loftus and palmer Flashcards
what was the aim of loftus and palmers experiment 1
to see if the use of leading questions will change a witnesses estimate of speed after watching a car crash
describe the method/design used in experiment
lab experiment with an independent measures design
IV: verb used in the leading question (smashed, collided, bumped, hit, contacted)
DV: speed estimate in MPH
who were the sample in experiment 1
45 students from washington university
describe the procedure for experiment 1
participants were shown 7 films of car crashes taken from training videos used by the seattle police department and evergreen safety council. the speed of the cars were known in 4 of the films, these were 20, 30, 40 and 40mph.
after watching, participants had to write a written account of what theyd seen and were also given a questionnair full of filler questions and one critical question- ‘‘how fast were the cars going when they…into each other
both quantitative and qualitative data was collected but only the quantitative data was analysed
describe the results in experiment 1
we are generally poor at guessing speeds- the estimates of speed were not affected by the actual speed of the car. e.g when the actual speed was 20mph, the estimated guess was 37.7mph and when the actual speed was 40mph the estimated speed was 36.1mph
estimates of the cars speeds did significantly vary based on the verb in the critical question e.g smashed had an estimate of 40.5mph whereas contacted had an estimate of 30.1mph. the difference in 10mph could be the difference in whether a driver is arrested for dangerous driving vs it just being an accident
what did loftus and palmer conclude after experiment 1
there are two possible reasons for the variation in estimate of speed:
response bias- when the participant is unsure what speed to estimate the verb gives them a clue as to whether they should guess high or low
memory distortion- the verb used in the question actually alters the participants memory of the crash
what was the aim of loftus and palmers experiment 2
to see if a leading question will actually distort a witnesses memory
what was the method/ design of exp 2
lab experiment with an independent measures design
IV: the question asked (smashed, hit or control (not asked about speed))
DV: response (yes or no) to whether they had seen broken glass
who were the sample
150 students from washington university
describe the procedure of experiment 2
all participants watched only one video of a car crash. they were then asked to write a written account of what theyd seen and then fill out a questionnaire, again with one leading question and the rest were filler questions.
a week later participants returned and answered a further 10 questions that included one critical question -‘‘did you see any broken glass’’ (there was none)
describe the results
number of people who reported broken glass:
smashed-16
hit-7
control-6
the number of people who remembered broken glass who were asked the smashed question was over double the number of people who remembered it in the hit condition
people in the control group still remembered broken glass despite not being asked a leading question- memory is not perfect
what did loftus and palmer conclude after experiment 2
the verb in the leading question caused memory distortion
there are two kinds of information that go into someones memory for a complex occurrence
-info obtained from the actual event itself e.g the actual crash
-info supplied after the event e.g the leading question
over time the information can merge and create ‘‘one memory’’ so you cant tell which is part of the event itself and what came after. this is known as the reconstructive hypothesis
comment on the sample in the study
the sample is very restricted as they are all from the usa (ethnocentric) and it is also age biased as they are all students- this may be an issue as younger people may have less experience driving than older people, therefore their estimates on the speed of a car may be more likely to be impacted by the verb, therefore results may not be generalisable to the target population
however the sample size is fairly good with 195 participants across both studies, making results easier to generalise as it makes it less likely that results regarding memory distortion only applies to a few people
comment on the reliability of the study
high in reliability as standardised procedures were used e.g all participants watched the same video and completed the same questionnaire (only the critical question was different). these standardised procedures ensures the experiment can be replicated
comment on the ecological validity of the study
low in ecological validity as it is a lab experiment with controlled conditions- viewing a staged car crash in a lab setting is very different to seeing one in real life e.g it is missing all of the emotion you would feel if you saw it for real.
the participants attention was already fixed on the screen so they were waiting for something to happen, therefore they may not have responded to it in a way they would seeing it in real life, making the study invalid
