. There have been many skeptical articles on major media sites such as NYTimes, Fox News, BBC, etc, so it is hard to understand how this author could have failed to uncover it.
    Does this author have a conflict of interest of some sort–ie does Ben Thomas receive payment from any of the many companies promoting brain training software?
    Discover Magazine should follow basic journalistic ethics and require disclosure in such cases.

    • lisa741

      like Jamie replied I’m impressed that people can get paid $9883 in a few weeks on the computer. did you read this page w­w­w.K­E­P­2.c­o­m

  • antoine alam

    brain training is it relative to age


Discover's Newsletter

Sign up to get the latest science news delivered weekly right to your inbox!


See More

Collapse bottom bar