Where the excitement came from

In 2008, Jaeggi and colleagues published a study in which people who trained on Dual N-Back improved on tests of fluid intelligence — the ability to reason and solve novel problems — and the improvement scaled with how much they trained. That finding was striking because fluid intelligence had long been considered largely fixed in adulthood. It's the reason Dual N-Back became the brain-training task people talk about.

Why it's still debated

Follow-up work has been mixed. Several later studies and meta-analyses — for example work by Redick and colleagues, and reviews by Melby-Lervåg and Hulme — often found little or no transfer to general reasoning once the comparison was made against an active control group (people doing a different demanding task) rather than a group doing nothing. Effects in this literature tend to shrink as the study designs get stricter. Commonly cited reasons for the disagreement:

• Small sample sizes in many of the individual studies
• Passive vs. active control groups producing very different results
• Different tests used to stand in for "fluid intelligence"
• Expectancy effects — people who believe training works may try harder on the outcome test
• Publication bias toward positive findings

What is well supported

Near transfer. Training Dual N-Back improves your Dual N-Back score and your performance on tasks that closely resemble it — other n-back variants and short-term / working-memory span tasks. This part is not really in dispute. Working memory is trainable in the sense that you get measurably better at working-memory tasks.

What's contested is far transfer: whether that carries over to unrelated things like reading comprehension, problem-solving, or IQ. Treat any far transfer as a possible bonus, not the reason to train.

How to think about it

• As a workout, it's excellent. It's hard, it adapts to your level, and it gives you a single number (N) that goes up as you improve — which is motivating and measurable.
• Set the expectation at "I'll get good at this and build a daily habit," not "my IQ will go up by X points."
• Consistency matters more than duration. The studies that report effects use weeks of short, near-daily sessions, not occasional long ones.
• It's free to test the claim yourself. Track your N over a few weeks and see what changes.

Who tends to get the most out of it

• People who want a genuinely demanding mental task with a clear progression
• People building a daily training habit — it pairs well with a streak
• People who've found other "brain games" too easy or too shallow
• People comfortable with the honest framing: measurable practice effect, uncertain broader payoff