LOFTUS & PALMER AIMS AND CONTEXT
to investigate the accuracy, or inaccuracy of memory. particularly the effect of leading questions on the estimate of speed.
experiment 1: to see if the estimates given by participants were influenced by the choice of verb.
experiment 2: to see if leading questions bias a person's response or alter the memory.
in court people may believe they are telling the truth but their memories arent always 100% accurate, inaccurate eyewitness memory may be a leading factor to fals convictions.
Fillmore(1971) suggested words 'smashed' and 'hit' implies different rate of movement to people, influencing their estimate.
LOFTUS & PALMER PROCEDURE
- 45 students shown 7 film clips of different traffic accidents
- after each clip participants answered questions including critical question:
- "about how fast were the cars going when they _____ each other?"
- the blank was filled with HIT, SMASHED, COLLIDED, BUMPED, CONTACTED
- new set of 150 students
- shown a 4 second film of multiple car crash, then asked questions
- GROUP1- 'how fast were the cars going when they SMASHED into eachother'
- GROUP2- 'how fast were the cars going when they HIT eachother'
- GROUP3- a control roup were not exposed to any question
- one week later the paticipants were asked further questions. the critical question was 'did you see any broken glass?' there was no broken glass but those who thought the car was faster might expect there too be.
LOFTUS & PALMER FINDINGS & CONCLUSIONS
EXPERIMENT 1 EXPERIMENT 2
- smashed 40.8 smashed- 16 yes - 34 no
- collided 39.3 hit- 7 yes - 43 no
- bumped 38.1 controlled - 6 yes - 44 no
- hit 34.0
- contacted 31.8
- the form of a question can affect a witness' answer to a question, response bias factors and memory representation is altered - a critical word can alter memory and perception.
- experiment 2 suggests that the effect of leading questions isnt due to response bias, but its due to alteration of the memory, because different questions led participants too have an altered perception of the seriousness of the accident. understood in the relation too carmichael et al1932
LOFTUS & PALMER METHODOLOGY
DESIGN- lab experiment and independant measures
ETHICS- didn't gain fully informed consent from their participants
RELIABILITY- other studies have found similar results (Loftus & Zanni 1975)
VALIDITY- criticised for low ecological validity
- generalisability - in a lab, contrived setting
- mundae realism - participants observed film clips rather than real life where observers would feel scared or anxious
- demand characteristics - participants were aware they were in an experiment and may have suspected that they would be questioned about the film clips
SAMPLING- american college students
- students may have better memory - affects generalisability
- less driving experience
- oppurtnity sample
LOFTUS & PALMER STRENGTHS AND WEAKNESSES
- controlled experiment - controlled IV
- standardised film clips and questions
- control of extraneous variables
- mundae realism - may be asked leading questions in court
- less motivation to remember film then real events
- less emotional effect
- low ecological validity
- sample were unexperienced to the topic
LOFTUS & PALMER ALTERNATIVE EVIDENCE
Loftus and Zanni (1975)
asked paticipants if they saw 'a' broken headlight (7% said yes) or 'the broken headlight (17%)
showed participants man stealing red wallet, 98% identified colour correctly even if the wallet was described as brown participants persisted the wallet was brown
Loftus et al (1978)
showed participants were affected by visual leading questions. shown a car at a stop sign, then asked about a yield sign - gave less accurate answers
showed a film of a purse being stolen, only 14% picked the right person out of identity parade
Yullie and Cutshall (1986)
inerview 13 people, 4 months after armed robbery. despite 2 misleading questions - no change