“You may be the smartest person in the world and you don’t even know it. If you’re not the most intelligent person in the world yet, then, welcome to your gym!” “The more you train with our brain exercises, the smarter you can be.” “Designed by neuroscientists and game experts, it exercises your brain while you have fun at the same time.”

The statements above are part of actual sales pitches from websites advertising online brain training, which consists of computer games that supposedly boost players’ cognitive abilities. Programs like Lumosity, NeuroNation, and BrainHQ claim to provide a fun workout for the brain backed by neuroscience, even citing published research studies using their games as proof.

But some scientists are pushing back against these claims, citing false advertising and exaggerated promises. In 2014, seventy of the world’s leading cognitive psychologists and neuroscientists signed a statement challenging the promotional tactics of brain training companies. Overall, they note a lack of sufficient evidence to say that games of this nature improve cognitive performance in any meaningful way.

“If with 15 to 20 hours of training, you can actually make someone smarter, that’s a really big deal,” said Cyrus Foroughi, a researcher of human factors and applied cognition in the Department of Psychology at George Mason University. “Personally, I wasn’t convinced that the methods being employed in brain training studies were sufficient enough to really sell the results.”

Earlier this year, Foroughi and his colleagues released results suggesting that fluid intelligence gains seen in cognitive-training studies could be originating from a simple placebo effect. They recruited subjects for their experiment using two different flyers: one overtly suggested cognitive enhancement through brain training, while the other used generic language to get people to participate in an unnamed study. The idea was to self-select those that already had interest in brain training and getting smarter in order to tease out a possible placebo effect.

Individuals in the placebo group improved their performance after a single, 1-hour session of a working memory task that equates to a 5- to 10-point improvement on a standard IQ test. On the other hand, those in the control group showed no significant change in test scores. Foroughi purposely chose to administer only one hour of cognitive training because that amount of time isn’t long enough to have any effect.

“We don’t know the underlying mechanism of why people improved, but probably the best explanation is those individuals who self-selected either put more effort in or believed in themselves more,” said Foroughi. “They thought, ‘I’m supposed to be smarter now.’ We don’t have the evidence to support this idea, but it’s the simplest explanation.”

The study, published in the Proceedings of the National Academy of Sciences in July, does not necessarily disprove previous cognitive training research. However, out of 19 studies included in a recent meta-analysis, 17 used overt recruitment methods that could have introduced a self-selection bias.

In addition, the U.S. federal government is putting its foot down on the way cognitive training companies advertise their products. In January, the Federal Trade Commission charged the creators and marketers of popular brain game program Lumosity with consumer deception, and the company paid $2 million as part of the settlement.

“Someone asked me once what I would tell my grandma about cognitive training, and my answer was simple: Keep your mind and body active in the ways you enjoy the most,” Foroughi said. “If you like playing Lumosity’s games, then great. Or if you prefer Sudoku, crossword puzzles — just do what you enjoy, and you’ll probably see the most benefits out of it.”

Keep up to date with the BOLD newsletter