PSLE Science questions that ask “What happens if…?” are testing whether a pupil can change one condition mentally and follow the scientific consequences. If a switch opens in a simple circuit, the conducting path is broken and a bulb on that path cannot light. If a green plant receives no light, photosynthesis cannot manufacture sugar using light energy, though the living plant can still respire. If one prey species declines, a predator with alternative food may be affected but is not guaranteed to disappear.
The most important technique is to keep track of the before state, the changed condition and the after effect. A prediction is more reliable when it identifies a mechanism, respects the given assumptions and distinguishes an expected outcome from a guaranteed one. The student should not invent further changes that the question never described.
This guide contains thirty original predictive questions, each with the exact condition changed, a model answer and a contrasting variation. It covers Primary 3–6 topics across Diversity, Cycles, Systems, Energy and Interactions. These are fresh teaching cases rather than official past-paper questions. The published SEAB Standard Science 0009 format verified here is the revised paper examined from 2026. Check any 2027 subject-format update on SEAB’s site.
- Before: describe the initial condition clearly.
- Change: identify the one factor deliberately altered or removed.
- Mechanism: state the scientific process responding to the change.
- After: describe what is expected in the stated model.
- Limits: do not predict exact outcomes without adequate evidence.
- Compare: use “greater”, “less”, “faster”, “slower”, “remains” or “no longer” precisely.
- Conservation: matter and energy are not created from nothing merely because the device or process changes.
- Retest: reverse the changed condition and check whether the pupil can reverse the prediction appropriately.
Official basis: MOE Primary Science syllabus: predicting, explaining and applying knowledge; SEAB examination format examined in 2026. The scenarios in this article are original and do not forecast the 2027 examination.
Quick navigation
- How to make an accurate what-if prediction
- What may, must and cannot mean in Science
- Heat, matter, light, force and circuits: ten changes
- Plants and human systems: ten changes
- Food webs, habitats and investigations: ten changes
- Practice reversing a condition
- Why many predictions fail
- Four-week revision timetable
- FAQ and links
1. The four-line prediction framework
Start with the sentence “If X changes while the stated other conditions remain the same, Y is expected because Z.” This is a useful scaffold, not a fixed examination sentence. X identifies the altered condition. Y states the likely outcome. Z supplies the scientific mechanism. If a conclusion is not guaranteed, add the relevant limitation.
For a simple circuit, X could be opening the only switch. Y is that the bulb goes off or does not light. Z is that the conducting path becomes incomplete. The condition “only switch in a single-loop circuit” matters. In a parallel circuit, opening a switch in one branch need not affect another complete branch.
For evaporation, X could be increasing exposed water surface area while temperature and airflow remain comparable. Y is a faster rate of evaporation under the tested range. Z is that the liquid has more surface exposed to the surrounding air. The prediction cannot be carried over automatically to a case in which the container with the greater surface area also sits in a colder, more humid or differently ventilated environment.
For a plant, X might be removing light. Y is that photosynthesis stops under the dark conditions, because the process requires light energy. The child must not write that every living process stops. Respiration continues in living cells, so the prediction applies to a specific process rather than to all life.
For a food web, X might be reducing a particular prey population. Y may involve less food for a predator. Z is that the predator depends on the prey for energy. But the predator may have alternative food, and the exact population change cannot be predicted without more information. This is where “may” is scientifically appropriate.
A good what-if answer names the effect that changes, but also the features that remain. A torch bulb that loses its electrical supply no longer produces light from that supply; the bulb itself does not necessarily cease to exist, and the battery’s stored energy does not automatically vanish from the world. A seed dispersed by wind moves away from the parent plant; germination still requires suitable conditions.
The pupil should also check whether the question asks to predict a numerical change or a qualitative outcome. If the data give two temperatures and ask for a decrease, calculate it. If the model only shows that a cup with one wrapping cooled less, avoid inventing the exact final temperature after a new wrapper is added.
The before–after framework is especially powerful for multi-step changes. A change in light can affect photosynthesis, which may affect plant growth, which can alter food availability for consumers in a simplified food web. State each supported link. Do not jump from less sunlight to “every organism dies” without considering the other conditions.
The method trains flexible understanding because the child can reverse it. If closing a switch completes a path, opening it interrupts that path. If adding a fan increases drying rate in comparable conditions, removing the fan may reduce airflow and slow drying. The exact outcome still depends on the setup, but the causal logic remains recognisable.
| Question part | What to look for | Illustrative answer |
|---|---|---|
| Before | Initial configuration or value | Bulb is in a complete loop |
| Change | What the stem alters | Switch is opened |
| Mechanism | How the system responds | Conducting path is broken |
| After | Expected result | Bulb no longer lights |
| Limit | Important missing or alternative condition | Other branches may still work in parallel |
2. Which language makes a prediction scientifically accurate?
Some outcomes are logically forced by the simplified system described. In a one-loop circuit with the only conducting path opened, the bulb cannot receive current along that path. Other predictions depend on biological variation or environmental conditions. A seed may fail to germinate without water, but giving water does not guarantee that every seed grows because viability, oxygen and temperature still matter.
The word “will” may be suitable when the model explicitly fixes all relevant conditions and the process is deterministic at the teaching level. “May” or “is likely to” is more suitable when the information allows alternative outcomes. Do not use cautious wording everywhere as a way to avoid reasoning. The pupil should match certainty to evidence.
“Compared with” directs attention to a baseline. A plant receiving less water may grow less under suitable comparable conditions, but if the question gives actual measured plant heights, use them to describe the observed difference. Avoid replacing data with a vague prediction when the question asks for a specific calculation.
“Only” and “all” create stronger claims. If two prey species are available, removing one does not prove every predator must die. If an experiment changes light and water simultaneously, it cannot establish that the result came only from light. The child should be trained to see what conditions protect or weaken a claim.
“Most likely” is not “certain”. “Cannot conclude” is not “definitely false”. A possible prediction can be scientifically useful without being guaranteed. These distinctions make Science explanations more accurate and reduce the temptation to write dramatic absolute outcomes.
For numeric change, “decrease” refers to final versus starting value, “rate” refers to change per time, and “extension” refers to added length. A wrong quantity can make a correct scientific prediction unusable. Check labels and units before choosing the direction of change.
3. Ten physical-system changes: matter, heat, light and electricity
Physical prediction 1: Open switch
Before and change: A bulb is on in a simple single-loop circuit. The only switch is opened.
Predicted effect and mechanism: The bulb goes off because the conducting loop is interrupted.
What the prediction does not establish: The battery may still store energy; the bulb need not be broken.
Reverse or vary it: What happens if that switch closes again with all other parts working?
Physical prediction 2: Open one parallel branch
Before and change: Two bulbs are in separate complete parallel branches. Branch A opens.
Predicted effect and mechanism: A’s bulb stops lighting; B can stay lit if its own path remains complete.
What the prediction does not establish: A series-circuit rule cannot be applied to all arrangements.
Reverse or vary it: What happens if the common battery connection rather than only A opens?
Physical prediction 3: Increase surface area
Before and change: Two comparable amounts of water are exposed under the same conditions; one tray area increases.
Predicted effect and mechanism: Evaporation tends to be faster with more exposed liquid surface.
What the prediction does not establish: The exact volume lost still depends on time, temperature and other conditions.
Reverse or vary it: What happens when the larger tray is placed somewhere colder?
Physical prediction 4: Cool a sealed bottle
Before and change: A closed cold bottle is moved into moist room air.
Predicted effect and mechanism: Water vapour near its cool exterior may condense into droplets.
What the prediction does not establish: The droplets generally originate outside, not by passing through the wall.
Reverse or vary it: What if the bottle warms to match the air and no longer cools it sufficiently?
Physical prediction 5: Add insulating wrap
Before and change: Identical cups begin at the same temperature; one receives a better thermal insulator.
Predicted effect and mechanism: The better-insulated cup generally has a smaller temperature fall over the comparable period.
What the prediction does not establish: The wrapper does not create heat.
Reverse or vary it: What if water volumes differ as well?
Physical prediction 6: Move card toward lamp
Before and change: A card moves nearer a small lamp while the screen stays fixed.
Predicted effect and mechanism: Its shadow usually becomes larger due to the spread of blocked light.
What the prediction does not establish: The physical card does not grow; the result is a projection.
Reverse or vary it: What if the card instead moves towards the screen?
Physical prediction 7: Add a light-blocking panel
Before and change: A panel that blocks most light is placed in front of a bright lamp.
Predicted effect and mechanism: Less direct light can reach the region behind the panel; a shadow may form.
What the prediction does not establish: Other light sources or partially transmitting material may alter appearance.
Reverse or vary it: What if the panel is replaced by clear glass?
Physical prediction 8: Change sliding surface
Before and change: The same block moves over a rougher surface with the same release.
Predicted effect and mechanism: Its travel distance may decrease because friction resists sliding.
What the prediction does not establish: The outcome depends on materials and comparable conditions.
Reverse or vary it: What if the surface becomes smooth but the launch push is also reduced?
Physical prediction 9: Stretch spring
Before and change: The same elastic spring receives a larger suitable load.
Predicted effect and mechanism: Its extension may increase over the tested elastic range.
What the prediction does not establish: Do not assume unlimited extension without damage.
Reverse or vary it: What if the original length is different for another spring?
Physical prediction 10: Remove a battery from a suitable series supply
Before and change: A working low-voltage classroom circuit with batteries in series loses one source cell.
Predicted effect and mechanism: The supply arrangement changes; bulb performance may change, and it may not operate if the conducting path is broken.
What the prediction does not establish: Exact brightness depends on cell arrangement and circuit, so the diagram matters.
Reverse or vary it: What if the cell is replaced with an appropriately connected one?
4. Ten biological-system changes: plants, animals and the human body
Living-system prediction 11: Remove light from green plant
Changed condition: A watered green plant has access to carbon dioxide but is placed in complete darkness.
Scientific prediction: Photosynthesis cannot make sugar through the light-dependent process without light.
Important limit: The plant can continue respiration while alive.
New question: What changes when suitable light is restored?
Living-system prediction 12: Block plant water transport
Changed condition: Water-carrying tubes to a group of leaves are damaged.
Scientific prediction: Those leaves may receive insufficient water for normal functions and photosynthesis.
Important limit: The exact outcome depends on other pathways and plant conditions.
New question: What if food-carrying instead of water-carrying tubes are disrupted?
Living-system prediction 13: Block manufactured food pathway
Changed condition: A specific pathway carrying sugar from leaves towards roots is interrupted.
Scientific prediction: Roots may receive less food and their functioning may eventually be affected.
Important limit: The affected part and duration depend on the given anatomy and extent of damage.
New question: What if only the path towards one branch is damaged?
Living-system prediction 14: Reduce water for germination
Changed condition: Similar viable seeds receive no water while comparable controls receive a suitable amount.
Scientific prediction: The water-deprived group is less likely to germinate normally.
Important limit: Water alone does not guarantee every watered seed will germinate.
New question: What if both groups get water but have very different temperatures?
Living-system prediction 15: Remove pollinator access
Changed condition: A flowering plant ordinarily receives insect visits, but visits are prevented in a simplified setup.
Scientific prediction: Pollen transfer by those insects may decrease, potentially affecting reproduction.
Important limit: Other pollination mechanisms may remain, so seed failure is not guaranteed without more detail.
New question: What if wind or another insect still transfers pollen?
Living-system prediction 16: Add protective cover to animal
Changed condition: An animal has a hard shell protecting vulnerable body parts against some threats.
Scientific prediction: The structure may reduce vulnerability in the described habitat.
Important limit: No defence guarantees safety from every predator.
New question: What if a predator can overcome the shell?
Living-system prediction 17: Increase physical activity
Changed condition: A student changes from sitting to running.
Scientific prediction: Active muscles need energy; breathing and circulation can increase to support oxygen transport and carbon dioxide removal.
Important limit: Exact heart rate and medical implications cannot be inferred from the story alone.
New question: What happens when activity returns to rest?
Living-system prediction 18: Digestive absorption disrupted
Changed condition: A hypothetical digestive system absorbs fewer usable nutrients from food.
Scientific prediction: Body cells may receive fewer transported nutrients for their functions.
Important limit: A specific disease cannot be diagnosed from this simplified scenario.
New question: What if absorption returns while blood transport remains impaired?
Living-system prediction 19: Reduce food for herbivore
Changed condition: A garden’s green plants decline while insects feed on the plants.
Scientific prediction: Some plant-eating insects may have less food, affecting survival or reproduction.
Important limit: Alternative plants and other conditions can alter the effect.
New question: What if different suitable plants become available?
Living-system prediction 20: Remove nesting tree
Changed condition: An area loses the trees where certain birds nest.
Scientific prediction: Suitable nesting sites may become less available and reproduction may be affected.
Important limit: The birds may find other suitable habitats; exact numbers are unknown.
New question: What if new suitable nesting sites are restored?
5. Ten environment and investigation predictions
Environmental and inquiry prediction 21: Reduce prey with alternatives
Changed condition: A snake has two prey species, mice and frogs. Mice become less common.
Expected outcome: Food from mice declines, but frogs remain a possible resource; total snake numbers cannot be fixed from the web alone.
Limit or trap: Do not claim certain extinction of the snake.
Check through a new condition: What changes if both listed prey species become scarce?
Environmental and inquiry prediction 22: Remove producer
Changed condition: A pond’s green aquatic plants decrease, and small animals feed on them.
Expected outcome: Those animals may have less food, with indirect effects on consumers further along the chain.
Limit or trap: Other producers or food sources could matter.
Check through a new condition: What if alternative aquatic producers increase?
Environmental and inquiry prediction 23: Introduce pollution
Changed condition: A river section receives harmful substances in a teaching scenario.
Expected outcome: Some aquatic organisms may face reduced survival or reproduction depending on pollutant and exposure.
Limit or trap: Exact species counts are not provided.
Check through a new condition: What data would help evaluate harm?
Environmental and inquiry prediction 24: Restore vegetation
Changed condition: Suitable native plants are restored to a damaged habitat.
Expected outcome: Food or shelter may become available to some organisms under appropriate conditions.
Limit or trap: Recovery cannot be guaranteed for every species.
Check through a new condition: What if the water supply remains unsuitable?
Environmental and inquiry prediction 25: Change two variables
Changed condition: Two plant groups vary in both water and light.
Expected outcome: Different growth is possible, but the method cannot isolate either cause.
Limit or trap: Repeating the confounded test does not make it fair.
Check through a new condition: How would you isolate the effect of light?
Environmental and inquiry prediction 26: Add repeats
Changed condition: An already fair towel test is performed several more times with comparable samples.
Expected outcome: Additional measurements can reveal consistency or unexpected variation.
Limit or trap: Repetition does not ensure perfect accuracy.
Check through a new condition: What if the scale has a consistent calibration error?
Environmental and inquiry prediction 27: Change survey duration
Changed condition: A bird population is surveyed for ten minutes in one park and one hour in another.
Expected outcome: Raw sightings may differ partly because observation effort differs.
Limit or trap: Counts cannot directly establish relative abundance without method controls.
Check through a new condition: What if both parks are observed for equal durations and comparable times?
Environmental and inquiry prediction 28: Increase temperature in evaporation test
Changed condition: Identical water trays are held under different controlled temperatures.
Expected outcome: Higher temperature generally increases evaporation rate when other relevant conditions are comparable.
Limit or trap: The effect is not necessarily identical for all temperatures or humidity levels.
Check through a new condition: What if airflow also changes?
Environmental and inquiry prediction 29: Compare two readings at unequal intervals
Changed condition: A cloth loses 10 mL in 10 minutes; another loses 12 mL in 40.
Expected outcome: The first has greater average water-loss rate (1 versus 0.3 mL/min).
Limit or trap: The greater total loss by the second is not evidence of faster evaporation.
Check through a new condition: What if both measurements last 40 minutes?
Environmental and inquiry prediction 30: Overload a spring
Changed condition: A spring is stretched far beyond a suitable elastic range.
Expected outcome: It may be permanently deformed and fail to recover its original length.
Limit or trap: Primary investigations should not assume ideal spring behaviour under damage.
Check through a new condition: What if loads are reduced to a safe elastic range?
6. Turn one question into three without memorising the answer
A good tutor can use one original scenario in three directions: predict the effect of adding something, removing it and changing a related factor. This is especially useful when a student has memorised a model answer but cannot explain it.
For a lamp in a single-loop circuit, the pupil first predicts what happens when the switch opens. Then the switch closes; the pupil explains why the path is complete again, assuming the source and components work. Finally the teacher adds a second parallel branch and asks whether the same simple rule still applies. The change reveals whether the child understands topology rather than only the word “open”.
For plants, begin with enough light for photosynthesis. Reduce light and predict the possible effect on sugar production under otherwise suitable conditions. Then restore light but reduce water. The pupil must recognise that photosynthesis depends on several necessary inputs; no single resource guarantees unlimited production.
For a cooling cup, ask which one has the greater temperature decrease, then which ends at a higher temperature, and then what happens if starting volumes differ. These are different questions using similar numbers. The child must respond to the quantity and condition asked.
For seed dispersal, first ask how a winged seed may travel in wind, then ask whether it must germinate after landing, and finally ask what happens if the landing site lacks suitable water. This connects structural adaptation to a separate life-cycle requirement.
For a food web, identify a predator’s prey, remove one prey, and then remove its alternative prey. The first change may leave another food source; the second has a different effect. The model should lead to reasoned possibilities rather than a fixed species count.
Always distinguish hypothetical prediction from measured evidence. If the question does not include a table of outcomes, do not write as though the experiment has already been performed. If it does include data, use the measured results and explain them within the method’s limits.
When a pupil finds a what-if question difficult, ask what has stayed unchanged. The phrase “all other conditions remain constant” is a powerful clue; it protects the comparison. Without it, the student should identify important missing factors before making a strong causal claim.
7. Five mistakes that make a sensible prediction wrong
1. Changing more than the question changes
A child invents a weak battery, different water supply or new predator not supplied by the diagram. Keep the stated conditions and ask only what follows from the named change.
2. Using a universal rule without checking the setup
The statement that every bulb turns off when one branch opens ignores parallel paths. Trace the actual connection.
3. Confusing a prediction with evidence
Expected outcomes from known science are not measured observations unless the question provides results. Signal which you are giving.
4. Overconfident ecological outcomes
When organisms have alternative food or habitat, a decline in one resource may matter but does not guarantee exact population changes.
5. Answering the wrong quantity
A higher final temperature is different from a larger temperature decrease, and a greater total loss is different from a faster rate.
8. Four-week what-if practice calendar
| Week | Practice focus | Independent check |
|---|---|---|
| 1 | Simple physical changes: heat, matter, shadow and circuits | Pupil identifies before, changed condition and after |
| 2 | Living systems: photosynthesis, respiration, seeds and transport | Pupil explains mechanism and conditions |
| 3 | Food webs, adaptations and investigation changes | Pupil avoids unsupported universal claims |
| 4 | Mixed what-if scenarios with reversed conditions and delayed retests | Pupil transfers reasoning without model answers |
Keep daily sessions manageable. Begin with two straightforward predictions, one changed-context contrast and one explanation. Ask the child to give the expected result and a reason. When the result depends on conditions, identify the key condition explicitly.
Review the most frequent misconception rather than every mistake at equal length. A pupil who confuses one open switch with the whole circuit needs a topology explanation. Another who knows the circuit but misreads “in Branch A only” needs practice reading diagram labels and qualifiers.
In group lessons, ask different students to defend the original, changed and reversed cases. Then give each pupil a fresh individual example. Group agreement is valuable discussion, but an independent response is needed to demonstrate mastery.
Parents can support without needing a Science laboratory. Use safe paper diagrams and scenarios about wet laundry, shadows or plants. Do not recommend hot-water burns, household mains wiring or unsafe wildlife experiments to make the topic feel realistic.
Delayed retests can be given after two days and about a week. A child who can predict the new outcome without copying the original answer has begun to understand the underlying scientific relationship.
Prediction is a broader life skill. It helps people think about the consequences of actions, whether conserving water, choosing a material or protecting habitats. The educational aim is not merely to guess the examiner’s story, but to understand how systems respond when a factor changes.
9. Further independent prediction challenges
Try a final series of questions without a model answer beside them. In a diagram with a torch, a cardboard object and a screen, suppose the torch remains fixed and the screen moves farther back. The shadow in the familiar small-source model generally becomes larger, because the light paths defining its boundary continue spreading. But the exact shadow height requires measurements or a scaled diagram. This exercise checks whether the pupil can distinguish a qualitative prediction from a numerical calculation.
In a plant investigation, suppose one group has sufficient water and light while another is deprived of water. The second group may show reduced normal function, but predicting its exact height at the end of the week would require data or a justified model. If the children later restore watering, the plant’s recovery will depend on its health, the duration of the shortage and other conditions. The aim is to connect a needed resource with a possible effect without promising a guaranteed outcome.
For a food web, imagine that an insect population decreases and two predators eat that insect. One predator also eats other prey, while the other has no alternative prey drawn in the diagram. The change may affect the predators differently, but a simple web still cannot supply precise population counts. Students should trace the feeding links, say which food source is reduced, and distinguish an available alternative from an invented one. The missing population evidence should be identified only when the question asks for a stronger prediction.
For an electrical circuit, ask a pupil to compare two actions. Opening a switch on the only path breaks the circuit and prevents the bulb on that route from lighting. Opening a switch in one parallel branch can leave a different intact branch working. The same word “open” appears in both questions, but the wiring determines the effect. If the child predicts that all bulbs must go out in every case, the problem is not vocabulary; it is the model of which conducting paths remain complete.
A parent can use these four contrasts as a ten-minute oral quiz. First ask for the expected result, then the cause, then one condition on which the result depends. Hide the earlier worked examples and change a number or object. The correct response should be independently constructed and scientifically defensible, not copied from the nearest familiar sentence. This is how a before-and-after exercise becomes a test of transferable understanding.
Frequently asked questions
What does what happens if mean in PSLE Science?
It asks a student to predict a supported consequence when a named condition changes.
Must every prediction be certain?
No. Some follow necessarily from the simplified setup, while others depend on biological or environmental factors.
Can pupils use may or might?
Yes when the supplied evidence supports a possibility rather than a guarantee. Explain why.
Why does the original setup matter?
A change has meaning relative to the starting condition and controlled factors.
What if two relevant factors change together?
A single-cause conclusion may not be supported; identify what else changed.
Can a food web support precise future population numbers?
Not unless the question supplies adequate data or model assumptions.
Is opening a switch always enough to turn off all bulbs?
No. It depends on circuit connections and which branch is interrupted.
Do plants stop all life processes without light?
No. Photosynthesis needs light, while living plant cells continue respiration.
How can parents practise what-if questions?
Change one factor in a familiar diagram and ask the child for the mechanism and predicted effect.
Does the 2027 PSLE definitely contain these thirty cases?
No. They are original teaching examples for syllabus-aligned reasoning.
Do Foundation Science students learn predictions?
Yes, within their own learning outcomes and assessment format; use the correct registered level.
How can a teacher test transfer?
Reverse the change or switch to a different apparatus that uses the same scientific mechanism.
Official sources and eduKate next reading
- MOE 2023 Primary Science syllabus — prediction and scientific inquiry
- SEAB PSLE 2026 Science examination formats
- Thirty cannot-be-concluded evidence questions
- Twenty-four cross-topic PSLE Science questions
- Fifty common Science misconceptions and exam traps
- Booklet B wrong answers repaired with reasoning
- Original Standard Science full mock
- eduKate Singapore parent and student support
Reviewed 13 November 2026. The thirty examples are independently written hypothetical teaching scenarios, not examination questions or guaranteed predictions. The 2026 SEAB format remains the verified baseline; check examination-year 2027 updates when released. eduKateSG is independent of MOE and SEAB.