In Phaedrus, Plato's 2,400-year-old set of dialogues, Socrates narrates a conversation between the King of Egypt, Thamus, and a god, Theuth. Theuth had invented various fields of learning, including arithmetic, astronomy, geometry, and writing, and he wanted the king to share them with his people. While each discovery had its pros and cons, writing, Theuth said, was the best invention of all, since it would “improve both the wisdom and the memory of the Egyptians.” The king unconvinced, described writing as the “conceit of wisdom,” arguing that it would cause men to become forgetful, because they would rely upon written texts rather than remember things for themselves. The god's discovery wasn’t a tool for memory, said the king. It would only enable reminiscence.

Almost two-and-a-half millennia later, our impulse to remember things, and to do so efficiently, remains fierce. The tools we use include grocery checklists, photo albums, flash cards, smartphone memos, and even scrawls on Post-Its. Over the past decade, digital brain-training games have emerged as the newest way to sharpen memory skills. They’re often touted as having a wide range of benefits, from helping people remember names and childhood stories to possibly staving off dementia and Alzheimer's disease.

In October, Stanford University’s Center for Longevity and the Max Planck Institute for Human Development in Berlin asked a group of more than 70 neuroscientists, cognitive psychologists, and academics to share their views on these games. “There is little evidence that playing brain games improves underlying broad cognitive abilities, or that it enables one to better navigate a complex realm of everyday life,” the group wrote in a consensus report.

The validity of brain games has been a point of debate ever since the first iterations of them launched in the early 2000s. What’s notable about the Stanford/Max Planck report is its deliberation and scope, offering insight collected over the course of a year from experts based at 40 institutions in six different countries. (Disclaimer: I'm a research engineer in a physics laboratory based at Stanford University. However, my work is unrelated to the subject matter in the report.)

The report questions the “pernicious,” “exaggerated,” and “misleading” claims made by brain-game manufacturers and stresses that wide-reaching positive results are “elusive.” The signatories don’t call out any specific types of brain-training software, but their language is often damning. “Many scientists cringe at exuberant advertisements claiming improvements in the speed and efficiency of cognitive processing and dramatic gains in ‘intelligence,’” they write.

Brain training is projected to be a $3 billion industry by 2015. Its target demographic is vast, from kids lagging behind in school to seniors who’d love to be able to remember things more efficiently. Most people, in fact, would appreciate the opportunity to feel like they’re getting the most out of their brains, and when a particular product is advertised as being both educational and backed by science, it can be hard to resist.

That there’s a need for scientists to issue a signed letter like this one only goes to show how popular brain games are and to what extent experts feel the public may have been seduced by the promise that they’re endorsed by science. The letter is both a word of caution to the public and the signatories’ way of distancing themselves from the commercial products. It's one thing to advertise Kool-Aid, the letter seems to say. But it's quite another to say people should drink it because science says so.

* * *

The brain's plasticity has been a matter of discussion in scientific circles for more than a century. In 1890, the American doctor William James discussed the “extraordinary degree of plasticity” in nerve tissue in his book, The Principles of Psychology. James wrote that “the phenomena of habit in living beings are due to the plasticity,” and concluded that this shaping of the brain's neurons is the foundational mechanism for growing and learning. His work caught the eye of Edward Lee Thorndike. Over the course of the next 50 years, Thorndike became one of the most renowned educational psychologists, known for his experimental approaches to the field.

One of the things Thorndike and Robert Woodworth, his collaborator, are best known for is their investigation into how much one form of learning can affect another (say, how studying Latin helps students with other subjects). This phenomenon is broadly called “transfer of learning.” Thorndike found that the only way to have any similar transfer between education and everyday acuity was if the subject being studied related to normal life in a number of ways. Barring that, “improvement in any single mental function rarely brings about equal improvement in any other function, no matter how similar,” he wrote in 1901. There is such a thing as transfer, but it's more likely to be “near transfer,” when subjects are similar, or related. Broadly speaking, learning was, and is, specific to its field.

The brain is a massive, malleable silo. It is at once the simplest of things, and the most complex of things. Within an instant, we can recall habits (making coffee), facts (Sacramento is California's state capital), and episodes (a first bicycle ride without training wheels). We can remember things that happened yesterday (the first snow of the season) and things that happened twenty years ago (graduation). The brain has an infrastructure of about 100 billion neurons that connect and send signals. Proteins in the neurons help open the gates to let the ions drift in and out across about a quadrillion different connections, called synapses.

We can also divide memory up into categories of the kinds of things we remember. Sensory memory is the ultra-short memory that lasts for fractions of a second after we see, taste, hear, smell, or touch something. Long-term memory is at the other end of the spectrum, the kind of deep storage that we might tap into at a family reunion. It’s often the complex product of repeated strong signal patterns and new proteins, establishing a separate corridor for lasting memory to travel through safely. Short-term memory is in between: It’s a rearrangement of the proteins, gates, and ions that builds a signal that might last between seconds and minutes. It's the kind that helps us get through this sentence, or the kind that we might use when we repeat a string of unconnected words after seeing or hearing them.

Another way of looking at short-term memory is to see how we use it. This is called working memory, and it's often compared to a scratchpad. Working memory is what swings into action when we do mental math or carry out the steps of a cooking recipe. It’s this working memory that has been linked to various cognitive disorders including dyslexia, ADHD, Alzheimer's disease, language difficulties, and dementia. Strengthen working memory with individually designed brain-training games, the claims suggest, and possibly see benefits in this vast array of disorders.  

A 2008 paper by researchers at the University of Michigan and University of Bern did suggest that by training the working memory, people could improve their fluid intelligence—the ability to solve new problems and see patterns or complex relations (something that has typically been viewed as fixed). “Instead of regarding [fluid intelligence] as an immutable trait,” it said, “our data provide evidence that, with appropriate training, there is potential to improve it.” The paper was widely touted and various brain-training companies adopted its methods. It also prompted other researchers to probe the study.

One problem noted was that the paper measured intelligence using a single test and with weak control factors. So, in 2012, eight researchers from four universities worked with the authors of the 2008 paper and replicated the study, with two key differences. They gauged the “transfer” effects of the training by testing participants’ intelligence using 17 different methods. They also set up control groups—one to test the placebo effect by using a dummy visual search training game, and another that offered no training. Their results showed that while the participants improved their scores in the brain-training exercises, they showed no gains in the general cognitive ability tests compared to the control groups.

* * *

With an increasing number of brain-training tests on the market, the 73 researchers stated in their October letter that the field would benefit from having validity standards set by the American Psychological Association or a similar organization. Repeated studies would help, too, since they could poke deeper at the weaknesses of each past study, and over time, the collective results could further the findings. In 2013, two researchers at the University of Oslo and University College London conducted a meta-analysis of 23 different brain-training studies. What they saw was what every skeptic has seemed to notice. “Memory training programs appear to produce short-term, specific training effects that do not generalize,” they wrote. That is, training with games helped people get better at the games they were playing, but not much else.

Brain games aren’t a complete waste of time. They provide mental stimulation, the kind that doctors advise the elderly to get from crosswords and other mind games. They are fun, engaging, even competitive. They show how we can train the brain to get better at a task with repeated practice. But these are fairly obvious results given what we know about the brain's plasticity. The larger question (and promise) is whether playing on a phone for 15 minutes each day can help us in other aspects of our lives that we haven't been training for.

We have limited time in a day, and how we parcel out the hours often depends on the rewards we expect for the things we choose to do. That’s why it's important to identify what might be lost given the time spent playing a game whose benefits might be questionable. As the researchers write, “If an hour spent doing solo software drills is an hour not spent hiking, learning Italian, making a new recipe, or playing with your grandchildren, it may not be worth it.”

All that said, the things that do help shape a healthy brain are the things that have been tried and tested for years. Physical fitness forces more blood to flow into the brain, allowing for more neural connections. So exercise works, as does conventional training in reading and language skills for children with reading-comprehension and oral language difficulties.

Also helpful are curiosity and engagement with the world around us, and the body’s lifelong subconscious effort to keep the brain active. The report's summary is perhaps the biggest indictment of any pop-a-pill brain-game philosophy. “The promise of a magic bullet detracts from the best evidence to date,” they wrote. “Cognitive health in old age reflects the long-term effects of healthy, engaged lifestyles.”