Comprehension errors sometimes happen because the strategy from the previous question leaks into the next one. A child uses inference when the new question is literal, keeps searching for evidence when the task asks for vocabulary, or continues a comparison routine after the comparison is over. Strong readers reset between questions.
This page uses Fernvale Primary School only as the school-family context named in the preserved 2019 URL. Its reader job is question-to-question reset: stop the previous strategy from leaking into the next item.
The old page contained historic Yishun centre, tutor, telephone and generic syllabus claims. Those are not carried forward. eduKate Singapore is an independent tuition provider and is not affiliated with Fernvale Primary School, MOE or SEAB.
Every Question Starts Fresh
Before answering the next item, the student should briefly clear the previous question state and ask:
- What does this new question ask?
- What kind of answer is required?
- Where should I search?
- Do I need stated information, evidence, comparison or inference?
Strategy Leakage Is Common in Mixed Sets
| Previous question | New question | Possible leakage |
|---|---|---|
| Inference | Literal | Student over-interprets |
| Evidence | Vocabulary | Student searches too widely |
| Comparison | Single-character motive | Student keeps writing both sides |
| Cause | Sequence | Student invents a reason instead of stating what happened next |
The Reset Routine
STOP → READ NEW QUESTION → NAME NEW JOB → CHOOSE NEW STRATEGY → ANSWER.
- Release the previous answer.
- Read the next question from the beginning.
- Name its family.
- Build the answer contract.
- Search with the appropriate evidence radius.
Question-Family Labels During Training
- L = literal;
- V = vocabulary;
- I = inference;
- E = evidence;
- C = comparison;
- R = reference.
Students can mark the family briefly during practice, then fade the notation once switching becomes automatic.
Reset After a Difficult Question Too
A hard item can contaminate the next one emotionally as well as strategically. Teach the student to treat the next question as a new local task rather than continue replaying uncertainty from the previous item.
Reset the Evidence Radius
The previous question may have needed several paragraphs. The next may need one sentence. Do not inherit the same search radius automatically.
Reset the Answer Length
A two-part answer may require more writing. The next item may ask for one precise word or phrase. Length should follow the answer contract, not the previous response.
A Strategy-Leak Diagnostic
| Observed pattern | Likely issue |
|---|---|
| Several errors occur in a row | No reset after difficult item |
| Inference appears in literal answers | Strategy leakage |
| Every answer has similar length | Answer-contract leakage |
| Search radius stays broad | Evidence-search leakage |
Use Mixed Mini-Sets
Practice with short sets that deliberately alternate question families. The student should name the job before answering and explain what changed from the previous item.
How a 3-Pax Class Trains Resets
In a maximum three-student group, learners can classify each new question before solving. The tutor can see whether the wrong answer began with a failure to reset or with a later comprehension step.
Current MOE English Context
The MOE Primary English Language Syllabus 2020 develops flexible selection and application of reading strategies. Question-to-question reset is one practical way to prevent a previously useful strategy from becoming the wrong strategy on the next item.
School Context Boundary
This article uses Fernvale Primary School only as the school-family context named in the preserved URL. It does not imply endorsement, partnership or affiliation.
Signs Reset Control Is Improving
- Question families are recognised quickly.
- One difficult item causes fewer follow-on errors.
- Evidence radius changes appropriately.
- Answer length matches the task.
- Mixed-set accuracy becomes more stable.
Question-to-Question Reset: Almost-Code Summary
AFTER each_question:
release_previous_strategy()
NEW_QUESTION:
identify_family()
build_new_answer_contract()
choose_new_evidence_radius()
IF previous_item_was_difficult:
reset_attention()
OUTPUT:
less_strategy_leakage
stronger_mixed_set_accuracy
better_comprehension_control