Whether a brain-training game works depends on what improvement it promises. Getting better at the game is a different outcome from improving broadly across unrelated tasks or everyday activities.

Three levels of a claim

  • Trained task: performance improves on the activity you practiced.
  • Related task: improvement extends to a similar activity.
  • Broad transfer: improvement extends to substantially different tasks or daily functioning.

A progress chart inside an app may demonstrate the first level. It does not automatically establish the others.

What a major review found

The 2016 review discussed by the Association for Psychological Science found stronger evidence for improvement on trained tasks than for broad transfer. It also identified limitations in the design of many studies.

This is a reason to inspect the specific evidence behind a product's promise. It is not a claim that every game is useless or that no form of learning can transfer.

Ask what the comparison was

If a study compares people who practiced with people who did nothing, other differences may contribute to the result. A more informative design considers an appropriate comparison activity, relevant outcomes and whether the tests themselves resemble the training.

Also ask whether the reported gain is meaningful for the promised use. A faster response within one game is not the same outcome as improved performance at school or work.

Set a goal you can observe

You might choose a puzzle because it is enjoyable, because you want to learn a particular strategy or because you want a structured practice routine. Those are understandable reasons that do not require a guaranteed IQ increase.

For example, learning to distinguish reflections from rotations is a specific skill you can demonstrate on new shapes.

How this applies to starp

starp provides reasoning questions, explanations and optional repeated testing. It should be used as practice and personal insight, not as a treatment or a promise of general cognitive enhancement. Evaluate progress by looking at the methods you understand and the errors you can explain, alongside any score history.