The science

Short version: you will get better at the six tasks in UTOK if you practise them. Any carry-over to the rest of life is small at best, and still disputed. UTOK claims the first thing and not the second.

Where the excitement came from

In 2008 Susanne Jaeggi and colleagues trained people on a dual n-back task, a harder version of Position Back, for up to nineteen sessions of about twenty-five minutes. They reported gains on a test of fluid reasoning that grew with the amount of training [1]. It was a small study with no active control group, and the reasoning test was given under a ten-minute time limit, but it launched a decade of research and a lot of apps.

What happened when people checked

Redick and colleagues repeated the training with an active placebo group, who did a different adaptive task for the same time, and seventeen transfer tests. They found no improvement in any of the ability measures [2]. Meta-analyses by Melby-Lervåg and Hulme, first in 2013 and again with Redick in 2016 across 87 publications, reached the same shape of answer: reliable gains on the trained task and on close variants of it, right after training, and no convincing gains on reasoning, reading, arithmetic or anything else further away once the comparison group is a fair one [3, 4].

In 2016 Daniel Simons and six colleagues went through every study the brain-training companies themselves cited. They found extensive evidence that training improves performance on the trained tasks, less evidence for closely related tasks, and little evidence that it helps with distantly related tasks or with everyday cognitive performance [5]. A 2019 review of the meta-analyses put far transfer at roughly zero once placebo effects and publication bias are accounted for [6].

The strongest result on the other side

The ACTIVE trial randomised 2,802 adults over 65 in 1998 and 1999 to memory, reasoning or speed-of-processing training: up to ten sessions of an hour or more, in person, with some people getting booster sessions later. Twenty years on, the group that did speed training with boosters was about 25% less likely to have a dementia diagnosis in their insurance records. Speed training without boosters showed nothing, and neither did the memory or reasoning groups [7].

That’s a real result. It’s also one specific task, in older adults, in one subgroup, measured as a secondary outcome. None of UTOK’s games is that task, and five minutes a day on a phone is not that dose. UTOK does not claim anything like it.

What UTOK does claim

Each game is a task that researchers use to measure one thing. Here is what they measure:

Colour Word is the Stroop task: holding back the urge to read (inhibition).
Position Back is a spatial n-back: keeping a short list in mind and updating it (working memory).
Number Grid is a Schulte table: scanning for a target (visual search). It has the thinnest research base of the six.
Tap Unless is the Sustained Attention to Response Task: staying with a boring, steady stream and catching the rare exception (sustained attention).
Block Span is the Corsi block-tapping test: remembering where things were and in what order (spatial memory).
Rule Switch is colour and shape task switching: changing rules on the fly, and what that change costs you in time (flexibility).

If you play them, your scores on them will probably go up, partly from practice and partly from getting used to the controls. UTOK shows you those numbers, your own and no one else’s, and that is the whole promise.

Why this page exists