Do brain games actually work? An honest scorecard

The argument that refuses to end

Every couple of years the same question flares back up in the press: do brain games actually do anything? The answer you hear depends entirely on who is talking — and unusually for a science story, some of the loudest voices on both sides are credentialled researchers who have read the same studies and reached opposite conclusions. So rather than pick a side, let's referee the fight.

2014: two open letters, one year, opposite verdicts

The modern debate has a clear flashpoint. In 2014, the Stanford Center on Longevity and the Berlin Max Planck Institute for Human Development published a consensus statement signed by around 70 scientists. Its conclusion was blunt: claims promoting brain games are "frequently exaggerated and at times misleading," and there was no compelling evidence they hold off cognitive decline.

The pushback was almost as loud. A separate group of researchers — more than a hundred of them, gathered under the banner of cognitivetrainingdata.org — answered with their own open letter, arguing the sceptics had understated a real and growing body of positive evidence. Two large groups of serious scientists, same field, same year, flatly contradicting each other. That, more than any single study, is the honest state of the field: genuinely contested.

The referee's report

The most thorough attempt to settle it arrived in 2016, when Simons and colleagues published an exhaustive review in Psychological Science in the Public Interest. They read the literature closely and graded it against best-practice standards. Their verdict, boiled down:

• Brain-training programs reliably improve the specific tasks you train on.
• There is less evidence that they improve closely related tasks.
• And very little evidence that they improve distantly related tasks or everyday cognitive performance.

They also flagged that many of the supporting studies had design weaknesses — missing active control groups, small samples — that make strong claims premature. In plainer terms: near transfer is real, far transfer is thin, and the marketing had been sprinting well ahead of the science.

What happens when marketing laps the data

You don't have to wonder where the over-claiming leads. In January 2016, the US Federal Trade Commission announced that Lumosity would pay $2 million to settle charges of deceptive advertising. The complaint was specific: Lumosity had claimed its games improved performance at work and school and could delay age-related decline and protect against dementia and Alzheimer's — without the science to support it. (The headline figure was $2 million; the full judgment was $50 million, largely suspended because the company couldn't pay.)

That settlement is the cautionary tale the whole category now lives with — and, frankly, a big part of why this site exists.

Both camps are partly right

Here's the un-dramatic truth the shouting tends to bury. The sceptics are right that broad, life-changing, medical-sounding claims outrun the evidence badly. But the hard-line "it's all snake oil" take overshoots too. You do reliably get better at the skills you practise. A handful of specific, focused programs have shown specific transfer. And engagement, enjoyment and a measurable sense of progress aren't nothing — they're just not the same thing as a cure.

The honest scorecard

If we tally only what's actually defensible:

• Settled: you get better at what you practise. Play a reaction test, your reaction score improves. Real and useful.
• Genuinely unclear: whether that carries to closely related tasks. Sometimes, modestly, in some programs. Don't bank on it.
• Not supported: broad IQ gains, a transformed everyday memory, or protection against dementia. The evidence simply isn't there.

Where we stand

CortexCrunch is built to live entirely on the settled side of that scorecard. We treat our games as exactly two things: measurable practice for specific skills, and something genuinely fun to do. We show you honest numbers — real milliseconds, real digit spans, percentiles clearly labelled as estimates — and we make zero medical claims. No "reverse cognitive decline." No "unlock your brain's hidden potential." If you want our full, cited position, it lives on the science page.

So — do brain games work? Yes, at the one thing they can honestly promise: making you a little sharper at the game in front of you, and handing you a number to beat tomorrow. Everything past that, we'll let the evidence decide — out loud, and without the BS.