Loftus And Palmer

Front 1

Loftus and Palmer - Eyewitness testimony

Back 1

Background to the study (context):

Psychological research has found that people are very inaccurate. For example, Marshall (1969) found that when the Air Force personnel who knew in advance that they would be asked to estimate the speed of a vehicle, observed a car traveling at 12mph, their estimates ranged from 10 mph to 50mph.

This suggests that it might be quite easy to influence the answers that people give to such numerical questions. One way to influence answers is phrasing a question in such a way that it leads a person to give a particular answer. some questions are more 'suggestive' than others: in legal terms such questions are called leading questions.

Front 2

Aim and hypothesis: Experiment 1

The aim of the first experiment was to see if the estimates given by participants about the speed of vehicles in a traffic accident would be influenced by the wording of the question asked. For example, participants asked about how fats the cars were travelling when they 'hit' each other would give different speed estimates and have different expectations than participants asked the same question with the word 'smashed' instead.

Back 2

A laboratory experiment was used and self report methods in the form of a questionnaire.

Strengths:

- There is a clear IV and a clear DV which allows a distinctive cause and effect relationship to be made

- The self-report allows a lot of information from the participant to be collected in a short period of time which will allows data to be collected quickly and from the participant themselves - it also means that the participant won't feel pressured to answer a certain way so therefore decreasing social desirability bias.

- extraneous variables are normally able to be controlled and limited.

Weaknesses -

- low ecological validity - artificial setting not the same as witnessing a real life accident, won't experience same senses etc..

- low mundane realism - watching a car accident on a tv screen is not an everyday task that someone would do.

Front 3

Participants: Experiment 1

The participants used in this study were 45 American students, divided into 5 groups.

This makes the sample ethnocentric as all the participants were from America. Also as all the participants were students this means they were all of the same age and suggests they may not have had as many years driving experience as say someone a lot older. This would be a major factor that could affect the results of the study as they may not be able to estimate the speeds of a car as accurately as someone with more years driving experience.

Back 3

IV (used in experiment 1): Verb used in the leading question

DV (used in experiment 2): Speed estimations of how fast the car was travelling

Front 4

Procedure: Experiment 1

The participants were shown seven film clips of different traffic accidents. the clips originally made as part of a driver safety film.

After each clip the participants were given questionnaire which asked them to describe the accident and then answer a series questions about the accident.

Back 4

There was one critical question: 'How fast were the cars going when they hit each other?' One group of participants were given this question. the other four groups were given the verbs 'smashed', 'collided', 'bumped' or 'contacted' in place of the word 'hit'. thus there were five experimental groups in this laboratory experiment.

This makes it an independent measures design.

Front 5

Results: Experiment 1

The mean speed estimate was calculated for each experimental group. The group given the word 'smashed' estimated a higher speed than the other groups (40.8mph). The group given the word 'contacted' estimated the lowest speed (31.8mph).

Back 5

Smashed: 40.8 mph

Collided: 39.3 mph

Bumped: 38.1 mph

Hit: 34 mph

Contacted: 31.8 mph

Front 6

Conclusion: Experiment 1

The results show that the form of a question can have a significant effect on a witness' answer to a question. In other words, a leading question can affect the accuracy of a memory.

Loftus and Palmer proposed two explanations for this:

Back 6

1 Response- Bias factors

The different speed estimates occur because the critical word influences or biases a person's response. In other words hearing the verb 'smashed' leads you to think of a noisy high impact accident and therefore leads you to estimate a higher speed than for the verb 'hit'.

Front 7

2 The memory representation is altered

The critical verb changes a person's perception of the accident - some critical word would lead someone to have a perception of the accident being more serious. this perception is then stored in the person's memory of the event.

Back 7

If this were to be true we would expect participants to 'remember' other details that are not true. Loftus and Palmer tested this in their second experiment.

Front 8

Aim and Hypothesis: Experiment 2

Loftus and Palmer conducted a second experiment to further investigate the effects of leading questions on accuracy.

This time going one step further to see if the leading question altered subsequent expectations about the likely consequences. In particular they wanted to see if such questions simply create a response bias (explanation 1) or if they actually alter a person's memory representation (explanation 2)

Back 8

Participants: Experiment 2

The participants in this study were again American students.

A new group of 150 students were used.

Front 9

Procedure: Experiment 2

Part 1 -

Participants were shown a one-minute film which contained a four-second multiple car accident.

They were asked a set of questions including the critical question about speed.

Back 9

There were three groups:

- Group 1 - 'How fast were the cars going when they smashed each other?'

- Group 2 - 'How fast were the cars going when they hit into each other?'

- Group 3 - were asked no question about the speed of the vehicles. this was a control group.

Front 10

Part 2 -

One week later the participants were asked to return to the psychology laboratory.

They were asked some further questions via a questionnaire including 'Did you see any broken glass?'

Back 10

There was no broken glass in the film but presumably, those who thought the car was travelling faster might expect that there would be broken glass.

Front 11

Results: Experiment 2

Part 1 -

Participants gave a significantly higher speed estimate in the 'smashed' condition. as before at 10.46mph

Whereas 'hit' was estimated at 8mph.

Back 11

Part 2 -

A week later ...

they found that participants in the 'smashed' condition were also more likely to think they saw broken glass.

Front 12

Conclusion - Experiment 2

Part 1 -

Results from part one of experiment 2 show again the effect of leading questions on the influence of an answer and confirm that the word smashed creates a perception that the car is travelling faster.

Back 12

Part 2 -

The results from part 2 of experiment 2 suggest that this is not due to response bias factor because leading questions actually altered the memory a participant had for the event.

Front 13

Propose memory is determined by two sources:

1 One's own perception gleaned at the time of the original event.

2 External information supplied after the fact (such as leading questions)

Over time information from these two sources is integrated in such a way that we are unable to tell which source any particular piece of information came from. All we have is one 'memory'.

Back 13

In the second experiment the two pieces of information combine to form a memory of an accident that appears quite severe and therefore generates certain expectations for example that there is likely to be broken glass.

This is known as a schema.

Front 14

The research method:

This study was a laboratory experiment.

Strengths:

- There is a clear IV and a clear DV which allows a distinctive cause and effect relationship to be made

- extraneous variables are normally able to be controlled and limited.

Weaknesses -

- low ecological validity - artificial setting not the same as witnessing a real life accident, won't experience same senses etc..

- low mundane realism - watching a car accident on a tv screen is not an everyday task that someone would do.

Back 14

The research method:

The experimental design used was an independent measures design.

Front 15

The sample:

American students were used in this study.

Limitation:

- This makes the sample ethnocentric as all the participants were from America.

- Also as all the participants were students this means they were all of the same age and suggests they may not have had as many years driving experience as say someone a lot older. This would be a major factor that could affect the results of the study as they may not be able to estimate the speeds of a car as accurately as someone with more years driving experience.

Back 15

Ethics:

As a whole the study was ethical however watching car accidents could lead to psychological harm of the participants.

However the participants weren't deceived and gave informed consent.

Front 16

Reliability:

"Loftus and Palmer's study is high in reliability due to the standardised procedure"

The same 7 video clips were used throughout the study as well as the same 9 filler questions (apart from the verb in the critical question) were used.

This allows us to replicate research for new participants were treated in the same way so we can make sound conclusions about the results gathered.

Back 16

Ecological Validity:

"This study lacked ecological validity as it was set in a lab"

This means that the study had a artificial setting. Watching a film clip of an car accident rather than watching a real life car accident is a lot different and overall a completely different situation for the senses of the body to process.

Real-life witnesses may be feeling scared or anxious.

In this study eyewitness testimony was tested by showing participants video clips. The nature of the task might affect the generalisability of this study.

Front 17

Snapshot/Longitudinal:

Experiment 1 is a snapshot study as it took place over one day.

Experiment 2 however is a very very short longitudinal study because it takes place over one week period.

Back 17

Qualitative/Quantitative data:

The data collected in the study was quantitative data.

Strengths:

-Easy to analyse and compare results and interpret them into graphs.

-Easy to produce accurate, reliable results.

Weaknesses:

- Doesn't produce a great amount of detailed information

-You don't know why they answered the questions as they did

Front 18

Usefulness:

Cognitive Interviewing:

The cognitive interview is a method of interviewing eyewitnesses and victims about what they remember from a crime scene.

Back 18

This technique is used by the police to increase the amount of information gained from witnesses and increase the validity (truthfulness) and reliability.

This involves:

- avoiding specific leading questions

- perspective taking

- avoiding post event information

.......information supplied by officers/ other suspects

- unbiased interviewing

Front 19

Two changes and the implications:

1 Participants could be participants would be real eyewitnesses and would be interviewed after real accidents. Half the eyewitnesses for each accident could be asked a leading question while the other half were not. This might mean that there is less control altogether in the experiment because it is now a field experiment and there is no control over the actual events in the accident or the speeds or how much broken glass or other damage there was. However balancing conditions out between each actual accident should to some extent account for this. Probably in a real life situation, leading questions will have less effect on memory. This is because there would be so much more information about the event for the person to process - what happens before as well as the accident being much more perceptually rich and so the leading question will have less influence.

Back 19

2 Keep the experiment in the laboratory and with the same procedure but with different participants. The sample could be selected from a larger target sample - adults from the general public. They could be recruited opportunistically or through adverts in the newspaper. Obviously this sample would be more expensive and time consuming to recruit than a sample of students. But it would be more representative of the general population and therefore results would be more generalisable. This sample might have an impact upon the results. Leading questions would have less impact upon people's recollections of speed because they would be more experience of cars and estimating their speeds. However it is possible the leading questions would have some effect - just not as much.