Sport-specific knowledge beats generic brain training

Insporta tests what a player knows about their sport, not how fast they press a button or how well they solve abstract puzzles. The reason is empirical. In the largest recent head-to-head comparison, Reinhard, Mann and Höner (2025) found that a sport-specific decision-making test explained 34% of the variance in youth football playing level, a generic “Football Intelligence” composite explained 13%, and a reactive stress test explained 0%. Sport-specific measurement is where the signal is.

The study that tested the brain-training hypothesis

Commercial cognitive products for sport rest on a simple hypothesis: elite athletes have better general cognition, so a generic test of attention, reaction or working memory should find talent. Reinhard, Mann and Höner, published in the Journal of Science and Medicine in Sport in 2025, put that hypothesis to a direct test with 110 youth footballers.

Each player completed three kinds of assessment:

Assessment	What it uses	Variance in playing level explained
Sport-specific decision-making test (360-degree match video)	Real game situations, the player chooses the next action	34%
Generic “Football Intelligence” score	A composite of four generic cognitive tasks with no football content	13%
Reactive stress test (Vienna Test System type)	Abstract stimuli under time pressure	0%

The gap is not a matter of one test being slightly better calibrated. The generic reactive test carried no information about playing level at all, and the generic composite carried less than half of what the sport-specific test carried.

The authors note their own limits: the sample is 110 players, all male, all already inside an academy system, and they call for larger samples. Those limits are discussed in how to validate a sports knowledge test. They do not change the ordering of the three results.

Two traditions, one answer

The 2025 paper is the latest round in a debate that has run for more than thirty years.

The cognitive-component approach (Nougier et al., 1991) holds that athletes are better than non-athletes at general cognitive functions, so you can measure them with generic stimuli: arrows, coloured circles, letter sequences. This is the tradition behind brain-training apps and generic sport-cognition batteries.

The expert-performance approach (Ericsson and Smith, 1991) holds that experts outperform others only, or mainly, in tasks that belong to their domain. A chess master remembers real board positions far better than a novice, but random piece placements no better. On this view the right stimulus for a footballer is football.

Kalén et al. (2021), a meta-analysis in Psychological Bulletin covering hundreds of studies, settled most of the argument. They sorted cognitive measures into three levels and found that sport-specific decision-making skills, the third level, discriminate between athletes of different standards far better than basic or higher generic functions. The three-level hierarchy is the frame Reinhard, Mann and Höner adopt, and the frame Insporta works within.

Why generic tests keep failing

Three findings from the wider literature explain the 13% and the 0%.

Executive functions develop the same way in everyone. Beavan et al. (2020) followed 343 players of a professional German club, ages 10 to 34, across three seasons with 1,018 observations. The developmental trajectories of working memory, inhibition and flexibility in these high-level players matched the general population. Years of elite training and selection left no visible mark on generic executive function. The authors conclude that including executive functions in talent identification is questionable. See what a knowledge test does not measure.

Names on tests are not evidence. Furley, Schütz and Wood (2023) describe the jingle-jangle fallacy: tests with the same name do not necessarily measure the same thing, and tests with different names may measure the same thing. A score called “Football Intelligence” that is built from four generic tasks and contains no football stimulus is the textbook case.

Response format matters. Travassos et al. (2013), a meta-analysis of 111 effect sizes, found that expertise effects grow as the required response becomes more like the real action: verbal reports show the smallest effect, in-situ actions the largest. Abstract button pressing sits near the weak end.

Where Insporta sits, and where it does not

Insporta is a knowledge test: ten questions, four options, one key, on the rules, tactics, patterns and history of a specific sport. Every question is tagged with the topic it tests, and response time is recorded as one dimension alongside accuracy. This places Insporta in the sport-specific tradition, at Kalén’s third level of decomposition.

It is important to be precise about what that does and does not mean. The 34% figure belongs to a 360-degree video decision-making test, not to a written knowledge test. Knowledge and decision-making are close relatives: a player cannot make the right tactical call without knowing what the options are, and the recall-versus-applied split in Insporta’s questions is designed to reach toward applied judgment. But they are not the same construct, and Insporta does not borrow the decision-making test’s numbers as its own. The construct Insporta claims is sport-specific knowledge, stated plainly, in line with the transparency Furley and colleagues ask of every cognitive tool in sport.

That precision is also why the methodology excludes whole classes of question: nothing on general cognitive ability, executive functions or reaction speed, and nothing on the physical or biomechanical traits of the athlete. Those things may matter for sport, but a knowledge test cannot measure them, and pretending otherwise is the error the research warns against.

What this means for a coach or a parent

A player’s score on a generic brain-training app tells you little about their level in their sport. The best-supported way to assess the cognitive side of sport is with material from the sport itself, whether that is a video decision task in a research lab or a knowledge test on a phone.

The size of the effect should be kept realistic. Even the best sport-specific test in Reinhard et al. explained about a third of the variance in playing level; the other two thirds are physical, technical, motivational and situational. A knowledge test is one instrument among several, and the full review of the twelve sources is the best guide to how much weight each one can bear.