Leading questions and eyewitness testimony | AQA A-Level Psychology Revision
Updated: Sep 9
For 7182 specification, first teach in September 2025
AQA A-Level Psychology | Free Revision Notes
Estimated study time: 40 minutes
These Leading questions and eyewitness testimony A-Level Psychology revision notes explain how the wording of a question can influence a witness’s account. A leading question may encourage a particular answer, alter an estimate or introduce information that was not present in the original event. You will examine controlled research into leading questions, apply the findings to unfamiliar investigations and evaluate whether laboratory studies accurately represent real eyewitness experiences. Leading questions are one of the factors AQA identifies as affecting eyewitness accuracy.
Learning Objectives 🎯
By the end of this revision page, you should be able to:
Define eyewitness testimony and leading questions.
Explain how wording can influence a witness’s answer.
Explain the response-bias and substitution accounts of misleading questions.
Describe research into the effects of leading questions.
Apply leading questions to unfamiliar research scenarios.
Evaluate research into misleading information.
Revision Notes 📚
Leading questions and eyewitness testimony A-Level Psychology revision overview
Eyewitness testimony, or EWT, is the account given by someone who has witnessed an event.
An eyewitness may be asked to recall:
What happened.
Who was present.
What a person looked like.
The order of events.
The speed or direction of movement.
Objects seen at the scene.
Details of a conversation.
Eyewitness accounts can be important in criminal investigations, but memory is not a perfect recording of an event.
The wording used by an interviewer can affect what the witness reports.
AQA identifies three factors affecting eyewitness accuracy:
Leading questions.
This lesson focuses specifically on leading questions.
What is misleading information?
Misleading information is incorrect or suggestive information presented after an event.
It may be introduced through:
A question.
An interviewer’s wording.
A written description.
A conversation with another witness.
A news report.
Information supplied by someone who did not witness the event.
Misleading information may affect a person’s later account, even when the information conflicts with what they originally saw.
Leading questions and post-event discussion are both forms of misleading information, but they are not identical.
Leading questions | Post-event discussion |
The suggestion is contained in a question | Information comes from discussing the event with other people |
Usually introduced by an interviewer or researcher | Usually introduced by another witness or source |
Wording may imply a particular answer | Different accounts may become combined |
Focus of this lesson | Focus of the next lesson |
What is a leading question?
A leading question is a question whose wording encourages or suggests a particular answer.
For example:
“How fast was the car travelling?”
This asks for a speed estimate but does not suggest that the car was moving particularly quickly.
By contrast:
“How fast was the car travelling when it smashed into the other vehicle?”
The verb smashed implies a forceful collision. This may encourage a higher speed estimate.
A leading question may influence a witness by:
Suggesting that an event was more severe than it appeared.
Implying that a particular object was present.
Assuming that something happened.
Directing attention towards one interpretation.
Encouraging the witness to report information they are uncertain about.
Introducing information that becomes confused with the original event.
Neutral and leading questions
More neutral question | More leading question |
“What happened to the two vehicles?” | “How violently did the vehicles crash?” |
“Did you see any broken glass?” | “Did you see the broken glass?” |
“What colour was the person’s coat?” | “Was the person wearing a black coat?” |
“Was anything being carried?” | “Did you see the knife being carried?” |
“Describe the driver’s behaviour.” | “Why was the driver acting aggressively?” |
The leading versions contain assumptions or suggestions that may affect the answer.
Leading questions and presuppositions
A presupposition is an assumption contained within a question.
Consider:
“Did you see the broken headlight?”
The word the suggests that there was a broken headlight.
A witness who is uncertain may assume that the interviewer knows a broken headlight was present.
Compare this with:
“Did you see a broken headlight?”
This wording does not assume the object was definitely present as strongly.
A presupposition may therefore encourage a witness to report an object or event that they did not actually observe.
Why eyewitnesses may accept a suggestion
Witnesses may be uncertain about details because:
The event happened quickly.
Attention was directed elsewhere.
Visibility was poor.
Some details were not considered important at the time.
Time has passed since the event.
The witness expects the interviewer to have accurate information.
When the original memory is incomplete, suggestive wording may provide an apparently plausible answer.
The witness may not be deliberately lying. They may genuinely believe that the information was part of the event.
Loftus and Palmer’s research
The aim of the research
Loftus and Palmer investigated whether the wording of a question could affect a person’s memory of a traffic accident.
Their research examined whether changing one verb in a question would influence:
Participants’ estimates of speed.
Their later reports of details that were not present.
The study is important because it demonstrates that small changes in wording can affect eyewitness testimony.
The first experiment
Participants watched film clips showing traffic accidents.
After viewing the clips, they answered questions about what they had seen.
The critical question asked participants to estimate the speed of the vehicles. Different groups received different versions of the question.
The verbs included:
Smashed
Collided
Bumped
Hit
Contacted
For example:
“About how fast were the cars going when they smashed into each other?”
Other participants received the same question with a different verb.
Independent and dependent variables
The independent variable, or IV, was the verb used in the critical question.
The dependent variable, or DV, was the participant’s estimated speed.
The study used different participants in the different wording conditions. This is an independent groups design.
A clear operationalisation would state:
IV: whether the verb was “smashed”, “collided”, “bumped”, “hit” or “contacted”.
DV: the speed estimate given by each participant.
The importance of clearly defining measures is explored further in operationalising the variables.
Findings of the first experiment
Participants who received stronger verbs generally gave higher speed estimates.
For example, the word smashed produced higher estimates than the word contacted.
The film clips were the same. The important difference was the wording of the question.
This suggests that the verb influenced the participants’ reports.
The finding can be summarised as:
More forceful verb → more severe interpretation → higher estimated speed
What does the first experiment show?
The findings demonstrate that a leading question can affect an eyewitness’s answer.
However, the experiment alone does not show whether the participants’ memories were permanently changed.
It is possible that participants still remembered the event accurately but adjusted their answer to fit the wording of the question.
This possibility is called response bias.
Response bias
The response-bias explanation
The response-bias explanation proposes that a leading question changes the answer a person gives without changing their underlying memory.
The participant may think:
“The researcher used the word ‘smashed’, so the cars must have been travelling quickly.”
They then give a higher speed estimate that seems consistent with the question.
Under this explanation:
The original memory remains unchanged.
The wording influences how the witness responds.
The effect may be temporary.
A different question might produce a different answer.
Applying response bias
A witness sees two cars make minor contact.
An interviewer asks:
“How fast was the car going when it crashed violently into the other vehicle?”
The witness may give a high estimate because the wording suggests a severe collision.
This does not necessarily mean that the memory of the accident has changed. The witness may simply be adjusting their answer to match the interviewer’s apparent expectation.
Response bias as a demand characteristic
A participant may attempt to work out what a researcher expects.
The wording of the question can provide a clue about the desired response.
This may create demand characteristics.
For example, the participant may assume that:
“Smashed” requires a higher estimate.
“Contacted” requires a lower estimate.
The researcher expects a particular object to have been present.
If participants respond to these expectations, the study may measure their interpretation of the question rather than a genuine change in memory.
The second Loftus and Palmer experiment
Testing whether memory had changed
Loftus and Palmer conducted a further experiment to investigate whether leading questions could alter the memory itself.
Participants watched a film of a traffic accident.
They were then asked a question using either:
The verb smashed.
The verb hit.
No critical speed question.
Later, participants were asked whether they had seen broken glass.
There was no broken glass in the film.
Findings
Participants who had previously received the word smashed were more likely to report seeing broken glass than participants in the other conditions.
The word “smashed” was consistent with a severe accident in which broken glass might reasonably be expected.
This suggests that the wording introduced information that affected later recall.
Why this supports memory alteration
The later question did not ask for another speed estimate.
Instead, participants were asked about an object that had not appeared.
If the first question had caused only a temporary response bias, it would be harder to explain why it affected a later memory question.
The findings suggest that the misleading verb may have become incorporated into the participant’s representation of the event.
This supports the substitution explanation.
The substitution explanation
What is substitution?
The substitution explanation proposes that misleading information becomes incorporated into memory and changes the original representation of the event.
The misleading information may:
Replace part of the original memory.
Become combined with the original details.
Make the witness uncertain about the source of a detail.
Produce a reconstructed version of the event.
The person may later report the misleading detail because they believe it was genuinely observed.
The substitution process
The process can be represented as follows:
The witness observes an event.
A memory of the event is formed.
A leading question introduces misleading information.
The misleading information is combined with or substituted for the original detail.
The witness later recalls the altered version.
Eyewitness accuracy is reduced.
Applying substitution
A witness sees a person carrying a mobile phone.
Later, an interviewer asks:
“What colour was the knife the person was holding?”
The question falsely suggests that a knife was present.
If the misleading information becomes incorporated into memory, the witness may later report seeing a knife.
The witness may not be knowingly providing false information. Their reconstructed memory may now include the suggested object.
Response bias and substitution compared
Response bias | Substitution |
The answer changes | The memory representation changes |
The original memory remains intact | Misleading information becomes incorporated |
Effect may be caused by expectations | Effect may continue into later recall |
The witness may adjust an estimate | The witness may remember a suggested detail |
Supported by immediate answer changes | Supported when later memory is affected |
The two explanations are not necessarily mutually exclusive.
A leading question could produce response bias in some people and memory substitution in others.
Research involving articles in questions
“A” compared with “the”
Research has also investigated whether changing a small article in a question can influence eyewitness reports.
For example:
“Did you see a broken headlight?”
compared with:
“Did you see the broken headlight?”
The word the contains a stronger presupposition that the broken headlight existed.
Participants receiving the more suggestive wording have been more likely to report the implied object.
This demonstrates that a leading question does not need to be dramatic. Even a small change in wording can affect a response.
Why this matters
Police interviews may involve many detailed questions.
Small differences in wording could influence whether a witness:
Reports an object.
Selects a particular description.
Estimates a speed or distance.
Assumes an event occurred.
Becomes increasingly confident in a suggested detail.
Interviewers should therefore avoid questions that contain unsupported assumptions.
Applying leading questions to research scenarios
Identifying the leading element
In an application question, identify the exact word or phrase that makes the question leading.
For example:
“How fast was the reckless cyclist travelling when they crashed into the pedestrian?”
Possible leading elements include:
“Reckless”
“Crashed”
These words suggest dangerous behaviour and a serious collision.
Simply stating that the question is leading is not enough for a developed answer.
Explaining the direction of the effect
You should also explain how the wording may affect the answer.
For example:
The words “reckless” and “crashed” suggest that the cyclist was travelling quickly and caused a severe collision. Witnesses may therefore give higher speed estimates than witnesses asked a neutral question.
Applying the response-bias explanation
A complete response-bias application might state:
Witnesses may assume that the researcher expects a high estimate because the word “crashed” implies a forceful collision. They may adjust their answer to fit the wording even if their original memory has not changed.
Applying the substitution explanation
A complete substitution application might state:
The suggestion that the cyclist was reckless may become incorporated into the witness’s memory. The witness may later remember dangerous behaviour that was not visible in the original event.
Four steps for application
Use this structure:
Quote or identify the leading wording.
Explain what the wording suggests.
Predict how the witness’s answer may change.
Link the change to response bias or substitution.
Weak application
The question is leading, so the witness will be inaccurate.
This lacks detail.
Stronger application
The interviewer asks how fast the van was travelling when it “smashed” into the car. “Smashed” implies a forceful collision, so the witness may report a higher speed. This could be response bias because the witness adjusts their answer to match the wording.
Designing research into leading questions
A simple experimental design
A psychologist could show participants the same film of an event.
Participants could then be allocated to one of two conditions:
Leading question condition.
Neutral question condition.
For example:
Leading question
“Did you see the damaged traffic light?”
Neutral question
“Did you see a damaged traffic light?”
The psychologist could record whether each participant reports seeing the object.
Independent variable
The IV would be the wording of the question.
This must be operationalised precisely.
For example:
Whether participants were asked about “the damaged traffic light” or “a damaged traffic light”.
Dependent variable
The DV would be the measured eyewitness response.
For example:
Whether the participant answers yes or no.
The speed estimated in miles per hour.
The number of correct details recalled.
Whether a suggested object is reported.
A good research answer states exactly how the DV is recorded.
Controlling variables
To make the comparison fair, the researcher should keep other factors constant, including:
The film shown.
The length of the film.
The time between viewing and questioning.
The interviewer’s tone.
The order of other questions.
The room used.
The instructions given.
The time allowed for recall.
If these factors differ, they may provide alternative explanations for the results.
Standardisation
The researcher could use standardised instructions and questions.
This means that participants within each condition receive the same:
Instructions.
Film.
Timing.
Testing procedure.
Interviewer behaviour.
Only the critical wording should differ between conditions.
Avoiding investigator effects
An interviewer’s behaviour might unintentionally influence a participant.
For example, the interviewer might:
Emphasise a particular word.
React positively to one answer.
Repeat a question.
Use different facial expressions.
Provide feedback suggesting an answer is correct.
A written questionnaire or recorded questions could reduce variation in interviewer behaviour.
Using an independent groups design
A repeated measures design may create a problem because the same participant would hear both versions of the question.
They might:
Notice the purpose of the study.
Compare the wording.
Guess the hypothesis.
Change their answer deliberately.
An independent groups design avoids presenting both versions to the same participant.
However, individual differences between groups could affect the results. Random allocation could help distribute these differences more evenly.
Ethical issues
Research into eyewitness testimony may involve deception.
Participants may not be told that:
The question contains misleading information.
The object mentioned was absent.
The researcher is testing memory distortion.
Revealing the full aim beforehand could create demand characteristics.
Researchers should therefore:
Obtain appropriate consent.
protect participants from harm.
Allow withdrawal.
Keep data confidential.
Debrief participants fully afterwards.
These issues connect with ethical issues in psychological investigations.
Evaluating research into leading questions
A strength is controlled experimental evidence
Research into leading questions is often conducted under controlled conditions.
Researchers can:
Show every participant the same event.
Change only the wording of a critical question.
Standardise the delay before recall.
Record answers objectively.
Compare the results across conditions.
If the groups give different answers, the wording of the question provides a plausible cause.
This gives the research good internal validity.
The research can demonstrate cause and effect
Because the researcher manipulates question wording, the study can investigate a causal relationship.
The proposed relationship is:
Leading wording → altered eyewitness response
A correlational study could identify a relationship between interview style and accuracy, but it would be more difficult to establish whether the wording directly caused the difference.
Replication is possible
Standardised film clips and questions allow researchers to repeat the study.
Replication helps psychologists determine whether the effect is reliable.
If similar wording effects appear across repeated investigations, confidence in the finding increases.
Laboratory tasks may lack ecological validity
Watching a short film is different from witnessing a genuine accident or crime.
Laboratory participants usually know that:
They are safe.
The event is not real.
Their answer is part of an experiment.
Their testimony will not decide a real person’s future.
Real witnesses may experience:
Shock.
Personal involvement.
Genuine danger.
A longer and more complex event.
Pressure associated with a police investigation.
Serious consequences if their testimony is inaccurate.
This may limit ecological validity.
The laboratory findings show that leading questions can influence memory under controlled conditions. They do not prove that every real witness will be affected to the same extent.
The research may lack mundane realism
Mundane realism concerns whether an activity resembles an ordinary real-life situation.
Estimating the speed of vehicles in a film may resemble part of an eyewitness interview, but the overall situation is still artificial.
Participants may pay different attention to a film than a person would pay during an unexpected real event.
This limits how confidently the findings can be generalised to genuine testimony.
Demand characteristics may affect responses
Participants may notice that the wording has changed or infer what the researcher expects.
For example:
“Smashed” appears to invite a high estimate.
“Contacted” appears to invite a low estimate.
“The broken light” suggests that the light existed.
Participants may cooperate with the perceived aim.
If so, the findings may reflect response bias rather than an altered memory.
However, the later broken-glass finding is harder to explain through a temporary adjustment alone. This provides some support for substitution.
Immediate estimates may measure judgement rather than memory
A speed estimate is not a direct copy of visual memory.
Many people are not skilled at estimating vehicle speed.
Participants may base their response on:
The wording of the question.
General beliefs about accidents.
Guesswork.
The severity of visible damage.
What seems like a reasonable answer.
This means that changes in estimates do not necessarily demonstrate that the visual memory itself has been altered.
The second experiment is valuable because it tested recall of a specific absent detail rather than only an uncertain numerical estimate.
Real witnesses may be more motivated
Real eyewitnesses may try harder to be accurate because the consequences are important.
Laboratory participants may give quick estimates without carefully checking their memory.
Greater motivation might make real witnesses:
More cautious.
Less willing to guess.
More likely to say they do not know.
More resistant to suggestive wording.
Alternatively, pressure and uncertainty could make some witnesses more vulnerable. Laboratory findings should therefore be generalised cautiously.
Practical applications are important
Research into leading questions has clear practical value.
Police officers, lawyers and other interviewers should avoid:
Suggestive verbs.
Questions that assume an object was present.
Repeated questions implying that the first answer was wrong.
Feedback that confirms a suggested detail.
Questions containing unsupported descriptions.
More neutral questions include:
“Tell me everything you remember.”
“What happened next?”
“Describe the person.”
“Was anything being carried?”
“What, if anything, did you see near the vehicle?”
The findings have contributed to techniques that encourage detailed recall without imposing an interviewer’s assumptions, including the cognitive interview.
Open questions may reduce suggestion
An open question allows a witness to answer in their own words.
For example:
“Describe everything you remember about the vehicle.”
A closed question restricts the possible response:
“Was the vehicle blue?”
Closed questions are not automatically leading. However, they may become leading if they contain assumptions or encourage a particular answer.
Open questions may reduce the amount of information introduced by the interviewer.
Leading questions cannot explain every inaccuracy
Eyewitness errors can occur even when interviewers use neutral questions.
Other influences include:
Post-event discussion.
Anxiety.
Poor attention.
The passage of time.
Retrieval difficulties.
Expectations about what normally happens.
Leading questions are therefore one factor affecting accuracy rather than a complete explanation of eyewitness error.
Not every witness is affected equally
The impact of a leading question may vary according to:
How clearly the event was seen.
The witness’s confidence in the original memory.
The time between the event and questioning.
The strength of the suggestion.
The importance of the detail.
Whether the witness is willing to say they do not know.
A clear and well-established memory may be more resistant than an uncertain one.
Research distinguishes report changes from memory changes
One of the most valuable features of the research is the distinction between:
A change in the answer given.
A change in the underlying memory.
The first Loftus and Palmer experiment could be explained through response bias.
The later question about broken glass provides stronger evidence that misleading wording may affect subsequent recall.
However, researchers cannot directly inspect a memory. They infer memory alteration from participants’ responses.
This means that separating response bias and substitution remains difficult.
Ethical benefits of controlled research
Researchers cannot ethically create genuine dangerous accidents simply to study witness memory.
Film-based experiments provide a safer alternative.
Although they may be less realistic, they allow psychologists to investigate misleading information without exposing participants to genuine harm.
This creates a trade-off:
Controlled laboratory study | Genuine eyewitness event |
Safer and more controlled | More realistic |
IV can be manipulated | Important variables cannot be controlled |
Easier to replicate | Difficult to reproduce |
May lack emotional involvement | May involve genuine distress |
Stronger causal conclusions | Greater ecological validity |
A balanced conclusion
Research provides convincing evidence that the wording of questions can alter eyewitness reports.
Controlled experiments show that:
Stronger verbs can affect estimates.
Presuppositions can encourage false reports.
Misleading wording can influence later recall.
However, laboratory studies use artificial events and may involve demand characteristics.
The safest conclusion is:
Leading questions can reduce eyewitness accuracy, particularly when the original memory is uncertain, but they are one of several influences on testimony.
Answering examination questions
Outline questions
For an outline question, include:
A clear definition of leading questions.
An explanation of how wording suggests an answer.
Relevant research procedure or findings.
A link to eyewitness accuracy.
Do not spend the whole answer evaluating the research if evaluation has not been requested.
Application questions
For an apply or explain question:
Identify the exact wording.
Explain the suggestion it contains.
State how the witness’s response may change.
Use a relevant concept such as response bias or substitution.
Research-methods scenarios
When a leading-question study is described, be ready to identify:
The IV.
The DV.
The experimental design.
A possible extraneous variable.
A suitable control.
Ethical issues.
Whether the data are nominal, ordinal or interval.
How validity or reliability could be improved.
Evaluation questions
A strong evaluation paragraph should:
State the strength or limitation.
Include relevant evidence or methodological detail.
Explain its effect on the conclusions.
Link back to leading questions and eyewitness accuracy.
Weak evaluation
The study was artificial.
Stronger evaluation
Participants watched film clips rather than witnessing a genuine accident. Their attention and emotional involvement may therefore have differed from those of real witnesses, limiting the extent to which the findings can explain eyewitness testimony in genuine investigations.
Hints from the Examiner Reports 💡
Examiner hint: Identify the factor precisely. If a scenario contains “the” rather than “a”, or a forceful verb such as “smashed”, explain why that wording is leading.
Examiner hint: Do not merely name a study. Strong responses use the research to explain the findings or behaviour in the scenario.
Examiner hint: Application requires an explicit link. State how the wording in each condition is likely to affect the participants’ responses.
Examiner hint: Avoid copying details from a scenario without explanation. Quoting the critical question gains little value unless you analyse the suggestion it contains.
Examiner hint: Research such as Loftus and Zanni can illustrate the effect of presuppositions, but it must be connected directly to the question being answered.
Examiner hint: In evaluation questions, generic criticism of laboratory research is not enough. Explain how artificiality or demand characteristics affect conclusions about leading questions and eyewitness accuracy.
Examiner hint: Keep different factors separate. Leading questions involve suggestive wording, while post-event discussion involves information received from other people.
AQA examiner commentary has particularly emphasised the need to connect research and specialist terminology directly to the conditions or scenario rather than simply describing a study.
Common Mistakes ⚠️
Mistake: Defining a leading question as any question asked after an event
Why this is incorrect:
A question is leading only when its wording suggests or encourages a particular response.
How to improve:
Identify the word, phrase or presupposition that makes the question suggestive.
Mistake: Saying that leading questions always create false memories
Why this is incorrect:
A leading question may cause response bias without changing the underlying memory.
How to improve:
Explain both response bias and substitution where relevant.
Mistake: Assuming that a participant is deliberately lying
Why this is incorrect:
The participant may genuinely believe that the suggested information was present.
How to improve:
Explain how misleading information can become incorporated into memory.
Mistake: Describing “smashed” as the dependent variable
Why this is incorrect:
The researcher manipulates the verb, so it is the independent variable.
How to improve:
Remember:
Wording = IV.
Speed estimate or eyewitness response = DV.
Mistake: Naming the IV as “leading questions”
Why this is incomplete:
The variable should be operationalised precisely.
How to improve:
State the exact wording used in each condition.
Mistake: Describing the findings without the direction of the effect
Why this is incorrect:
Saying that the estimates were “different” does not show full understanding.
How to improve:
State that stronger verbs generally produced higher speed estimates.
Mistake: Saying the first Loftus and Palmer experiment proves memory alteration
Why this is incorrect:
The speed-estimate finding could be explained by response bias.
How to improve:
Use the later broken-glass question as stronger evidence for substitution.
Mistake: Confusing leading questions with post-event discussion
Why this is incorrect:
Leading questions introduce suggestions through wording. Post-event discussion involves information from other people or accounts.
How to improve:
Identify the source of the misleading information.
Mistake: Giving a generic ecological-validity criticism
Why this is incorrect:
A criticism must explain how the research situation differs from genuine eyewitness experiences.
How to improve:
Refer to the lack of real danger, emotional involvement or consequences.
Mistake: Assuming closed questions are always leading
Why this is incorrect:
A closed question can be neutral if it does not suggest a desired answer.
How to improve:
Examine the wording and assumptions rather than only the answer format.
Mistake: Copying the question from a scenario
Why this is incorrect:
Application requires psychological explanation, not repetition.
How to improve:
Explain what the wording suggests and how it may alter the witness’s answer or memory.
Mistake: Criticising the research without linking back to EWT
Why this is incorrect:
A methodological comment gains value only when its effect on the conclusion is explained.
How to improve:
Finish the point by stating whether the research provides strong or limited evidence about real eyewitness accuracy.
Exam-Style Questions ✍️
Question 1
Which one of the following is the clearest example of a leading question?
A. “Describe everything you remember about the vehicle.”
B. “What happened after the vehicle stopped?”
C. “How fast was the reckless driver travelling?”
D. “Was anything being carried by the driver?”
[1 mark]
reveal mark scheme and indicative content
Answer: C. “How fast was the reckless driver travelling?”
Question 2
Define a leading question in the context of eyewitness testimony.
[2 marks]
reveal mark scheme and indicative content
Mark scheme
AO1 = 2
2 marks: Clear definition referring to a question whose wording suggests or implies a particular answer.
1 mark: Limited or partial definition.
Example answer
A leading question is a question worded in a way that suggests a particular answer or introduces information that may influence an eyewitness's response.
Question 3
Explain one difference between the response-bias explanation and the substitution explanation.
[3 marks]
reveal mark scheme and indicative content
Mark scheme
AO1 = 3
Award marks for a clear distinction:
1 mark: Response bias suggests wording affects the answer a witness chooses to give.
1 mark: The original memory itself is not necessarily altered.
1 mark: Substitution suggests misleading information becomes incorporated into or alters the original memory.
Example answer
The response-bias explanation suggests that a leading question changes the answer a witness gives without necessarily changing the underlying memory. The substitution explanation suggests that misleading information actually alters or becomes incorporated into the witness's memory of the event.
AQA's 2025 mark scheme explicitly accepts both response bias and substitution explanations for leading-question effects.
Question 4
A witness sees two bicycles make slight contact. Later, an interviewer asks:
“How fast was the cyclist travelling when they smashed into the other bicycle?”
Explain how this question might affect the witness’s answer.
[4 marks]
reveal mark scheme and indicative content
Mark scheme
AO2 = 4
Award 1 mark each for:
Identifying “smashed” as leading wording.
Explaining that “smashed” suggests a more forceful collision than the slight contact actually witnessed.
Under response bias, the wording may encourage the witness to give a higher speed estimate.
Under substitution, the misleading wording may distort the witness's memory so the incident is remembered as more severe.
Example answer
The word “smashed” is a leading term because it implies that the collision was more severe than the slight contact Eva actually witnessed. This could create a response bias, causing the witness to estimate a higher speed because this seems consistent with “smashed”. Alternatively, the wording could become incorporated into the memory itself, so the collision is later remembered as more violent.
Question 5
Outline one experiment conducted by Loftus and Palmer into leading questions.
[4 marks]
reveal mark scheme and indicative content
Mark scheme
AO1 = 4
Award 1 mark for each relevant point, up to 4 marks.
Indicative content
Participants watched films of car accidents.
They were asked to estimate the speed of the cars.
The critical question used different verbs, such as “smashed”, “collided”, “bumped”, “hit” or “contacted”.
More forceful verbs produced higher estimates of speed.
This demonstrated that wording of a question can influence eyewitness responses.
AQA specifically identifies this experiment and the manipulation of these critical verbs as relevant research into misleading information.
Example answer
Loftus and Palmer showed participants films of car accidents and asked them to estimate how fast the vehicles had been travelling. The critical question contained different verbs, such as “smashed”, “hit” or “contacted”. Participants given stronger verbs such as “smashed” produced higher speed estimates, showing that leading wording can influence eyewitness testimony.
Question 6
A researcher shows participants a film of a shop. No window is broken.
Half the participants are asked:
“Did you see the broken window?”
The remaining participants are asked:
“Did you see a broken window?”
Explain why the researcher might expect different answers in the two conditions.
[4 marks]
reveal mark scheme and indicative content
Mark scheme
AO2 = 4
Award 1 mark each for:
“The broken window” is more leading than “a broken window”.
The definite article “the” presupposes or implies that a broken window existed.
Participants in this condition may therefore be more likely to report seeing one.
This could be explained by response bias or by the misleading information becoming incorporated into memory.
AQA's 2025 mark scheme uses precisely this distinction, noting that “the” implies that an object was present whereas “a” is more ambiguous.
Example answer
The question using “the broken window” implies that a broken window definitely existed, even though it did not. Participants may therefore be more likely to answer that they saw one. The wording could simply bias their response, or it could become incorporated into their memory so that they later believe they actually saw a broken window.
Question 7
A psychologist investigates the effect of question wording on eyewitness testimony.
Participants watch the same video. One group is asked how fast a car was travelling when it “smashed” into another vehicle. A second group is asked how fast it was travelling when it “contacted” the vehicle.
Identify the independent variable and dependent variable. Explain one variable that the psychologist should control.
[6 marks]
reveal mark scheme and indicative content
Mark scheme
AO2 = 6
Award up to 2 marks for each element.
Independent variable
The wording of the critical question.
Specifically, whether the collision is described using “smashed” or “contacted”.
Dependent variable
The participant's estimate of the car's speed.
Control variableCredit any appropriate variable with explanation, for example:
All participants should watch the same video, so differences in speed estimates cannot be caused by seeing different collisions.
The time between viewing the video and answering the question should be kept constant.
Instructions and viewing conditions should be standardised.
Example answer
The independent variable is the wording of the question, specifically whether participants hear “smashed” or “contacted”. The dependent variable is the speed estimate given by each participant. One variable that should be controlled is the video shown. Every participant should see exactly the same collision so that any difference in speed estimates is more likely to result from the wording rather than differences in the event they observed.
Question 8
Explain how the findings of the broken-glass experiment provide stronger evidence for memory substitution than the original speed-estimate experiment.
[4 marks]
reveal mark scheme and indicative content
Mark scheme
AO3 = 4
Award 1 mark each for:
In the original experiment, participants only gave speed estimates.
Different speed estimates could be explained by response bias, with participants adjusting their answer to match the verb.
In the later experiment, participants were subsequently asked whether they had seen broken glass, although there was none.
Participants exposed to stronger wording were more likely to report broken glass, suggesting the misleading information had altered their memory rather than simply their immediate response.
AQA explicitly lists Loftus and Palmer's broken-glass experiment alongside the original verb manipulation as evidence relevant to leading questions.
Example answer
The speed-estimate experiment could be explained by response bias because participants might simply choose a higher speed when they hear a stronger word such as “smashed”. In the broken-glass experiment, participants later reported seeing something that had not actually been present. This provides stronger support for substitution because it suggests that the misleading wording had changed their memory of the event, rather than simply changing their speed estimate.
Question 9
Outline one strength and one limitation of controlled research into leading questions.
[6 marks]
reveal mark scheme and indicative content
Mark scheme
AO3 = 6
Award up to 3 marks for the strength and 3 marks for the limitation.
Indicative content
Strength
Laboratory studies can use highly controlled and standardised procedures.
Participants can see the same event, with only the wording of the critical question changed.
This reduces the influence of extraneous variables and allows stronger conclusions about the effect of leading questions.
Limitation
Watching a film of an accident is less realistic than witnessing an unexpected real-life event.
Participants may experience much less emotion or consequence than genuine eyewitnesses.
This may reduce ecological validity and make generalisation to real eyewitness testimony difficult.
AQA accepts validity, control, sampling, replication and practical implications when evaluating research into eyewitness testimony.
Example answer
One strength is the high level of control. Researchers can show every participant the same event and change only the wording of the critical question. This makes it easier to conclude that differences in recall were caused by the leading question rather than another variable.
One limitation is that controlled studies often involve watching films rather than witnessing genuine events. Participants know that they are taking part in research and experience fewer real-world consequences, so their responses may not accurately represent the behaviour of genuine eyewitnesses.
Question 10
Discuss leading questions as a factor affecting the accuracy of eyewitness testimony.
[16 marks]
Comments