Can Sudoku make you better at PSLE Science?
Can a memory game raise Mathematics marks?
Can a brain-training website make a student generally “smarter” before an examination?
This page first appeared in November 2014 and recommended several online brain-game sites for students preparing for examinations.
The instinct was understandable.
If an examination uses the brain, perhaps exercising the brain with puzzles should improve examination performance.
The missing word is transfer.
Getting better at one cognitive task proves that learning happened. It does not automatically prove that the learning transfers to a different task.
This 2026 rebuild keeps the original “exercise your brain” idea, but changes what the exercise should mean.
Practice usually improves what is practised
If you practise Sudoku, you can become better at Sudoku.
If you practise a working-memory task, performance on that task—or closely related tasks—can improve.
If you practise typing, typing becomes faster.
This is ordinary learning.
The more ambitious claim is that practising one task produces broad improvement somewhere else.
That is called far transfer.
Far transfer is much harder to establish.
Brain training is not one single scientific result
Research in cognitive training is complicated.
A 2026 meta-analysis in npj Digital Medicine found moderate behavioural benefits from computerised working-memory training across the studies it analysed, with outcomes affected by task type, compliance, training dose and age.
That is real evidence that targeted cognitive training can change performance.
It does not mean every commercial brain game broadly raises school achievement.
APA discussion of brain-training research makes the same caution clear: people often improve on the tasks they practise, but the improvement may not transfer strongly to unrelated tasks.
Sources: npj Digital Medicine — 2026 working-memory training meta-analysis and American Psychological Association — brain training and transfer.
Near transfer and far transfer
Suppose a student practises remembering sequences of symbols.
There are several possible outcomes.
- Task learning: better at the exact trained sequence task.
- Near transfer: better at a similar untrained memory task.
- Far transfer: better at a substantially different activity such as algebra, essay writing or comprehension.
The farther the target moves from the trained task, the more mechanism we need to explain why transfer should occur.
Similarity creates a bridge. Without a bridge, transfer is a hope rather than a plan.
PSLE does not ask for generic brain power
PSLE asks students to perform specific language, mathematical and scientific tasks.
English requires vocabulary, grammar, reading, writing, listening and oral communication.
Mathematics requires concept knowledge, number fluency, representation, method selection and problem solving.
Science requires knowledge, inquiry, evidence interpretation and explanation.
A general puzzle may use attention or memory.
It does not contain the specific knowledge structures needed for those subjects.
The best brain exercise for an examination is often the examination’s underlying cognitive job
If the student struggles to retrieve Science concepts, practise retrieving Science concepts.
If the student struggles to select Mathematics methods, practise distinguishing Mathematics problem types.
If the student struggles to infer from English passages, practise inference from evidence.
This produces a shorter transfer distance.
Train close to the performance you need—then vary the surface enough to prevent memorising only one example.
Retrieval practice is a real cognitive workout
Reading notes is cognitively easier than producing the answer without notes.
Retrieval forces the learner to reconstruct.
- What is the definition?
- What are the steps?
- Why does the process happen?
- Which formula applies?
- What evidence supports this inference?
IES guidance recommends active retrieval and quizzing as tools for durable learning. IES — Organizing Instruction and Study to Improve Student Learning.
This is brain exercise with direct curricular content inside it.
Spacing makes the workout harder in a useful way
An answer retrieved ten seconds after reading is still supported by freshness.
An answer retrieved three days later requires more reconstruction.
Spacing allows some forgetting to occur, then asks the learner to recover the knowledge.
IES research summaries repeatedly identify spacing as a strong learning principle.
The important point is not to chase a perfect spacing formula.
Return more than once across time.
Interleaving trains discrimination
A worksheet containing ten nearly identical problems gives the student a strong cue about which method to use.
An examination removes that cue.
Interleaving mixes related problem types so the learner has to distinguish them.
This is especially useful in Mathematics after the underlying methods are already understood.
IES describes interleaved practice as requiring students to choose a strategy rather than repeat the same one automatically. IES — Study Effectively: Interleaving.
Explanation exercises a different layer
A student can recognise the correct answer without understanding why it is correct.
Ask for explanation.
- Why does this answer follow?
- Why are the other options wrong?
- Which evidence supports the conclusion?
- What would change the answer?
- Can you explain the concept without the textbook wording?
Deep questions expose whether the learner has a relationship or only a memorised label.
Representation switching is another useful cognitive exercise
Move the same idea between forms.
- words → diagram;
- diagram → equation;
- table → graph;
- graph → verbal explanation;
- story problem → model;
- paragraph → one-sentence summary.
The learner discovers what stays invariant when the representation changes.
IES also recommends connecting concrete and abstract representations and combining graphics with verbal descriptions.
Working memory matters—but reducing unnecessary load matters too
Students sometimes respond to working-memory difficulty by trying to “increase memory capacity” directly.
Another route is to reduce what working memory has to carry.
- Make number facts fluent.
- Write intermediate steps.
- Use diagrams.
- Chunk information.
- Learn vocabulary well enough that word meaning no longer consumes attention.
- Use consistent notation.
Fluency converts some effortful operations into faster, more stable routines.
This frees attention for the difficult part of the question.
Brain games can still be enjoyable
None of this means Sudoku, chess, crosswords, puzzles or memory games are useless.
They can be enjoyable.
They can provide challenge.
They can develop task-specific skill.
They can be social.
They can become hobbies worth having.
The correction is only this:
do not promise transfer that the activity has not demonstrated.
Chess is an excellent example
Strong chess players recognise patterns, calculate variations and manage time.
Those are sophisticated abilities.
A child becoming better at chess has genuinely learned.
It does not follow that algebra or composition marks must therefore rise.
The domains share some broad cognitive resources but contain different knowledge and representations.
Exam preparation needs task-specific automaticity
A student under time pressure cannot consciously reconstruct every low-level operation from first principles.
Some operations need fluency.
- multiplication facts;
- common fraction conversions;
- grammar patterns;
- high-frequency vocabulary;
- reading familiar notation;
- basic scientific relationships.
Fluency is not mindless learning.
It is freeing higher-level attention from repeatedly solving the same lower-level problem.
A useful “brain circuit” for Mathematics
- Retrieve one older concept.
- Solve two familiar examples.
- Mix the topic with two alternatives.
- Explain why the selected method fits.
- Try one unfamiliar wrapper.
- Return several days later.
This trains knowledge, selection, explanation, transfer and delayed retrieval in one subject-relevant circuit.
A useful “brain circuit” for Science
- Retrieve the scientific relationship.
- Explain it in plain language.
- Read a diagram or experiment.
- Separate observation from inference.
- Connect evidence to cause.
- Answer in concise examination form.
The task is cognitive training—but it trains the cognition Science actually requires.
A useful “brain circuit” for English
- Read a short text.
- Summarise it without looking.
- Infer one unstated idea from evidence.
- Retrieve two useful words from earlier study.
- Use them in new sentences.
- Explain one grammar or style choice.
Again, the exercise is close to the actual language system.
Difficulty has to be calibrated
A task that is impossible every time does not create useful practice.
A task that is automatic every time may no longer create enough adaptation.
The useful zone is challenging enough that the learner must retrieve, discriminate or reason—but not so difficult that the only behaviour being practised is failure.
Feedback is part of cognitive training
If a student practises the wrong reasoning repeatedly, the brain is still learning.
It is learning the wrong pattern.
Useful feedback identifies:
- where the first divergence occurred;
- why it matters;
- what the correct relationship is;
- how to recognise the same issue next time;
- when the repair will be retested.
Practice and feedback are a loop.
Sleep and physical state are part of the cognitive system
The brain is not a separate device sitting above the body.
Attention and memory operate through sleep, health, movement, nutrition and emotional state.
This does not require turning every lifestyle habit into an optimisation programme.
It means a revision plan that depends on chronic sleep loss is fighting its own learning system.
Metacognition may be the most useful general “brain training”
Students improve when they can judge their own state more accurately.
- I recognise this but cannot retrieve it.
- I can do it only when the method is labelled.
- I know the concept but cannot explain the evidence.
- I can solve it but I take too long.
- I keep making the same checking error.
IES research summaries describe feedback-driven metacognition alongside retrieval, spacing and interleaving as a powerful teaching approach.
The student learns not only the subject.
The student learns how their own performance fails.
What happened to the three 2014 brain-game websites?
The original page recommended BrainMetrix, Games for the Brain and BrainHQ.
We no longer make them the centre of this guide.
Websites change, commercial offerings change and the scientific question is bigger than any product.
If a family uses any cognitive-training app or game, ask:
- What exact skill is being trained?
- What evidence exists for transfer?
- What age group was studied?
- What outcome improved?
- How large was the improvement?
- Would subject-specific practice produce a more direct return for the current goal?
Do not confuse engagement with transfer
A game can be beautifully designed.
A child may spend twenty focused minutes on it.
That proves the game is engaging.
It does not prove PSLE performance will improve.
Engagement is a delivery advantage.
Transfer is an outcome question.
Do not confuse neuroscience language with evidence
Products can sound scientific because they use terms such as:
- neuroplasticity;
- working memory;
- cognitive speed;
- brain fitness;
- executive function.
Those are real concepts.
The presence of a scientific word does not prove the product’s specific claim.
Ask for the actual study and outcome.
A better exam-preparation hierarchy
- Repair missing subject knowledge.
- Retrieve it after delay.
- Mix related question types.
- Explain and switch representations.
- Transfer to unfamiliar contexts.
- Practise timed execution.
- Use general puzzles as recreation or supplementary challenge—not as a substitute for the above.
Frequently asked questions
Do brain-training games work?
They can improve performance on trained or related cognitive tasks. Broad transfer to unrelated academic performance is much less automatic and depends on the training and outcome being studied.
Is Sudoku good for the brain?
Sudoku is a legitimate puzzle that trains Sudoku-relevant pattern and constraint solving. Enjoy it if you like it. Do not assume it substitutes for Mathematics or language study.
Should children stop playing thinking games before exams?
No. Recreation and challenge can remain part of a balanced routine. The issue is opportunity cost: do not displace high-value subject repair with a game because the game is marketed as general brain training.
What is the best way to exercise memory for exams?
Use active retrieval of the actual knowledge needed, spaced across time, with feedback and later transfer checks.
Can working-memory training help?
Research, including a 2026 meta-analysis, reports measurable benefits in computerised working-memory training. The practical question is how much those gains transfer to the specific academic outcome a student needs.
So how should students exercise their brains for exam preparation?
Not by searching for one magical cognitive game.
Exercise the functions that the exam actually needs.
Retrieve.
Discriminate.
Explain.
Represent.
Transfer.
Execute.
Reflect on the error.
Return later.
The brain adapts to what it repeatedly has to do. Make the practice resemble the capability you want to keep.
Sources and related eduKate guides
- npj Digital Medicine — Meta-analysis of computerised working-memory training, 2026
- American Psychological Association — Brain training and transfer
- IES — Organizing Instruction and Study to Improve Student Learning
- IES — Retrieval, spacing, interleaving and feedback-driven metacognition
- eduKate — Top 5 Tips to Study for PSLE
Historical note: first published in November 2014 as a list of online brain-training websites for exam preparation. Rebuilt in 2026 to preserve the idea of cognitive exercise while making transfer explicit and redirecting exam preparation toward subject-relevant retrieval, spacing, interleaving, explanation and execution.