Quick answer: Brain games can make you better at the games and at closely related tasks, but the best available evidence does not show reliable broad transfer to unrelated academic performance. If the goal is a better examination result, most preparation time should therefore be spent on the knowledge, reasoning, language, retrieval and paper-control skills that the examination actually requires.
This article began in 2014 as a short list of brain-training websites for pupils preparing for examinations. The idea was sensible for its time: exercise the mind before an important test. Research since then allows us to make the idea much more precise.
The important distinction: near transfer and far transfer
When a person practises a task, improvement can appear in different places.
- Near transfer means improvement on the trained task or a very similar task.
- Far transfer means improvement spreads to substantially different abilities or outcomes — for example, from a working-memory game to broad reasoning, language performance or school achievement.
A 2019 second-order meta-analysis covering many cognitive-training programmes and populations found that near transfer occurs, while far-transfer effects are small or null and disappear when important biases are controlled. A separate meta-analysis of children’s executive-function training found a significant near-transfer effect but no convincing far-transfer effect to untrained executive functions. This does not mean the brain cannot learn. It means that learning is often more specific than marketing language such as “train your brain” suggests.
For a student, that distinction changes the preparation question from “How do I make my brain generally stronger?” to “What capability must be available when this particular examination asks for it?”
Why getting better at a puzzle is still real learning
If a child repeatedly solves Sudoku, memory sequences, visual-search games or logic puzzles, the child may become faster and more accurate at those tasks. That improvement is genuine. The mistake is to assume automatically that the gain will travel into PSLE English comprehension, Secondary Mathematics algebra, Science open-ended explanations or an IB essay.
The trained activity and the target activity need shared structure. A puzzle can be enjoyable, can provide a useful challenge and can help a learner notice strategies such as checking constraints or holding intermediate information. But if the examination requires knowledge of fractions, inference from a passage or a scientific explanation using evidence, those capabilities must themselves be practised.
The examination is a transfer problem
An examination rarely asks a learner to reproduce the lesson exactly as it was first taught. The student must recognise what is relevant in a changed situation, retrieve the right knowledge, choose a method, execute it and monitor the answer under time pressure.
That means useful preparation should move through a sequence:
- Understand: know what the concept, method or language feature means.
- Retrieve: produce it without the notes open.
- Recognise: identify when it applies.
- Vary: practise it in changed examples.
- Integrate: mix it with other topics or skills.
- Execute: use it under realistic paper conditions.
- Review: use mistakes to locate the next weak link.
This is a much closer match to the transfer students actually need.
What should a student practise instead of generic brain training?
1. Retrieval from memory
Close the notes and attempt to recall, explain, write or solve. The gap between what feels familiar and what can actually be produced becomes visible. This is especially important for vocabulary, formulae, scientific relationships, definitions, procedures and factual knowledge.
2. Discrimination between similar problems
Students often know several methods but choose the wrong one. Mixed practice makes the learner decide which tool applies rather than being told by the worksheet heading. This is a deeper form of preparation than doing twenty identical questions in sequence.
3. Explanation
Ask the learner to explain why an answer is correct, why an alternative is wrong, or what changed between two examples. Explanation exposes shallow imitation. In Science, English and Mathematics alike, being able to articulate relationships can reveal whether the learner has a usable model or only a memorised surface pattern.
4. Error correction
A wrong answer is valuable when the learner can identify the cause. Repeating the same type of paper without repairing the cause can simply generate more evidence of the same weakness.
Useful error categories include missing knowledge, misreading, wrong method selection, incomplete reasoning, language weakness, calculation accuracy, careless transcription and poor time allocation. The categories are not grades; they are routes to action.
5. Timed execution
Once accuracy exists, students need practice using it at examination speed. Timing should be learned inside the actual subject and paper. A Mathematics question, an English composition and a Science structured response place different demands on reading, planning, production and checking.
See also Top 5 Tips to Study for PSLE: A 2026 Guide to Preparation, Memory and Exam Execution.
Can brain games still have a place?
Yes — if their role is defined correctly.
- For enjoyment: puzzles can be intrinsically satisfying.
- For a short warm-up: a brief challenge may help some students transition into focused work, provided it does not consume the study session.
- For a specific skill: if the trained task closely resembles a capability the learner genuinely needs, near transfer may be useful.
- For metacognition: a teacher can use a puzzle to discuss strategy, checking, persistence, assumptions or constraint management.
- For enrichment: logic and strategy games can broaden a learner’s experiences even when they are not examination interventions.
The key is not to confuse an enjoyable enrichment activity with a validated shortcut to general academic performance.
A simple test for any “brain training” claim
Before adopting a training programme, ask five questions:
- What exactly is being trained?
- What outcome is being claimed?
- How similar is the trained task to the claimed outcome?
- Was improvement measured against an appropriate control group?
- Does the improvement persist and appear outside the training task?
This is useful beyond education. A claim becomes stronger when the measured outcome is close to the real-world outcome we care about, and when competing explanations have been tested.
What “brain fitness” should mean for a student
For examination preparation, a useful version of “brain fitness” is not a single hidden mental capacity. It is a collection of available capabilities:
- knowledge that can be retrieved;
- concepts that can be recognised in unfamiliar forms;
- methods that can be selected rather than blindly repeated;
- language that can carry meaning accurately;
- attention that can return after interruption or difficulty;
- checking routines that catch preventable errors;
- timing decisions that protect the whole paper;
- the willingness to expose uncertainty and ask for help.
Those capabilities are trainable, but they are trained most reliably by working with the material, decisions and conditions in which they must later operate.
For parents: do not mistake activity for transfer
A child can look intensely busy while practising something that has little relationship to the weakness limiting school performance. Before adding another app, game, worksheet or course, look at the evidence already available: marked work, teacher comments, recurring errors, oral explanations and timed performance.
If the child cannot explain a Science relationship, more generic memory games may not solve it. If the child understands Mathematics but repeatedly loses marks through notation and checking, the intervention should target execution. If English comprehension fails because background knowledge and vocabulary are too thin, the learner needs sustained reading, language development and passage work.
The question is always: what is the earliest weak link that prevents the desired performance? Train as close to that link as practical, then test whether the improvement reaches the real task.
Sources
- Sala, G. et al. (2019), Near and Far Transfer in Cognitive Training: A Second-Order Meta-Analysis, Collabra: Psychology.
- Kassai, R. et al. (2019), A meta-analysis of the experimental evidence on the near- and far-transfer effects among children’s executive function skills, Psychological Bulletin.
- Current exam preparation should also be checked against the relevant official examination specification, such as the SEAB PSLE formats for Primary 6 pupils.
Archive note: First published in 2014 as “Exercise your brains for exam preparation.” The original post recommended several brain-game websites. The URL has been preserved, while the article has been rebuilt around what later evidence tells us about training, transfer and examination-specific practice.
A Deeper Reader: Eight Questions for Testing Whether Training Actually Transfers to the Real Task
The phrase “brain training” is attractive because it suggests one hidden capacity can be strengthened and then used everywhere. Real learning is usually more specific. Practice changes performance most reliably where the trained task and the target task share knowledge, representations, decisions or procedures. The farther apart they are, the more carefully transfer has to be demonstrated rather than assumed.
1. What is the actual target of training?
Before choosing a game, app or exercise, name the desired outcome precisely. Is the goal faster arithmetic, stronger vocabulary retrieval, better inference, improved working-memory performance on a particular task, or fewer examination timing errors? “Sharper brain” is too vague to test.
Once the target is concrete, the learner can compare the training activity with the real task and ask what structure they genuinely share.
2. Which distinctions keep training claims honest?
- Practice effect is not broad cognitive improvement.
- Near transfer is not far transfer.
- Task familiarity is not underlying academic knowledge.
- Improved speed is not automatically improved reasoning.
- Short-term gain is not durable retention.
- Enjoyable challenge is not automatically an examination intervention.
- Correlation between players and high performers is not proof the game caused the performance.
3. How does transfer happen when it does happen?
Transfer becomes plausible when the learner can recognise a shared structure across different surfaces. A student who learns to check constraints in a logic puzzle may later use a similar habit in Mathematics, but only if the relevant constraint-checking process is noticed and practised in Mathematics too. The bridge cannot always be assumed to build itself.
This is why varied subject practice is powerful. It keeps the important structure while changing wording, context or representation. The learner has to carry the method across a meaningful distance that resembles the distance required in the actual examination.
4. What evidence should support a training claim?
Good evidence asks whether improvement exceeds ordinary practice or expectation, whether an appropriate comparison group was used, whether the outcome was measured independently from the training game, whether gains remained after time passed and whether they appeared in the real target task.
It is easy to demonstrate that repeated practice improves the practised activity. The stronger claim—improvement in a different academic outcome—requires stronger evidence.
5. Which counterexamples reveal weak transfer?
The Sudoku expert. Pattern skill grows while English comprehension remains unchanged. The memory-app champion. Sequence scores improve but Science explanations still fail because causal knowledge is missing. The fast clicker. Processing speed in a game rises but careful examination accuracy does not. The motivated user. Performance improves partly because the programme created regular study time rather than because its special cognitive mechanism was unique.
6. How should this change examination preparation?
Train close to the capability that must appear on the paper. Retrieve the vocabulary that must be used. Solve the mathematical structures that must be recognised. Practise scientific explanation. Read unfamiliar passages. Rehearse timing and recovery inside the actual paper type. Then vary the surface enough that the learner cannot survive by memorising one template.
Generic puzzles can remain enrichment, warm-up or recreation. They simply should not displace the higher-value practice when examination time is scarce.
7. What does a transfer-testing cycle look like?
- Define the real target performance.
- Take a baseline on that target.
- Choose training with a plausible shared structure.
- Practise long enough to change the trained task.
- Retest the real target using new items.
- Check again after a delay.
- Compare improvement with ordinary subject practice where possible.
- Stop or redirect if the promised transfer does not appear.
8. What should remain after the brain-game claim is stripped away?
The useful idea is specificity with variation. Build the actual knowledge and decisions the learner needs, then make them flexible by practising across contexts. Enjoy puzzles because they are worthwhile puzzles; use them as teaching objects when the shared strategy is explicit; but demand direct evidence before claiming they improve unrelated academic performance.
Next route: connect this page to retrieval, spacing, interleaving and PSLE preparation. Keep this article as the transfer-evidence owner rather than a catalogue of brain-game products.