Twelve cards lie face up; each shows one, two or three symbols that vary in shape, colour and shading. Anyone who sees a 'set' — three cards where each feature is either all the same or all different — calls it, takes the cards, and three more are dealt. Most sets at the end wins.
Why it may be useful
Finding a set means checking four attributes at once and switching which one you are matching on. Harvard's executive-function guide names SET among card games that exercise this kind of cognitive flexibility, and it plays solo too.
Keep in mind
Real-time, first-to-spot play favours the fastest eyes, so mixed-age groups need a handicap or the turn-based variant, and children who never win a round give up quickly.
Written 2026-09-15 · editorial assessment, not a measured effect · how we rate →
What the game is
SET is a card game of eighty-one cards and one rule. Every card carries one, two or three identical symbols, and cards differ in four ways: how many symbols, what shape they are, what color, and how they are filled. Each of those four features has exactly three possible values, which is why the deck is exactly eighty-one cards with no repeats.
Setup is dealing twelve cards face up in a grid, and that is the whole preparation. A set is any three cards where each of the four features is either the same on all three or different on all three. There are no turns: everyone looks at the same twelve cards at once, and the first person to see a set says so and takes it. Three new cards fill the gap.
The game ends when the deck runs out and no sets remain, and whoever collected most wins. It plays with one to eight people in roughly twenty to thirty minutes, and works alone as a puzzle. Created in the 1970s and published commercially in 1991, it is now sold by PlayMonster.
What you actually do
Your eyes go over twelve cards looking for three that agree. You cannot search on one feature alone: two red cards and one green is fine only if the other three features also resolve, so a promising pair has to be checked all the way through and then dropped. Most players start by fixing on color, then learn to hold a pair and ask what the third card would have to be, which is faster because that card is uniquely determined. Getting better means switching which feature you anchor on, and abandoning a pair the moment it fails.
What it exercises
Checks three cards against four features at once before calling
Works out which single third card would complete a pair
Switches which feature to search on when one approach stalls
Drops a promising candidate quickly instead of forcing it
Why we rate it Low Worksheet score 51 / 100
FitHow much real knowledge or skill does the play itself make you use?50weight 45%
Fit scores 50. Almost nothing sits between the player and the named skill: there is no theme, no story and no luck beyond the deal, and the only activity is checking combinations against several attributes and shifting which one you anchor on. Speed pressure, which can crowd out careful checking, is what keeps this a solid rather than exceptional fit.
TransferWill what you practise here work outside the game?38weight 30%
Transfer scores 38, the lowest of the three. Visual search over a closed set of abstract attributes is a narrow task, and the honest expectation is that a player gets better at SET rather than at reading or arithmetic. The game earns its place because the attribute-shifting is unusually pure, not because transfer to other tasks is established.
PracticeDoes the game make you use that skill again and again, as it gets harder?70weight 25%
Practice scores 70. Every round contains many searches and a session contains many rounds, so the same act repeats hundreds of times in an evening, and the deck never lays out the same way twice. What holds the number in this band rather than higher is that the difficulty never scales: the puzzle is identical on day one and day two hundred.
What the evidence says
Practice-backedNo published study, but people who answer for the outcome use it: schools, clinics, museums, research labs.
The evidence level is Practice-backed, meaning expert recommendation rather than a measured outcome. Harvard's Center on the Developing Child names SET in its executive-function activity guide for five- to seven-year-olds, as a card game that practices cognitive flexibility by requiring matches on more than two attributes at once. That is a recommendation from people who study the area, not a trial of this game. Small student-journal studies of SET and visual search exist but are not strong enough to cite. The PlayMonster product page supplies the eighty-one-card contents, the age guidance and the rule as the publisher states it.
Editorial assessment - not a measured scientific effect.
Fit
50/100
Transfer
38/100
Practice
70/100
Why it makes the cut
The 'aha' of spotting a set before everyone else is the whole game, played in maths clubs and at kitchen tables since the 1990s without any teaching framing.
Practice-backed
Harvard's Center on the Developing Child names SET in its executive-function guide as a card game that practices cognitive flexibility. That is expert recommendation, not a trial of this game.
