Five Questions to Ask About a Validated Cognitive Test

A cognitive test described as validated was validated for something specific: a purpose, a population and a comparison. Those three details decide what a score from it means in a new setting. Five questions get a buyer to them, and they work on any instrument, including ours.

A rugby club, a supplement and a cognitive test

Bristol Bears announced on Oct. 14 that CONKA, a cognitive performance company, becomes the club's Official Cognitive Performance Partner for the 2026/27 season. The announcement says the partnership will use the Integrated Cognitive Assessment, which it describes as "a validated two-minute cognitive test," to help track cognitive readiness in first-team players. Adam Roberts, the club's head of medical, and CONKA co-founder Harry Glover are both quoted in it.

Nothing in that announcement is unusual. It is a useful example because the word validated is carrying real weight in that sentence, and a reader cannot tell from the sentence what it covers. The published record on that instrument is public and specific, so walking through it shows what the five questions do.

Validated for a purpose, in a population

The Integrated Cognitive Assessment, sold as CognICA by Cognetivity Neurosciences, is a rapid categorisation task delivered on a tablet. Its main validation paper is Kalafatis and colleagues, published in Frontiers in Psychiatry on July 22, 2021, under the title "Validity and Cultural Generalisability of a 5-Minute AI-Based, Computerised Cognitive Assessment in Mild Cognitive Impairment and Alzheimer's Dementia." The sample was 95 healthy participants at a mean age of 66.8, 80 participants with mild cognitive impairment at a mean age of 69.6, and 55 with mild Alzheimer's dementia at a mean age of 71.2. The test separated mild cognitive impairment from healthy participants with an area under the curve of 0.814, and mild Alzheimer's from healthy participants at 0.883. The authors conclude that it "can support clinicians by aiding accurate diagnosis of MCI and AD" and that it "is appropriate for large-scale screening of cognitive impairment."

That is a strong result, and it is a result about detection, in older adults, against a clinical comparison. FDA cleared the test for commercial distribution in the United States in October 2021, and trade coverage at the time described it as an assessment for the early detection of dementia. A clearance is tied to an intended use statement that lives in FDA's own record rather than in anyone's press release, and that record is the place to read what a cleared device is cleared for. We have written at more length about the difference between clearance, approval and the general wellness category.

There is also published work on this instrument in athletes, and a reader asking the five questions would find it. Daniel Glassbrook, Paul Chazot and Karen Hind measured the same instrument's same-day precision in 31 healthy, uninjured athletes aged 18 to 40, in contact and non-contact sports, before and after a normal training session, and published in Applied Neuropsychology: Adult in February 2025. They report high precision, with every measure inside a 5% coefficient of variation except the pre-session combined index at about 7.1%, and they conclude the test "is a practical test which can be used to measure cognitive function before and after sport participation." They also state their own limits: 31 participants, training duration and intensity unrecorded, and no comparison yet between contact and non-contact sports or between amateur and professional athletes.

So the honest picture on this instrument has two public parts, and the word validated points at whichever one covers the use in front of you. One small discrepancy is worth noting plainly. The club's announcement calls it a two-minute test, and the validation paper's own title calls it a five-minute test. Each is quoted accurately here, and the current administration time is a question for the manufacturer.

Five questions to ask about a validated cognitive test

• Which instrument produced this number, by name? A score with no named instrument cannot be checked by anyone.

• What was it validated to do? Detection, diagnosis support, screening and performance tracking are four different jobs with four different evidence requirements.

• In which population, at what ages, against what comparison? A result in adults with a mean age near 70 and a clinical comparison group is evidence about that setting.

• Who ran the study, and who funded it? A manufacturer-funded study is still evidence, and the funding belongs in a reader's view of it.

• Where is it published, and does the paper state its own limits? A paper that names its limits is easier to trust than one that does not.

Five answers take a few minutes to collect and they settle most of what a buyer needs. A number with no instrument behind it is an opinion with a decimal point.

Percentages without a named study

The same announcement reports a 13-week trial with first-team players producing an average 7.9% improvement in cognitive performance, with focus up 19.3% and memory up 7.9%, described as statistically significant. It also cites an independent study of professional athletes reporting a 14.86% improvement for product users against a 0.69% decline in a placebo group, described as peer-reviewed and conducted with UK universities. The announcement names no journal, no university and no study.

That is question five with nothing behind it yet. The research may be published and sound, and a reader cannot check it from the announcement, which is the whole difficulty. A figure outlives the document it came from. It gets pulled into a deck, then into a summary of the deck, then into an answer an assistant gives someone who never saw either, and at each step the denominator and the study name are the first things dropped.

This is a problem for the category rather than a complaint about one club. Most announcements in this space are built the same way, and ours were until we changed how we write them.

The practice we hold ourselves to

Neurova Labs publishes figures from its own cohorts, and the rule we work to is that a figure travels with its participant count and its study design in the same sentence. Our own measurement is the iXpressGenes CoreSignals Panel, which reads 22 markers of RNA gene expression from whole blood and is designed to track inflammatory and immune signals associated with stress and recovery. It is not used to diagnose any condition. The limits of that panel are written out on our measurement page, including the point that the same signals move with sleep loss, illness, exertion and diet.

Neurova Academy is a facilitator-led general wellness and cognitive training program from Neurova Labs. Neurova Labs makes no claim to measure cognitive readiness and reports no cognitive readiness outcome. No single accepted measure of cognitive readiness exists, and a score reported under that name describes the instrument that produced it. The fuller version of that position is on our claims page.

Ask us the five questions. If a figure of ours ever reaches you without its participant count attached, that is a defect on our side, and we would rather hear about it than have it travel. Write to info@neurovalabs.com.

Neurova Academy is a general wellness and cognitive training program. It is not a medical device and is not offered to diagnose, treat, cure, or mitigate any disease or condition, including concussion, traumatic brain injury, or CTE. Outcomes described reflect observational pilot findings and individual experience, and are not a prediction of results for anyone else. Individual results vary. Laboratory testing provided by iXpressGenes, a CLIA-certified, CAP-accredited laboratory.