Classic Study: Loftus and Palmer (1974) Flashcards
what was the aim of the study ?
- interested in the idea that human memory is fallible
- wanted to see if memory can be influenced by information which occurs after an event (post-event information)
- this included leading questions
what type of sampling was used in experiment one ?
opportunity
what was the sample of E1 ?
45 students
what was the procedure of E1?
- shown short clip of traffic accident
- ppts divided into five groups
- following the film, ppts asked: “about how fast were the cars going when they ______ (into) each other?”
what was the range in length of the clips shown in E1 ?
5-30 seconds
what were the five verbs used in the question, depending on the group ?
- smashed
- collided
- bumped
- hit
- contacted
what was the independent variable for E1?
the verb used in the question
what was the dependant variable for E1 ?
the mean estimate of the speed per group (mph)
what verb was attributed the highest average speed ?
‘smashed’ - 40.8 mph
what verb attributed the lowest average speed ?
‘contacted’ - 31.8 mph
what were the conclusions from E1 ?
- the experiment provided powerful evidence of the effect that post-event information can have on the recall of an event
- however, the question remained whether this was because the information in the question biased a person’s response or actually altered the person’s memory of the event
why was a second experiment conducted ?
to test whether the post-event information altered the person’s memory or if it just biased their answer
what was the sample for E2 ?
150 students
what was the procedure of E2 ?
- ppts shown one minute film that included a short scan of a car accident
- 50 ppts were asked:”how fast were the cars going when they smashed (into) each other”
- another 50 asked the same but instead of ‘smashed’, it was ‘hit’
- the third control group was not asked to estimate the speed of the vehicles
- after a week, they were asked back in
- given a questionnaire based on the film (not given second viewing)
- randomly placed in it was the critical question: “did you see any broken glass?”
- there was no broken glass in the film
how many conditions were there, and what were they ?
3
- ‘smashed’
- ‘hit’
- control
what was the independent variable of E2 ?
the wording of the critical question
what were the findings of E2 ?
yes, to seeing broken glass:
- ‘smashed’ = 16
- ‘hit’ = 7
- control = 6
what were the conclusions from E2 ?
- findings from E2 suggests that post-event information did not simply create response bias
- post-event information actually altered the person’s memory of the event and generated expectations - such as the likelihood that there would be broken glass
what is a strength of the study ?
aspects of the study were well-controlled
How was the study made internally valid ?
- took place in a lab so there could be strict control variables
- the content of each was standardised for each ppt - as was the stimulus material
what is the competing argument to the strong validity of the study ?
- watching a film of a car crash is unlikely to produce the same emotional response as witnessing one in real life
- there wasn’t the same amount of emotional investment- there was no risk of someone going to prison for the answer they gave - makes them more careless with their answer
- the lab study took away from the ecological validity
assess the applicability of the study
- led to a review of the American justice system
- research into EWT and the fallibility of human memory was a major consideration in the Delvin Report, published in 1976
- recommended that juries should not convict on the basis of a single eye-witness testimony
- this shows that Loftus’ work has implications that stretched beyond the lab
assess the generalisability of the study ?
- they used university students - their level of education may have affected the results as they will most likely be more intelligent than the general average and thus be better witnesses
- students often have less experience with cars - this may have affected their judgement of speed
- either way the sample is not representative of the general population and so is not generalisable
assess the reliability
- Yuille and Cutshall (1986) showed that witnesses to a real-life shooting in Canada gave very accurate reports of the crime four months after the event even though they had initially been given two misleading questions.
- this suggests that, in real situations (when the stakes might be higher), witnesses are less influenced by post-event information.
