Loftus And Palmer Flashcards
Procedure of Loftus and Palmer
Loftus used two experiments for her study
The first experiment consisted of 45 students in five groups of 9.
They were shown 7 clips of traffic incidents ranging from 5-30 seconds, in a random order.
After watching the clips they were given a questionnaire which asked them leading questions to describe the clips they had just watched.
Then a series of critical questions regarding the estimated speed that the cars were going when they… ‘hit’ (group one)
The other groups were given other verbs ‘contacted’ ‘smashed’ ‘collided’ and ‘bumped’
The remaining questions were filler questions, used to disguise the experimental aim.
The independent variable was the verbs used in the leading question and dependent variable was the estimated speed in mph.
The second experiment consisted of 150 students, split into 3 groups of 50, they were shown a traffic accident lasting 4 seconds.
After the screening they were asked how fast the cars were going when they ‘crashed’ into each other
The other group’s verb was replaced with ‘hit’ and the last group was not asked to make an estimate, this was the control group.
A week later the groups were brought back in to answer a further question, which was if they had seen any broken glass in the clip, this was the dependant variable. The independent variable was the verb used to describe the accident.
Context and Aims of Loftus and Palmer
Carmichael et al conducted a study in 1932, found that giving participants a picture with a different accompanying word, affected how they reconstructed the images. Concluding that reconstruction can be affected by different accompanying labels.
The Devlin Committee in 1974 found when EWT was the only evidence there was a 74% conviction rate.
Prior to conducting this study Loftus had conducted many other studies regarding EWT, in 1974 she conducted a study with 3 groups of mock jurors to listen to evidence and vote on a guilty/non guilty verdict on a mock shop robbery. One group did not have any EWT, the second group had EWT which was discredited as the witness had bad eyesight and the last had initial EWT testimony and a reliable witness, it was found that 72% had voted guilty verdict for the last procedure, however 68% still counted the discredited EWT.
Loftus’ aims were
To see how accurately people remember details of a specific event.
Whether people were accurate at reporting numerical details regarding a traffic collision
Whether different verbs in leading questions could affect how participants (EW) recall damage.
Findings of Loftus and Palmer
Gained quantitative data,
Experiment one, investigated the affect of verbs on leading questions.
Using 5 different verbs, varying by group collided, hit, smashed, contacted and bumped, Loftus and Palmer created estimates from each verb.
Loftus and Palmer found that smashed had the highest estimated speed at 40.8mph and contacted with the lowest at 31.8mph.
From experiment one, Loftus and Palmer concluded that leading questions can distort memory, if smashed was used in a question for EWT they may reconstruct the memory incorrectly based on the verb.
Experiment 2 was testing whether leading questions could distort people’s memories of an event.
Participants in the first two groups were asked a question containing ‘smashed’ or ‘hit’ and those in the third were not asked a question as they were the control group.
A week later they were brought back in to answer a question about broken glass, similar to EWT.
16/50 in the smashed group claimed to have seen broken glass.
7/50 in the hit group and 6/50 in the control group.
From experiment two it was concluded that leading questions can distort a persons memory up to a week later.
Evaluation of the methodology of Loftus and Palmer
In the first experiment Loftus and Palmer screened the clips at random, this was beneficial by reducing order effects such as bored ness and tiredness which can act as confounding variables on the study, however it is a strength as ppts were more likely to give genuine answers. However a weakness of randomised procedure meant that when they were asked questions about the study less important variables were focused on, this is different to real life EWT.
The Loftus and Palmer experiments were ethically sound as participants had given informed consent, they were not exposed to real life events but shown videos which lowered their likeliness to distress and protected them from emotional harm, also they had been fully debriefed after the study finished, so they weren’t deceived about anything.
Reliability: Loftus and Palmer used a standardised procedure, they used videos and questionnaires which could be replicated with different groups of people, to find consistency in the results.
Validity, it could be said that there is low ecological validity as the participants were not observing a real incident, if they were it would be likely that they would have more motivation to recall certain details such as if there was broken glass.
There was good validity however as the control measures put into place meant that the participants responses were affected by the change of verb for example the findings for ‘smashed’ and ‘contacted’.
Sample the sample for experiment one was quite small with only 45 students, this meant that it is unrepresentative of students and not generalisable to other groups.
Whereas the sample of experiment 2 was larger with 150 students, it is more likely to be generalisable to students, but still not generalisable to people of other ages and occupations.
Alternative evidence of Loftus and Palmer
Loftus and Zanni conducted a study with a similar procedure to before, participants were shown a video of a traffic incident and then one group were asked whether they had seen ‘a’ broken headlight and the other group was asked whether they had seen ‘the’ broken headlight.
It was found that most of the participants who were asked if they had seen the broken headlight were more likely to answer ‘YES’ as ‘the’ implies the existence of a broken headlight.
Control of all extraneous variables similar to the original Loftus and Palmer study.
Yullie and Cutshall conducted a study similar to Loftus and Zanni, however this was on real Eye Witnesses of a gun shop robbery where the owner was injured and robber murdered. 13 out of 21 eye witnesses interviewed by police had agreed to take part in a study 5 months later, where the researchers asked them two misleading questions, I.e did you see ‘a’ yellow quarterpanel and did you see ‘the’ yellow quarterpanel?
This study refutes Loftus and palmers study as the findings show that the participants were not affected by the misleading questions. A strength of this study is that it real witnesses of a crime and gave it ecological validity, it could also be generalised to real life.
However the study did not have a large amount of participants , so may not be generalisable to the wider population.
Unethical to distort the memory of people who have witnessed a real life crime for psychological research.
