Quick Read: Brain games can make people better at the games or tasks they practise. The harder question is whether that improvement transfers to school outcomes such as reading, Mathematics, Science, writing or examination performance. Research on working-memory and cognitive-training programmes has repeatedly found that near transfer is easier to demonstrate than far transfer. If the goal is an examination, the most reliable preparation remains learning the subject, retrieving it, receiving feedback and practising the actual forms of reasoning and execution the exam requires.
One-sentence answer: Sudoku, memory games and cognitive-training apps may be enjoyable and may improve closely related tasks, but students should not assume they will automatically raise PSLE, O-Level, IB or other exam scores.
Why this 2014 article needed rebuilding
The original post recommended several “brain-training” websites for students preparing for examinations. That reflected a popular idea of the time: train general cognitive abilities and school performance should improve downstream.
The problem is that transfer is not automatic. Becoming faster at one cognitive task may mainly mean becoming better at that task.
This updated page therefore removes stale website recommendations and asks the more important educational question: what kind of training transfers, how far, and under what conditions?
Near transfer versus far transfer
Near transfer occurs when practice improves performance on a very similar task. If you repeatedly practise remembering sequences, you may improve on another sequence-memory task.
Far transfer would mean the training improves a substantially different outcome—for example reading comprehension, Mathematics attainment, intelligence-test performance or exam marks.
The further the target task is from the trained task, the stronger the evidence needs to be before claiming transfer.
What large reviews have found
A major meta-analytic review of 87 publications and 145 experimental comparisons found that working-memory training could improve performance on closely related working-memory tasks, but found no convincing evidence of reliable far transfer to verbal ability, non-verbal ability, word decoding, reading comprehension or arithmetic when compared with active control conditions.
That does not prove that every conceivable cognitive-training programme can never transfer. It does mean parents should be cautious when a product claims that practising one narrow game will broadly increase intelligence or school achievement.
Why brain games can feel more powerful than they are
- Scores rise visibly inside the game.
- Difficulty adapts, so the learner feels challenged.
- Practice effects appear quickly.
- The interface may label progress as “memory”, “attention” or “brain speed”.
- The trained task can resemble a laboratory measure of cognition.
All of this can be real without proving that the improvement will appear in a Science explanation or Mathematics problem.
Being good at a task is still valuable
None of this means puzzles are useless. Sudoku, chess, logic games, memory challenges and strategy games can be worthwhile because they are enjoyable, demanding activities in their own right.
The correction is about the claim. “This is an interesting puzzle that practises planning” is different from “this will raise your Mathematics score”.
Why subject-specific learning transfers better to subject exams
Examinations reward specific knowledge and specific forms of reasoning. A Mathematics paper requires mathematical concepts, representations and procedures. English requires vocabulary, grammar, reading, writing and communication. Science requires domain knowledge plus interpretation and causal explanation.
So the training target should resemble the outcome at the level that matters.
- To improve inference, practise reading and justifying inference.
- To improve algebraic selection, practise choosing among algebraic methods.
- To improve Science explanations, practise connecting evidence to scientific cause.
- To improve exam timing, practise realistic papers and analyse where time is lost.
Retrieval practice is different from generic brain training
Retrieval practice asks the student to recover the actual knowledge they need later. The subject content is not incidental—it is the object being strengthened.
Examples include:
- recalling a formula and explaining when it applies;
- reconstructing a Science process;
- answering vocabulary questions without notes;
- planning a composition from memory;
- solving a previously learned method after a delay.
Because the retrieved knowledge is the knowledge needed later, the pathway to the target is much clearer.
Practice still needs feedback
Repeatedly attempting a task without checking can strengthen the wrong response. Good practice includes correction.
- What did I get wrong?
- Was the knowledge missing?
- Did I choose the wrong method?
- Did I misunderstand the question?
- Was the final execution careless?
The answer determines the next practice.
What about attention?
Attention matters enormously for learning, but training attention inside one game does not guarantee broad academic improvement. Students can often improve school attention more directly by changing the environment and task structure.
- remove avoidable phone interruptions;
- work in defined blocks;
- make the task explicit;
- use active rather than passive study;
- sleep adequately;
- break difficult work into manageable stages.
What about working memory?
Working memory is relevant whenever students have to hold and manipulate information. But one practical response is to reduce unnecessary load rather than assuming capacity can simply be expanded through games.
- Automate basic facts where useful.
- Chunk related information.
- Write intermediate steps down.
- Use diagrams.
- Teach prerequisite knowledge first.
- Reduce irrelevant task complexity during initial learning.
These methods support the task the learner actually needs to perform.
Puzzles can still support useful habits
A good puzzle can encourage persistence, checking, representation and strategy change. Those habits may become useful elsewhere if the learner consciously recognises and reuses them.
But that transfer should be demonstrated, not assumed. Ask the student to articulate what principle from the puzzle appears in a school task.
Chess is not automatically Mathematics tuition
Chess involves planning, pattern recognition and constrained decision-making. Mathematics also involves reasoning. The overlap does not mean chess practice directly teaches fractions, algebra or geometry.
Activities can share general features without being educational substitutes.
What students should do if the exam is the goal
- Identify the actual knowledge and skill tested.
- Learn weak prerequisites.
- Practise retrieval after delays.
- Mix question types so method selection becomes necessary.
- Use changed examples to test transfer.
- Practise realistic exam formats.
- Analyse errors and repair the cause.
- Protect sleep and recovery.
This is less glamorous than “train your brain”, but much closer to the required outcome.
When a brain-training product makes a big claim
- What exact outcome improved?
- Was it the trained game or a genuinely different skill?
- Was there an active control group?
- Did the improvement last?
- Was academic attainment measured?
- Was the study independent of the product company?
- How large was the effect?
Near transfer is not failure
Near transfer can be completely useful when the trained task is itself valuable. Typing practice should improve typing. Mental arithmetic practice should improve mental arithmetic. Reading fluency practice should improve reading fluency.
The problem appears only when a narrow improvement is marketed as a universal cognitive upgrade.
Parent decision rule
If a child enjoys a puzzle game and it fits family screen-time boundaries, it can be recreation. If the parent is paying for it specifically to raise exam scores, demand evidence that measures the exam-relevant outcome.
Common misconceptions
- “A higher brain-game score means higher intelligence.” It may mainly reflect practice on the game.
- “Working-memory training reliably raises school achievement.” strong far-transfer evidence is lacking.
- “Puzzles are useless if they do not raise marks.” They can still be enjoyable, challenging activities.
- “Subject study is rote while brain training builds thinking.” High-quality subject learning itself develops reasoning within real knowledge domains.
- “General cognitive skill can replace knowledge.” Experts reason with rich domain knowledge, not empty processing power alone.
Related eduKatePunggol guides
- Understanding Lag Time in Studies — spacing, retrieval and transfer.
- Top 5 PSLE Study Tips — practical study behaviour.
- How to Build Discipline in a Student — self-regulation.
Evidence note
A widely cited meta-analysis by Melby-Lervåg, Redick and Hulme reviewed 87 publications with 145 experimental comparisons and found short-term improvements on trained or closely related working-memory measures but no convincing reliable far-transfer improvements in important real-world cognitive outcomes when compared with treated controls. This article uses that evidence to set a cautious boundary around broad “brain training” claims.
Updated from eduKatePunggol’s 2014 brain-game links. The old websites have been removed as recommendations; the article now teaches students and parents how to distinguish task practice, near transfer and far transfer.
