The Skeptics Society & Skeptic magazine

In our health-conscious culture permeated by people eating kale, meditating, and working out, it seems tempting to regard the brain as just another muscle—one whose relevant parts can be “exercised” to keep them from getting flabby and plump. In this article, Dr. Carol Tavris examines the evidence to see if working memory training programs really work.

I hate writing this issue’s column. Sometimes it is really annoying to have to review information you don’t want to hear, but it’s my moral and scientific obligation to do it, so you don’t have to. Like most of humanity over the age of 37,¹ I have been awaiting the scientific discoveries that will improve my memory. In my lifetime, I have seen enough of these to realize that memory-improvement advice generally depends on whatever model of memory is current at the time. When people erroneously believed that memories were “filed away,” the solution was to improve our card-catalog system. When people erroneously believed that memories were recorded as if on tapes, the solution was to find a way to rewind them (and remove the dust and scratches). When people erroneously believed that memories were “buried,” like pirate treasure or potatoes, the solution was to find ways to “uproot” them. Some of these memory-improvement methods were benign; others, like “recovered memory therapy,” hypnosis, and truth serum, had malevolent consequences that destroyed many lives and families. I am sympathetic to the desire to improve our muddy memories and restore forgotten ones. We are our memories.

Today, in our health-conscious culture permeated by people eating kale, meditating, and working out, it seems tempting to regard the brain as just another muscle, one whose relevant parts can be “exercised” to keep them from getting flabby and plump. Memory exercises and meditation to the rescue! Puzzles, games, and challenges are today’s mental weights.

In recent decades, memory scientists have revved up their study of “working memory,” one of the key mental systems responsible for storing and manipulating information. Working memory is a cognitively complex form of short-term memory—that little bin where we store a new phone number until we use it once or enter it in our contacts list. Working memory is where, well, the work gets done: it involves active mental processes that control retrieval of information from long-term memory and interpret that information appropriately for a given task. No wonder that people who do well on tests of working memory tend to do well on intelligence tests and on tasks requiring complex cognition and the control of attention, such as understanding what they read, following directions, taking notes, learning new words, estimating how much time has elapsed, and many other real-life tasks. When they are engrossed in challenging activities that require concentration and effort, they stay on task longer, and their minds are less likely than other people’s to wander.

My own working memory is taking more naps these days, and, when awakened, it tends to go wandering off into the brush. I am, therefore, an obvious candidate for the working memory training programs available. Each of them claims to be supported by ample empirical evidence of its beneficial effects, but I try to live by my own patented skeptical formula: $$(TI) + (WT/RP) = MDD, where TI is “Time Invested,” “WT” is Wishful Thinking, “RP” is Real Problem, and MDD is Money Down Drain.

Working memory training originated in 1999, when a cognitive neuroscientist named Torkel Klingberg created a computer training program designed to improve working memory in children with ADHD. By 2001, Cogmed was officially launched, and initial studies of children with attention deficit problems were promising. Naturally, that early success led to the hope that it would help people with other impairments of working memory—everything from mild learning difficulties to drinking problems to strokes and other forms of brain damage. Today healthy people use it in hopes of improving their creativity, reasoning, intelligence, everyday lapses of attention, and recall of what the hell that movie was we saw last week. Cogmed begat Mindsparke, Lumosity, and Jungle Memory, which allege that they improve IQ scores and school grades. Cogmed claims that its training

improves attention, concentration, focus, impulse control, social skills, and complex reasoning skills by substantially and lastingly improving working memory capacity. The goal is improved performance and attentional stamina. Obviously, the results are what really matter.

You can guess what’s coming, can’t you? Monica Melby-Lervåg, who is in the Department of Special Needs Education at the University of Oslo, and her colleagues Thomas S. Redick and Charles Hulme recently published a meta-analysis of working memory training studies: 87 publications with 145 experimental comparisons.² All studies had a pretest-posttest design and a control group, and were designed to see whether working memory training transferred to other measures (nonverbal ability, verbal ability, word decoding, reading comprehension, or arithmetic). Typically, right after training, people reliably improved on measures of “intermediate transfer,” that is, tests of doing what they had just learned to do. (Practice doing puzzles and you get better at doing puzzles.) For measures of “far transfer” (nonverbal ability, verbal ability, word decoding, reading comprehension, and arithmetic), the researchers found “no convincing evidence of any reliable improvements when working memory training was compared with a treated control condition.” They conclude: “working memory training programs appear to produce short-term, specific training effects that do not generalize to measures of ‘real-world’ cognitive skills.”

There is no good evidence that working memory training improves intelligence test scores or other measures of ‘real-world’ cognitive skills.

At the Georgia Institute of Technology, the eminent memory scientist Randall W. Engel and his lab have been finding the same thing.³ In their own meta-analytic review, they concluded that “the claims made by Cogmed are largely unsubstantiated…there is no good evidence that working memory training improves intelligence test scores or other measures of ‘real-world’ cognitive skills.” They added:

Cogmed working memory training is sold as a tool for improving cognitive abilities, such as attention and reasoning…. The only unequivocal statement that can be made is that Cogmed will improve performance on tasks that resemble Cogmed training. However, for people seeking increased intelligence, improved focus and attentional control, or relief from ADHD, current research suggests that this training program does not provide the desired result.

This is not an innocuous issue. No one cares and you have nothing to lose if you play Scrabble morning, noon, and night. But a person taking the full Cogmed program will pay up to $1,500, and when educators and clinicians invest in it lavishly in the belief that it will help troubled populations—notably underperforming or inattentive students— costs become a concern. That is why, Engel and his colleagues write with calm understatement, “The efficacy of Cogmed is not simply a scientific curiosity. Cogmed is a product that is actively marketed to school systems and to the parents of children with developmental disabilities. Due to the cost to tax payers and consumers, it is our opinion that demonstrating the validity of this program is of the utmost importance.”

Finally, Thomas Redick took a close look at five studies that claimed to have demonstrated benefits of working-memory training.⁴ Most had tiny samples (between 7 and 17 people in both training and control groups); where benefits occurred right after the intervention, they were gone in a few months; and that any apparent benefits of training were actually a statistical artifact. These findings were extended and confirmed in a major 2016 meta-analysis of studies of “brain-training programs” that specifically focused on the use of cognitive tasks or games as a means of enhancing performance on other tasks —which, after all, is what the test promoters claim they can do. But don’t.⁵

As long as I’m reporting unwelcome news, here’s another tidbit: mindfulness meditation, shown to have indisputably good effects on mind and body, has an unexpected down side: it increases susceptibility to false memories. In three experiments, Brent Wilson and his colleagues randomly assigned participants to either a mindfulness induction, in which they were instructed to focus attention on their breathing, or a mind-wandering induction, in which they were instructed to think about whatever came to mind.⁶ Then they took a standard memory test in which participants are shown a list of words, and later asked either to recall them (by writing down as many as they could remember) or recognize those they had seen among a larger list. Each memory test contains a word (the critical item) that is closely related to the words on the list but is not actually on it. For example, the list garbage, waste, can, refuse, sewage, junk, rubbish, sweep, scraps, pile, dump, and litter can activate the critical item trash. The overall number of correctly-recalled words did not differ between conditions, but people in the mindfulness condition were significantly more likely to report having seen trash (a false memory) than participants in the control condition. “Because mindfulness meditation encourages judgment-free thoughts and feelings,” the researchers suggest, it seems to suppress the usual mechanism of reality monitoring by which we decide whether a memory is true or false.

The moral? Mindfulness is good for focusing on the here and now, but maybe not so much for remembering the there and then.

That’s enough on this subject for now. I have a crossword to finish.