WJEC AS Psychology PY2 - Loftus and Palmer (1974)

Definitions:

• eye witness testimony - evidence/statement given by a witness to a court under oath
• leading question - question that either by form or content suggests answer desired, or leads witness to desired answer

Importance: to understand how eye witness testimonies can be influenced as if a witness testimony is wrong, it can lead to a wrongful conviction.

Previous Research:

• time, speed and distance are inaccurately reported
• speed is especially difficult to judge for humans
• Marshall (1969) - test with Air Force Personnel; knew they had to guess speed of moving automobile. Guesses ranged from 10 - 50mph; actual speed was 12mph; even if we're expecting to judge speed, humans are bad
• Fillmore (1971) - verbs "hit" and "smashed" imply differential rates of movement

Aims:

1) to investigate the effect of phrasing of a question on a witness account of speed

2) to see if leading questions can cause answer to suit the question (bias response) or change/distort memory

EXPERIMENT 1 - looking at how the verb used affects speeds estimates

• STEP 1: 45 US College students participated in a LAB EXPERIMENT
• STEP 2: students shown 7 films showing traffic accidents
• STEP 3: divided into 5 groups (independent groups design)
• STEP 4: students answered a questionnaire, each group had a different verb in the critical question asking the estimation of speed of car (IV); verbs were: H S B C C
• STEP 5: Estimated the speed of the car (DV)

EXPERIMENT 2 - if the verb "smashed" will alter memory of accident

• STEP 1: 150 US College students participated in LAB EXPERIMENT
• STEP 2: shown film of multiple car accident (film = 1 minute, accident = 4 seconds)
• STEP 3: divided into 3 conditions (independent groups design), 50 asked question with verb "smashed", 50 with "hit" and 50 asked no questions
• STEP 4: asked questions if in the 2 conditions 
• STEP 5: a week later, all participants were asked question, "did you see any broken glass?"

Results from Experiment 1: verb changed = influences judgement of speed

• verb that produced the highest speed was "smashed" = 40.5mph
• verb that produced the lowest speed was "contacted" = 31.8mph

Conclusions from Experiment 1:

• changing the verb = response-bias: if subject is unsure what to answer (e.g. 30/40mph), verb indicates right estimate (smashed = 40mph)
• verb "smashed" distorts memory representations to more violent car crash, a subject could remember details that were never there?

Results from Experiment 2:

• those who saw broken glass with verb "smashed": 32%, with verb "hit": 14%, with no prior questions: 12% 

Conclusions from Experiment 2:

• phrasing of a question distorts mental representations of event

STRENGTHS

• a lab experiment: cause (IV)-effect (DV) relationship established; extraneous variables controlled i.e. angle/distance of car crash same for all participants
• quantitative data (objective) easier to analyse and compare  
• standardised procedures easily replicated; increases RELIABILITY of results
• independent groups design means control of demand characteristics 

WEAKNESSES

• low ecological validity: difficult to generalise; not in a courtroom; testimony can affect conviction; in a lab, there is a lot less pressure
• lacks mundane realism: video of car crash, accurate memory = real car crash, impact of witnessing crash
• demand characteristics = aware of study; affect behaviour accordingly
• low population validity: all US College students make them ethnocentric and difficult to generalise to all age ranges of population
• ethics: deception, and no informed consent

1. Loftus and Zanni (1975) - SUPPORTS

• showed participants car accident; some asked "did you see A broken headlight" and others asked "did you see THE broken headlight?"
• 7% of those asked about 'A broken headlight' reported seing one
• 17% of those asked about 'THE broken headlight' reported one
• no broken headlight in the video

Criticism of L+P's experiment is that judging speed is complex, therefore led easily by leading questions. Research demonstrates leading questions do distort memory. SUPPORTS L+P as provides evidence that leading questions alter response

2. Loftus (1979) - CONTRADICTS

• participants shown pictures of man stealing red wallet from bag
• 98% able to identify the correct colour
• asked leading questions after recall, still persisted the wallet colour was red

CONTRADICTS: suggests leading questions have limited memory effect: colour is more difficult to affect a person's memory as less ambiguous whereas speed isn't

3. Buckhout (1980) SUPPORTS

• 13 second clip of mugging shown on TV; identity parade of 6 suspects shown; viewers asked to phone in saying who they thought had done it
• 6 suspects, chances says 17% guess right suspect; only 14% correct

Criticism of Loftus and Palmer was demand characteristics which affects behaviour; study SUPPORTS as shows recall in real life is still poor when unaware of expt; didn't get person right more than chance would predict

4. Loftus and Pickrell (1995) DEVELOPS

• interviewed participants about childhood events, planted a false memory of being lost in a mall as a child (imaginary event)
• 20% believed event happened, some even believing it after being debriefed

DEVELOPS EWT research; L+P studied leading questions affecting memory; this study shows leading questions can plant memories that were never there: verbal information can alter memory