loftus and palmer (1974) Flashcards
(20 cards)
Aims of study
- to investigate the accuracy we remember the detail of a complex event such as a traffic accident
- to test the hypothesis that the languages used in eye witness testimonies can alter memory, specifically the role of leading questions
Sample
- 45 American students
- 9 in each condition
- may not have been strong drivers due to being young- so speed perception may have been inaccurate (inexperienced on road)
Method
Lab
Procedure of 1st experiment
- split into 5 groups
- shown 7 clips of 5-30 seconds each of a staged traffic accident
- 4 staged
- taken from Evergreen safety council and the Seattle police department
- after watching the clips, the ppts received a questionnaire asking them to give an account of the accident as if they were an eye witness
- then to answer a series of specific questions about the accident
- the critical questiiion was the one that interrogated the subject about the speed of the vehicles involved int he collision
- “about how fast were the cars going when they hit each other” for 9 ppts
- the remaining ppts had the verb hit swapped equally with collided, bumped, smashed or contacted on the condition that the experimental group were in
- a different ordering of the films was presented to each group of ppts ( to control for the practice effect of getting better at estimating mph)
IV and DV
- IV= the verb used in the critical question - smashed, collided, bumped, hit, contacted
- DV= the estimated speed of the car in mph
Results for the mean speed estimates for various verbs
- smashed= 40.5 mph
- collided= 39.3 mph
- bumped= 38.1 mph
- hit= 34 mph
- contacted= 31.8 mph
Results for the staged car accidents WHICH WE KNOW THE ACCURATE SPEED OF
- 20 mph = 37.7 mph
- 30 mph = 36.2 mph
- 40 mph = 39.7 mph
- 40 mph = 36.1 mph
- so people are not very good at judging how fast a vehicle was actually travelling
Conclusion
- the verb used in the critical question influenced the speed estimate
- the more powerful verbs that suggested higher speeds before impact received higher estimates
2 reasons L+P suggested for why leading questions affected memory
- distortion= the verbal label such as smashed may change a subjects memory. Ppts have cognitively changed their memory through re-evaluation of what they already know
- response bias= demand characteristic where the ppt is uncertain to say 30-40 mph and the verb given biases their response towards the higher estimates. Adjusts their estimates to fit in with the expectations of the questions
Aims of the follow up experiment
- to investigate the origin of different speed estimates
- to investigate whether estimates were due to demand characteristics of response bias or memory alteration
Sample of 2nd study
150 students
Method of second study
Lab
Procedure of the second study
- 150 students were split equally into 3 groups
- a film was shown depicting a multiple car accident, film lasted less than 1 min and accident less than 4 seconds
- at end of film, ppts were given a questionnaire asking them to describe the accident in their own words and then answer a series of questions about the accident
- asked critical question of “how fast were the cars going when they hit each other”
- verb hit as changed for smashed, and for the third group there was no interrogation about vehicular speed
- 1 week later, ppts returned without viewing the film and answered series of questions
- critical question here was “did you see any broken glass”- answered using yes or no
- no broken glass, but broken glass is associated with higher speeds- so expected ppts who had smashed question to answer yes more to this critical question
Results of second study
- smashed = 16 yes, 34 no
- hit= 7 yes, 43 no
- control= 6 yes, 44 no
Conclusions
- smashed led to more yes responses and higher speed estimates (10.46 mph compared to 8 mph)
Generalisability
A weakness of lofts and palmer is that there is low generalisability. This is because in the 1st study, a sample of only 45 American students were used. Students are often inexperienced drivers so would not be strong on the roads. Therefore, their speed perception may have been inaccurate compared to a longer term, more experienced driver who has not recently passed their test. Therefore, we cannot generalise results on speed estimates to people with varying levels of driving experience.
Reliability
- standardised procedure of 7 5-30 second films of car accident, 4 of which were staged so have complete control over the presentation of each, so can be replicated to test for consistency.
- uses quantitative data which is easily measurable due to the estimate of car speed in mph- so easy to compare data if replicated
- consistent supporting evidence- loftus and zany found that asking ppts if they had seen “the” broken headlight instead of “a” broken headlight increases the incidence of the ppts saying yes- so strength as it provides consistency to the conclusion that leading questions to witnesses can cause memory to become distorted
Applications
A strength is that it can have positive applications to the police in terms of how they interview witnesses. The study concluded that leading questions can alter memory if something is suggested, even if it didn’t happen- therefore we should advise police to only ask general, open questions with no leading questions which could negatively influence witness testimony.
Has also led to the Devlin report, which has suggested that identification evidence was not sufficiently reliable to make a prosecution. This has changed the way that we interview subjects- try not to use leading questions, preventing wrongful convictions based on distorted memories.
Validity
- low task validity- ppts watched 7 videos, 4 of which were staged, that lasted between 7-30 seconds. This is a weakness as it acts mundane realism as ppts were expecting an accident so could focus on this, but in real life accidents are spontaneous events. Ppts were also not experiencing the emotional response of an actual car accident- meaning that realistic judgement was not collected, lowering validity of results.
- high internal validity- 4 accidents were staged so there is confirmation of speed- can be accurately measured, meaning cause and effect relationship can easily be established between leading questions and judgement of speed (due to high levels of control)
Ethics
A strength of the study is that there was no ethical concerns. Ppts were fully aware of the experiments so gave fully informed consent. Additionally, 4 of the accidents shown to ppts were staged which meant that ppts were not exposed to the stress of a real life accident. This is a strength as it shows that the duty of care towards ppts was upheld and follows BPS ethical guidelines.
