Loftus and Palmer's study Flashcards
What is the overall aim of the study?
To find out the effect of leading questions on eyewitness memory
What is Bartlett’s schema theory?
Our memory for new information is affected by previous knowledge
Memory is reconstructive
What is the aim of the 1st experiment in this study?
Whether using different verbs to describe the collision of 2 cars would affect estimates of the speed they were travelling during the crash
What was the type of experiment and its conditions?
- Lab experiment
- IV: Verb strength used in the critical question (‘Smashed’, ‘Collided’, ‘Bumped’, ‘Hit’, ‘Contacted’)
- DV: Speed estimate of the cars when they crashed (mph)
What tool was used to collect P’s estimation of speed and their account of the incident?
Questionnaire
What was the sample for this study?
45 STUDENTS (divided into 5 groups of 9)
Describe the procedure of the first experiment
- P’s shown films of vehicle crashes
- P’s then had to fill in a questionnaire asking: Give an account of the accident. And answer an array of Qs
- The critical Qs was “about how fast were the cars going when they —- each other?”
What were the findings of this study? (1st experiment)
- Actual speed at which the cars were travelling did not affect speed estimates
- Verb strength significantly affected the speed estimate.
“P’s in the smashed condition gave significantly higher speed estimates (40.8 mph) than P’s in the contacted condition (31.8 mph)
What were the conclusions made from this first experiment?
(Conclusion always use present tense)
- Eyewitnesses are poor at estimating vehicle speed
- Verb strength and the wording of the leading question affected the speed estimates given by the P’s
What is the aim of the second experiment?
To find out whether the strength of the verb in the critical question affected the likelihood of P’s remembering broken glass at the crash site
What was the sample of the 2nd experiment?
150 students (divided into 3 groups of 50)
Describe the procedure of the 2nd experiment
- P’s watch a film of a car crash
-P’s then fill out a questionnaire which asked them to give an account of the accident and other questions - The critical question : “about how fast were the cars going when they — into each other?”
- 3 groups: ‘Smashed’ condition, ‘Hit’ condition, and a controlled group
- P’s invited back to lab after 1 week
- Given a new questionnaire
- New critical question : “did you see any broken glass”
- No broken glass in the video
What were the findings of the 2nd experiment?
More P’s who had seen the verb ‘smashed’ falsely reported seeing broken glass (16 P’s) than P’s who had seen ‘hit’ (7 P’s) or no speed question
What were the conclusions of the 2nd experiment?
(Conclusions always use present tense)
The broken glass findings could not have occurred due to response bias as the broken glass question did not contain the verb cue
Therefore, the language used in a question can cause permanent distortions in memory
=> The fact that memory can be RECONSTRUCTED to incorporate new information after an event suggest that eyewitnesses are not reliable
EVALUATIONS:
Research Method
(+) Lab makes it highly controlled. E.g The same crash films were shown to all P’s .
This makes the study high in internal validity because we can be sure of that the differences in speed estimates were caused by verb strength and differences in film shown to P’s could not act as an extraneous variable
(-) Lab makes it low in ecological validity. Watching car crash on a video is not the same as watching a car crash in real life. E.g. eyewitnesses in real life would be focused on making sure people in the cars were alright, rather than recalling details of the incident.
This makes the findings less applicable to real life eyewitness situations
EVALUATIONS:
Data type
(+) Quantitative data makes it easy to compare speed estimates between verb conditions. E.g. Loftus and Palmer were able to run an inferential test which found a difference in speed estimates between the verb strength condition in E1.
We can draw objective conclusions about the data
(-) Lacks sufficient rich, contextual detail. E.g. We do not know any details about the P’s false memories of the broken glass. We do not know how sure they were that they saw the glass.
EVALUATIONS:
Ethics (Ethical issues)
(+) P’s were protected from psychological harm as the videos of the car crashes were not gruesome. E.g. The clips did not involve distressing scenes of people being hurt.
This improves the reputation of Psychology as an ethical field of science
(-) P’s did not give informed consent. They were not told the exact aim of the study as knowing about the verb manipulation in the critical question would have created demand characteristics. E.g. P’s would deliberately estimate more falsely the speed of the cars because they think that is what the experimenter expects.
However it was necessary to hid the aim to avoid demand characteristics
EVALUATIONS:
Reliability
Uses standardized materials. E.g. The films used were taken from standard driver safety video clips.
This is good as another researcher could easily use the video and replicate the study in the same way to find out if the same results recur.
(Always use same in reliability Qs)
EVALUATIONS:
Generalizability
Only students were used whose speed estimates may not be generalizable to the wider population. E.g. Students are less to be drivers than other adults thus would have poorer speed estimates due to lack of experience with cars.
Therefore, the findings that eyewitnesses tend to be poor at estimating speed could not be generalizable to non-students as they might drive cars more proficiently and can materialize the speed of cars more accurately than students.
EVALUATIONS:
Practical Applications
Influence the way that eyewitnesses are interview. - Suggest that police should avoid using emotive language as this can distort the severity of the event in the eyewitness’ mind.
- Suggests that witnesses should be kept apart after an incident so their memories of the event cannot be affected by each other’s account
=> Prevent source-monitoring error amongst others
