The N-Back game, reviewed: our hardest minute

A hands-on look at the toughest game we make: what it actually trains, how we score it, and an honest take on whether it makes you smarter.

Play it: N-Back Fluid memory · about 1 minute · free, no sign-up
Play N-Back →

The one game people email us about

Most of our games just get played. N-Back gets argued about. It's the game with a reputation — the one a 2008 study suggested might raise your IQ, the one forum threads still call a "brain gym." So it earns a proper hands-on review: what it is, how ours plays, and whether any of the hype survives contact with the evidence.

What N-Back actually is

N-Back is easy to describe and genuinely hard to do. A sequence of items appears one at a time — in our version, a square lighting up somewhere on a 3×3 grid. Your job is to respond whenever the current position matches the one from N steps earlier. In a 2-back, that means "is this square the same as the one two flashes ago?" You're always comparing now against a target that keeps sliding forward.

The original lab version — the one from the famous study — was a dual N-back: a spatial stream and a stream of spoken letters at the same time, each judged independently. It is as punishing as it sounds. Ours runs the single spatial stream, which is plenty demanding and far less likely to make you want to launch your laptop across the room.

Why it feels so hard: you can't simply "memorise" the sequence. Every new square forces an update — drop the oldest item, shuffle everything back one slot, hold the new one ready. That constant refreshing is the entire point.

How our version plays

Our N-Back runs a 2-back by default, with the option to drop to 1-back or push up to 3-back. Squares flash on the grid; you tap when the position repeats from two back. Crucially, false alarms cost you — flag a match that isn't there and your score takes the hit. That one design choice matters: it stops you mashing the button and rewards genuine restraint over trigger-happy guessing. A round lasts about a minute, and you're scored on accuracy: hits, minus the damage from false alarms.

Expect to feel clumsy at first. A 2-back humbles almost everyone on the opening go — that's normal, not a verdict on you.

What it trains, for real

N-Back is a workout for working-memory updating: holding a small amount of information in mind, comparing against it, and refreshing it as new items arrive. That's a real, measurable cognitive process, and it shows up in things like mental arithmetic and keeping track of a fast conversation. Getting better at N-Back is a legitimate, if narrow, achievement.

But does it make you smarter?

Here's the honest part, and it's exactly why this game has a reputation. In 2008, Jaeggi and colleagues reported in PNAS that dual N-back training improved fluid intelligence — raw, novel problem-solving — and that more training meant bigger gains. It was a genuinely exciting result, and the internet ran with it.

Then other labs tried to reproduce it. Redick and colleagues (2013) ran a careful version with a proper active control group and found no transfer to fluid intelligence. Larger reviews since — including Simons and colleagues' comprehensive 2016 analysis — landed on a sober consensus: brain-training tasks reliably improve the trained task, sometimes closely related tasks, and show little to no far transfer to general intelligence or everyday life. So N-Back makes you better at N-Back. The IQ boost hasn't held up.

We unpack the full replication saga in the N-back and IQ story. Short version: play N-Back because it's a satisfying, honest challenge — not because it will raise your ceiling.

Who'll love it (and who'll bounce)

You'll enjoy N-Back if you like a game that genuinely fights back — the kind where a two-point improvement feels earned. You'll bounce off it if you want quick, breezy wins; this is the opposite of a casual tap game. It's the most cognitively effortful minute on the site, and some people find that thrilling while others find it stressful. Both reactions are completely fair.

Getting better at the game

A few honest tips — for your score, not your IQ:

  • Find a rhythm. The flashes arrive at a steady tempo; settling into it frees up attention for the comparison.
  • Don't over-flag. Because false alarms hurt, when you're unsure it's usually better to hold off. Precision beats enthusiasm.
  • Start at 2-back and stay a while. Jumping to 3-back too early mostly teaches you to feel overwhelmed. Own the middle level first.
  • Play rested. Working memory sags hard when you're tired — a bad night will cost you more points than a month away from the game.

The verdict

N-Back is our hardest, most honest minute. As a test of working-memory updating it's excellent: demanding, cleanly scored, and immune to button-mashing. As an IQ pill, it's a myth we're happy to keep puncturing. Play it for the challenge and the satisfaction of a rising number you actually earned — and treat anyone selling it as a shortcut to genius with a large pinch of salt.

Take the 2-back challenge

One minute, honestly scored, no login and no miracle claims — just the hardest game we've got.

Play N-Back →

CortexCrunch is a cognitive practice tool, not a medical device. The games and articles here are inspired by research in cognitive science, but we make no claims about treating, diagnosing or preventing any condition. Think of it as a gym for your mind — the benefits depend on consistent practice.