TLDR: Nine PlayMemorize games mirror the formats psychologists have used for a century to measure cognitive ability · Matrix (Raven’s matrices), Spatial (mental rotation), Analogies (verbal reasoning), Stroop (interference), Odd One Out (classification), Sequences (numerical pattern), Color (visual-motor sequencing), Math (arithmetic fluency), and Pi (digit span). Each one isolates a sub-skill that has decades of validation behind it.
The classic IQ subtests were designed to isolate sub-skills that load on general cognitive ability while being short and replicable. Most of them have been in use for fifty to a hundred years and have a deep psychometric literature behind them · which makes them surprisingly good training formats too. Knowing what each test was designed to measure tells you what each game is actually exercising, and seeing the family map clarifies which sub-skills are independent versus which compound.
What you will get out of this article. The historical test each game is modelled on, what cognitive sub-skill it actually measures, an inline round of each, and a workout that mirrors a balanced cognitive battery.
A brief history of the formats
Cognitive testing as we know it begins around 1905, when Alfred Binet built the first practical intelligence test for the French school system. Over the next century the field assembled a stable toolkit:
Raven’s Progressive Matrices (1936). John Raven’s 3×3 grid with the bottom-right cell missing. A culture-fair non-verbal reasoning test that became the most-used cognitive subtest in the world. Matrix is the format.
Mental rotation (Shepard & Metzler, 1971). Two block figures, are they the same or mirror images? The format introduced the modern study of spatial cognition. Spatial is the format.
The Stroop task (1935). John Ridley Stroop’s word-colour interference paradigm. The single most-cited paradigm in cognitive psychology · runs in milliseconds, has thousands of papers behind it. Stroop is the format.
Verbal analogies (turn of 20th century). “Doctor is to hospital as teacher is to ___?”. Used on the SAT from the 1940s until 2005, when they were removed for being too predictive of overall score.
Digit span (early 20th century). Repeat a string of digits forwards or backwards. The classic working-memory test. Pi training stretches the same buffer.
Number-sequence problems. What comes next? A staple of arithmetic-reasoning subtests since the 1920s. Sequences is the format.
Odd-one-out / classification. Strip away the surface features to find the hidden category. Used on Wechsler tests since the 1940s.
Corsi block-tapping (1972). Reproduce a sequence of locations. The visual-motor cousin of digit span. Color is the format.
Mental arithmetic. Wechsler’s Arithmetic subtest has been part of the WAIS since 1955. Math is the format.
The single most striking pattern in this list is that each format isolates a different sub-skill, and the sub-skills are mostly independent · which is why a real cognitive battery uses several of them rather than relying on one. A balanced PlayMemorize routine that touches all nine is, in spirit, a self-administered short-form battery.
All nine classic-test games at a glance
Game-by-game
🧩 Matrix · Raven’s Progressive Matrices
Matrix mirrors Raven’s Progressive Matrices · a 3×3 grid with the bottom-right cell missing, where rules cascade across rows and columns. It is the single most-used non-verbal IQ subtest worldwide because it is culture-fair (no language needed), validates well across populations, and genuinely measures the rule-search engine.
🔄 Spatial · mental rotation
Spatial is the textbook mental-rotation paradigm · target shape on top, candidates below, three of them mirror images and one of them the target rotated. The format from Shepard and Metzler’s 1971 paper has stayed essentially unchanged because nothing replaces the cleanness of the design.
🌈 Stroop · interference
Stroop is the gold-standard interference test · word RED written in blue ink, name the ink. The 1935 paradigm has been cited tens of thousands of times because it isolates inhibitory control in a way that few other tasks can.
🔗 Analogies · verbal reasoning
Analogies presents pairs of words and asks you to complete the second pair so the relationship matches the first. The College Board removed analogy items from the SAT in 2005 because they were considered too predictive of overall score · which is, perversely, exactly what makes them a good training format.
π Pi · digit span
Digit-span tasks ask you to repeat a string of digits forwards or backwards. They are the classic working-memory subtest and have been used in cognitive batteries since the early 20th century. Pi training stretches the same buffer · with the bonus that the same digits come up again, so chunking and pegging strategies start to compound.
🔢 Sequences · numerical pattern reasoning
Sequences mirrors the number-sequence subtest that has been part of cognitive batteries since the 1920s · spot the rule and continue. The skill is the foundation of mathematical reasoning, and the format is unusually scalable in difficulty.
🎯 Odd One Out · classification
Odd One Out gives you four to six items and asks which one doesn’t belong. The format has been part of Wechsler-style batteries since the 1940s and exercises classification under abstraction · finding a category that contains all but one item.
🎨 Color · Corsi block tapping
Color mirrors the Corsi block-tapping paradigm · the visual-motor cousin of digit span. The 1972 task asks the subject to tap blocks in the order they were lit; the cognitive sub-skill is the visual-motor sequencing buffer, separate from the verbal one.
🧮 Math · arithmetic subtest
Math mirrors the Wechsler Arithmetic subtest, part of every WAIS edition since 1955. The skill is calculation fluency under time, and although arithmetic seems “lower” than reasoning in the cognitive hierarchy, it correlates strongly with general ability because it stresses working memory.
🫥 Nuance · the Ishihara plate as a streak game
Nuance is the catalogue’s nod to chromatic-sensitivity tests · the same family as Ishihara plates and pseudo-isochromatic charts. A digit or letter ramps from zero contrast against a flat-colour background up to a difficulty-driven peak gap (32% lightness on Easy, ~4.5% on Hard), and you tap it as soon as it is visible. Unlike Ishihara plates, Nuance varies hue per round, randomises the position, and grades on lightness rather than red-green or blue-yellow channels · so the test stays fair for players with colour-vision differences.
How to use these games for self-assessment
This is not an IQ test. A real psychometric battery has standardised norms, environmental control, and a trained administrator. PlayMemorize gives you the formats, not the scores. What it can do is show you which sub-skills are your relative strengths and weaknesses · which is the more useful piece of information anyway, because it tells you where to direct practice.
Three habits that turn the games above into a useful self-assessment: first, play each game cold, in a quiet room, fresh in the morning · the variance from environment dwarfs any honest skill difference. Second, log your score on each one and look at relative performance, not absolute · “I am unusually weak at Stroop relative to Matrix” is more useful than “I scored 7 on Stroop”. Third, return to the same battery after four weeks of focused practice · the comparison shows what training actually moved.
Don’t take the score personally. Cognitive batteries measure skill at the moment of testing · not “intelligence” as a fixed property. The single largest day-to-day predictor of score is sleep. If you score badly on a tired day, that is real information about how the same brain performs tired versus rested · not information about the brain’s underlying capacity.
A 25-minute battery workout
A balanced session that covers most of the classic subtests, ordered to match the energy demand:
- 3 minutes Matrix · alert-brain reasoning
- 3 minutes Spatial · mental rotation
- 3 minutes Sequences · pattern detection
- 2 minutes Analogies · verbal reasoning
- 2 minutes Odd One Out · classification
- 3 minutes Pi · digit span
- 3 minutes Color · visual-motor span
- 3 minutes Math · arithmetic fluency
- 3 minutes Stroop · interference closer
Run the same battery weekly for four weeks and chart your scores. The improvement is rarely uniform across all nine · the surprises in the chart are where the real signal is, and pointing focused practice at the weakest two or three usually pulls the whole average up.
Where this matters off the screen
The cognitive sub-skills measured by these formats are real · they predict job performance, academic outcomes, and the speed at which adults learn new domains. But the games are tools, not labels. The point of practising them is not “to be smart” but to know which corners of your thinking are slow, fast, brittle, or robust · and to spend a few minutes a week on the corners that matter most for whatever you actually do with your brain.
The test that matters: after a month of mixed practice, return to the most challenging real-world task you do regularly · debugging code, drafting a contract, writing an essay, planning a trip · and notice whether the small cognitive frustrations have eased. They are the honest signal; the games are how you got here.
Polymath
Cross-game streak roulette drawn from the whole PlayMemorize catalogue. Pure full-spectrum test · every round can be any game
Play now - it's freeNo account needed. Works on any device.
