This weekend, I came across this piece on ScienceDaily. Like any good press release, the message is one that rather invites you to click: speaking several languages could keep your brain up to thirteen years younger. This sounds promising, especially if you speak several languages. The newspaper articles almost write themselves, but before you download DuoLingo… is writing them a good idea?

Naturally, I wanted to read the study, and that is where I immediately ran into the first problem. I couldn’t find this study anywhere. At the moment, there is neither a paper nor a preprint available online. All we have is this press release about research presented at the FENS Forum 2026, a large neuroscience conference, this summer.

Don’t get me wrong. That does not suddenly make the research bad. But it struck me as a good opportunity for a small exercise: how do you handle a scientific press release when you cannot read the actual study?

1. Look for the study

Let’s start with the obvious, and with the step I had already taken. A press release is not a study. It is communication about a study, usually written for a very different purpose than a scientific paper and not necessarily written by the researchers themselves.

In this case, the study has not yet been published, and there is no preprint either. For now, we have to make do with what the researchers presented at the conference and the subsequent communications about it.

This immediately means that there are questions I cannot answer. That is worth keeping in mind for the rest of this story.

2. Check where the story comes from

ScienceDaily is doing here what ScienceDaily often does: publishing science news that originally comes from universities, research institutes or scientific organisations.

The original public source turned out to be communication around FENS Forum. That is useful to know. If you then find five websites reporting roughly the same thing, you have not necessarily found five independent confirmations. They may simply be five versions of the same press release or stories based on it. Several reports of the same study do not necessarily strengthen the evidence if they can all be traced back to a single source.

3. What did the researchers actually measure?

If you still want to continue with the press release without having access to the study, ask yourself what the researchers actually measured. Fortunately, we do get an answer here.

They used magnetoencephalography, or MEG. This allows researchers to measure brain activity and, among other things, examine patterns in the functional connections between different parts of the brain. (Yes, I had to look this up too.)

First, they used data from 728 people to build what is known as a brain ageing clock using machine learning. Put very simply, an algorithm learned which age typically goes with particular patterns in the brain. This was the basis.

In a second step, the model was applied to another group of 144 people from the Basque Country who spoke between one and four languages. This produced a striking result: compared with monolinguals, the estimated brain age of people who spoke two languages was about 6 years younger. For those who spoke three languages, it was around seven years younger, and for those who spoke four, as much as thirteen years.

But… did you notice what subtly happened along the way?

“Your brain is thirteen years younger” does not mean the researchers measured the biological age of people’s brains and found them to be literally thirteen years younger. It means that a statistical model based on patterns in brain activity predicted a younger age.

To be clear, that is still more than interesting. But it is not quite the same thing.

4. Look at the study design

There is another important shift in the press release. The headline rather quickly gives the impression that speaking several languages keeps your brain young. But to make that claim, you would really want to show that one causes the other.

Yes, the old correlation-versus-causation problem is back. And you can probably guess what comes next? That was not established here.

The researchers did not randomly assign the participants to groups that had to learn one, two, three or four languages. Nor did they follow them for years while some learned additional languages and others did not. In this study, the researchers compared people who already differed in their language experience.

In other words, this is observational research.

You can conclude that multilingualism is associated with a younger estimated brain age. It is much harder to say that multilingualism causes a younger brain age.

5. Who are those 144 people?

By now, we have actually come quite a long way in taking the story apart, but this is where I really started to miss the scientific paper.

Having 728 participants to train a model sounds fairly impressive. But the interesting comparison between people speaking one, two, three, and four languages was ultimately based on 144 people.

At that point, you want to know how many people were in each group. How many of those 144 participants actually spoke four languages? Also, how wide were the confidence intervals around that rather spectacular thirteen-year figure? How robust was the result across different analyses?  The answers may well be reassuring, but we simply don’t know.

Of course, you could start guessing. Don’t. If you don’t know something, don’t fill in the blanks yourself.

6. Could there be other explanations?

Back to something positive that we can learn from the press release. The researchers took factors including age, sex and education into account. But people who speak several languages can, of course, differ from people who speak only one language in many other ways.

Let’s consider a few possibilities. They may have different kinds of jobs. They may travel more. They may have different social networks. They may differ in other forms of cognitive or social activity. Perhaps people who remain cognitively stronger are also better able to learn several languages and continue using them.

That last possibility even reverses the reasoning suggested by the press release.

None of this means that multilingualism cannot have an effect. It simply means that this study cannot automatically rule out these alternative explanations.

And, to be fair, the researchers themselves acknowledge in the press release that factors such as lifestyle and social engagement may play a role.

Which brings us to the final tip.

7. Read the cautious sentences too

When reading scientific press releases, it is usually worth looking beyond the headline and opening paragraph and searching for words such as may, could, association, limitations, and further research.

From that point onwards, you often find a rather different study from the one promised by the headline. This is not because the university press office necessarily wanted to mislead you. They were doing their job: a press release needs to attract attention. Researchers, meanwhile, need to formulate their conclusions carefully. Tension can arise between those two aims.

And something curious can happen.

The story starts with the researchers stating that:

A statistical model estimates a lower brain age based on brain activity in multilingual participants.

And after a few steps, people read this message all over the world:

Speaking several languages keeps your brain up to thirteen years younger.

Finally, is this a bad study?

I genuinely have no idea. More than that, I find the result interesting. According to the press release, there even appears to be a kind of dose-response relationship. In this case, speaking more languages was associated with a younger estimated brain age. Earlier acquisition of a second language and greater language proficiency were also associated with it. These are hardly unimportant topics in education.

While searching, I also discovered that the same research group had previously published a study in which multilingualism was associated with less accelerated ageing. The new research now approaches the question using a very different measure. There is therefore plenty of reason to be curious about the eventual paper, if it makes it through peer review.

So I didn’t write this post to tear the study apart. It simply provided a nice opportunity to make a broader point: a scientific press release does not have to be wrong to put you on the wrong track.