clock menu more-arrow no yes

Brain training games don’t boost IQ. Here’s what does.

An experiment boosted IQ by 5 points in just one hour. It’s bullshit — and that’s the point.  

Thomas Reichhart / Shutterstock

Wouldn’t it be nice if we could all become a little smarter?

The world would be better off for it. Smarter people live longer and enjoy greater well-being across their lifetimes, psychologists have found. Intelligence gives people an edge in subtle ways. High intelligence is even thought to protect people from automobile crashes.

That’s why brain training games — computer-based memory puzzles — have ballooned into a billion-dollar industry in recent years. The idea is these memory games could help boost what’s known as fluid intelligence, which is the ability to reason (as opposed to crystallized, accumulated knowledge).

But here is the sad truth: Brain training games likely don’t work. In January, the Federal Trade Commission called out Lumos Labs, the makers of a popular brain-training app, for deceptive advertising, and hit them with a $2 million fine. "For years, researchers have looked into brain games and found that they simply don't have the real-world benefits they purport to," Vox's Julia Belluz explained at the time.

And now, there's even more evidence. A new paper in the Proceedings of the National Academy of Sciences demonstrates how the evidence ​in favor of brain training is likely just a placebo

How expectations compromise brain training research

There are some studies — and even meta-analyses — that have found evidence brain training can boost fluid intelligence. But the authors of this new PNAS paper, a group of psychologists at George Mason University, were extremely skeptical of those conclusions. They guessed the papers were flawed for a very simple reason: expectations.

Here’s their hypothesis: When people know they’re joining a "brain training" study, they perform better on intelligence tests — because they expect the training to work.

And this is often how these studies are run: The authors highlight one meta-review where 17 of 19 studies mentioned "brain" or "cognitive" training in their recruitment materials. These words alone bias the whole test, they claim.

To test this idea, the George Mason psychologists designed a study and sent out two fliers: one recruiting participants for a "brain training and cognitive enhancement" study, and the other simply recruiting for "a study."

People who respond to the ad on the left may also be more likely to believe that intelligence is malleable. And past research has found that a person’s beliefs about the malleability of intelligence can predict who will improve the most during brain training. Experiments that recruit with fliers like the one on the left may end up with a biased sample.

The researchers had 25 participants in each group. They then tested their fluid intelligence, had them play a memory training game for one hour, and then tested their intelligence again.

It’s important to note that there’s absolutely no reason one hour of training should make any reasonable difference when it comes to enhancing IQ. But amazingly, the placebo group’s IQ jumped, while the other group’s IQ remained flat.

The placebo group made huge gains on two different intelligence tests.

"It is extremely unlikely that individuals in the placebo group increased their IQ by 5-10 points with 1 hour of cognitive training," the authors write in discussion.

The placebo group grew "smarter" because they expected to grow smarter. (Perhaps they were more focused and concentrated on the tests, which increased their scores. It’s also evidence for how important motivation can be for cognitive performance.)

The authors suggest that if brain training research is to continue, it needs to become more rigorous, adopt "gold standard" double-blind placebo measures, and limit subjects’ expectations as much as possible.

So brain training doesn’t work to increase intelligence. What does?

Surprisingly (and maybe depressingly), little does.

Studies have found that IQ is highly influenced by genetics and is generally consistent throughout your life. While the research shows that all of our brains decline as we age, if you were a smart kid, you'll be a smart old person.

This chart shows a Scottish study where a group of 90-year-olds were given an IQ test they previously took at age 11. Even though intelligence generally declines with age, those who had high IQs as children were most likely to retain their smarts as very old people.

Intelligence: All that Matters

There are some promising ideas, however, about IQ enhancement that are worthy of considering.

Education: The longer time a person spends in school, the more likely they’ll see a boost in IQ. (Twelve years of mandatory education is more intense and rigorous mind training experience than a couple of hours on a video game.) Analyses have found evidence that each additional year in school adds 3.7 IQ points.

Nutrition and environment: Good nutrition might not boost IQ, but it’s believed that poor nutrition and toxins (lead, in particular) in a child’s environment does decrease it. Iodine deficiencies during development are believed to decrease IQ by around 10 points.

Time: The world is getting smarter every year. Really. Mean IQ scores appear to be increasing between 2 and 3 points per decade. It’s called the Flynn effect, and it is likely the result of increasing quality of childhood nutrition, health care, and education.

What Is Intelligence?: Beyond the Flynn

So progress can be made on intelligence — it’s just unlikely to happen on an individual level, in a big way.