Do Brain Training Apps Work? The Evidence in 2026

The Industry and the Claims

Brain training is a multi-billion dollar market. Apps like Lumosity, Elevate, Peak, and dozens of competitors promise to sharpen memory, boost processing speed, and improve cognitive performance across the board — often with impressive-looking graphs of user improvement over time. The marketing is sophisticated; the science is more complicated.

In 2016, the US Federal Trade Commission fined Lumosity $2 million for "unsubstantiated claims" that their games would improve performance in everyday tasks and protect against cognitive decline. This was not a fringe case — it reflected the consensus among academic researchers who had spent years studying transfer effects from cognitive training.

What the Research Shows

The cognitive training literature has one clear and highly replicable finding: people improve on the specific tasks they practise. If you play a game that requires rapid visual pattern matching, you will get better at rapid visual pattern matching. This is called "near transfer."

The question that matters is whether improvement on the trained task transfers to untrained cognitive abilities — and ultimately to real-world performance. This is "far transfer," and the evidence here is largely negative.

A 2014 consensus statement signed by over 70 leading cognitive scientists concluded that current evidence does not support claims that brain training games produce general cognitive enhancement or protect against age-related cognitive decline. A 2017 counter-statement, signed by a different group, argued the evidence was more mixed. The disagreement is partly methodological (what counts as evidence) and partly about the magnitude and persistence of any effects found.

The 2016 Comprehensive Review

The most thorough academic assessment of brain training research appeared in Psychological Science in the Public Interest (2016), authored by Daniel Simons and colleagues. They reviewed evidence across the full range of commercial and laboratory cognitive training claims and found the same pattern: near transfer is robust, far transfer is not. When positive transfer effects did appear in published studies, they tended to be small, methodologically fragile, and dependent on comparison to passive control groups rather than active controls doing a different engaging activity. When active controls (who played non-cognitive games or took music lessons) were used, the brain training groups frequently showed no advantage over the comparison activity. The suggestion was that any benefits were probably due to engagement and practice effects, not specific cognitive mechanisms — and that other engaging activities would produce similar or better results.

Why Transfer Is So Hard to Achieve

The fundamental problem with cognitive training is that improvement tends to be highly task-specific. The brain optimises for the exact demands it encounters repeatedly. Practising one working memory task makes you better at tasks using the same format and the same processing demand — not at working memory tasks in general.

This is the opposite of how domain expertise works. A doctor who develops expertise in diagnosis is not merely getting faster at the specific diseases they have already seen — they are building generalised pattern recognition that transfers to novel presentations. The difference is that expertise training involves meaningful variability and real-world consequence; cognitive training apps typically involve narrow, repeated practice on a fixed format.

The Action Video Game Exception

The one genre of "screen-based cognitive training" with more credible evidence for genuine transfer is action video games — not the puzzle-based, brain-training-app style, but fast-paced action games requiring rapid visual attention, decision-making under time pressure, and multisensory integration. Meta-analyses by Daphne Bavelier at Rochester University find that action game players show improved visual attention, faster processing speed, and better attentional tracking compared to non-gamers — and that training non-gamers on action games produces these benefits. The transfer appears to operate through the visual attention and processing speed systems rather than higher-level cognition. This is narrow but real, and it is notably absent in puzzle, memory, or problem-solving games — suggesting the mechanism is domain-specific and does not generalise to the "train your brain" premise of most commercial apps.

What Does Transfer

The interventions that do show meaningful transfer to general cognitive function are almost entirely physiological:

• Aerobic exercise — consistent transfer to executive function, working memory, and processing speed across multiple meta-analyses
• Sleep — protecting and consolidating cognitive function broadly
• Domain expertise — years of deliberate practice in a complex domain builds cognitive structures that transfer within that domain and to closely related ones
• Mindfulness meditation — modest transfer to attentional control

The Verdict

Brain training apps are not cognitive enhancers in any general sense. They are cognitive practice environments for specific tasks, and they produce specific improvements in those tasks. If you want to perform better on a particular type of visual attention task, practising that specific task will help. If you want to improve your general cognitive performance, the evidence points away from apps and toward exercise, sleep, stress management, and substantive domain learning.

The most honest use of a brain training app is entertainment combined with mild familiarity with specific task formats. As a cognitive enhancement investment, the money and time are better spent elsewhere.