PSLE Composition Brainstorming | Generate → Filter → Connect → Plan

Brainstorming is useful only if it helps the student choose a composition they can actually finish.

Generating ten ideas is not the goal. The goal is to move from an open prompt to one coherent route quickly enough that there is still time to write, develop and review the composition.

generate → filter → connect → plan → stop brainstorming → write

Start with the Actual PSLE Task

In the 2026 PSLE English Continuous Writing task, students write at least 150 words on a given topic and base the response on at least one of three pictures.

That means brainstorming should begin with four constraints:

What is the topic really asking?
Which picture or combination provides a useful route?
What story or response can be developed coherently?
Can the student complete it within Paper 1’s shared 1 hour 10 minute window?

The pictures are stimuli, not a demand to include everything shown.

Stage 1: Understand Before Generating

Students often brainstorm too early. They see one picture and immediately invent a dramatic plot.

Before generating, write three short notes:

Topic: what central situation, quality or experience must remain relevant?
Picture: what does the selected image clearly show—and what does it merely suggest?
Possible story job: problem, choice, discovery, misunderstanding, relationship, challenge or consequence?

This reduces the risk of developing an imaginative story that has drifted away from the prompt.

Stage 2: Generate Several Routes

Now generate three to five possible routes. Use whichever method creates ideas fastest for this student.

Method	Best when	Main risk
Listing	student thinks in quick options	list remains disconnected
Clustering / mind mapping	student needs to see relationships	map grows without selection
Starbursting / 5W1H	story lacks depth or causality	too many questions consume time
Freewriting	student is blocked or over-edits too early	raw writing becomes the final plan without filtering

No method is required by SEAB. These are planning tools, not marking criteria.

Stage 3: Filter the Ideas

Every idea should pass five tests.

Relevance: does it genuinely answer the topic?
Picture connection: can the selected image fit naturally?
Causality: can one event cause the next?
Manageability: can the story be completed and developed within exam conditions?
Language readiness: does the student have enough language to express the route clearly?

A spectacular plot that needs six locations, ten characters and a surprise ending may be less useful than a smaller story with clear pressure and consequence.

Stage 4: Connect the Story

Once one route is selected, connect it before drafting.

character goal → pressure → decision → consequence → changed state

Ask:

What does the main character want?
What makes that goal difficult?
What information changes the situation?
What choice does the character make?
What consequence follows from that choice?
What is different by the end?

This is the point where brainstorming becomes planning.

Stage 5: Build a Minimal Plan

A useful exam plan can be very short.

Scene	Planning note
Opening	normal situation + first pressure
Development	pressure increases / new information
Turning point	character must choose
Consequence	choice changes the situation
Ending	new state / reflection if earned

The plan should help the student remember the causal route. It should not become a second composition written in note form.

The Stop Rule

Brainstorming has diminishing returns.

Stop when the student can answer:

What is my story about?
Which picture am I using?
What does the character want?
What pressure changes the situation?
What is the turning point?
How will the ending follow from it?

Once the route is coherent, additional idea generation can become procrastination.

When Listing Works Best

Listing is useful for a student who can generate possibilities quickly but needs a disciplined filter afterwards.

Example topic: An Unexpected Discovery

lost wallet under a bench;
old letter inside a library book;
injured bird behind the school hall;
friend’s hidden mistake;
unlabelled box in a storeroom.

Then score each idea for relevance, causality, development and manageability.

When Clustering Works Best

Clustering helps when ideas exist but their relationships are unclear.

A centre node such as lost wallet can branch into:

owner;
money/cards;
temptation;
friend’s advice;
security office;
consequence of keeping or returning it.

The map becomes useful when the student draws cause-and-effect links instead of merely adding more bubbles.

When Starbursting Works Best

Starbursting is useful when a story is thin.

Who wants something?
What changes?
When does the pressure become urgent?
Where can the key event happen naturally?
Why does the character choose this route?
How does the consequence follow?

Do not answer every possible 5W1H question. Use questions that expose missing causal information.

When Freewriting Works Best

Freewriting can help a student who freezes because every first sentence feels inadequate.

Write freely for two or three minutes, then mine the paragraph for:

a useful character goal;
a conflict;
a surprising consequence;
a setting detail that matters;
one line of dialogue with narrative pressure.

The raw freewriting is not the composition. It is material to select from.

A Three-Minute Exam Brainstorm

30 sec: interpret topic and pictures.
60 sec: generate three routes.
45 sec: filter for relevance and causality.
45 sec: write goal → pressure → choice → consequence.

This is a practice target, not an official timing rule. Students should test their own planning speed inside full Paper 1 practice.

How Tuition Can Teach Brainstorming

compare three possible routes from the same prompt;
reject ideas that drift from the topic;
repair stories with no cause-and-effect chain;
time the planning process only after the method is understood;
let different students use different generation methods;
reduce planning scaffolds as students become faster and more independent.

At eduKateSG, our typical group is up to three students. One learner may prefer a quick list, another a map, and another a questioning method. The final standard is the same: a coherent plan the student can execute independently.