Loftus & Palmer (1974)

• Methodology:


  Experiment 1- 45 students. Experiment 2- 150 students. Lab experiment with independent groups design. IV: verb in critical question. DV: average speed estimate (mph) & broken glass (yes or no).
• What is a leading question?

  Question that either by its form or content, suggests to the witness what answer is desired, or leads him to the desired answer.
• Procedure of experiment 1:

  45 students (5 groups). Each participant watched 7 films on a traffic accident. Each film 5-30 seconds. After each film, each participant given questionnaire. First question asking for an account of what they'd seen. 2nd part was series of specific questions about the film. Critical question about speed of cars: 'How fast were the cars going when they _ eachother?'. Blankspace filled with 1 of the words: smashed, collided, bumped, hit & contacted.
• Procedure of experiment 2:

  150 students (3 groups). Shown film of a multiple car crash. Asked set of questions, including critical question about speed. Group 1 asked, 'How fast cars going when smashed into eachother?'. Group 2 asked, 'How fast were cars going when they hit eachother?' & group 3 control group so not exposed to any question. Week later, participants returned to lab. Asked, 'did you see broken glass?' ( no broken glass in film).
• Evaluate methodology & procedures:

  Casual conclusions drawn as experiment controlled in lab. Large sample used in experiment 2 (150) so more reliable, kept some info back to ensure validity of results. Hard to estimate speed of a car (might influence their responses more than a simpler task), smaller sample in experiment 1, students may not be representative of whole population & may respond differently, lacks ecological validity-findings may not represent real life.
• Ethics:


  Low risk of harm (psychological, due to watching car accident clips) & lack of valid consent (mild deception).
• Findings of experiment 1:

  Mean speed estimate calculated for each experimental group. Group given word 'smashed' estimated higher speed than other groups (40.8 mph). Group given word 'contacted' estimated lowest speed (31.8 mph).
• Findings of experiment 2:


  Part 1: Participants gave higher speed estimates in 'smashed' condition. Part 2: Participants in 'smashed' condition were more than twice as likely to report seeing broken glass- 16 reported they had seen, 34 didn't. Group 2 ('hit condition'): 7 reported they'd seen & 43 reported they hadn't. Group 3 (control): 6 had seen & 44 hadn't.
• Conclusions:

  Form of question can markedly & systematically affect witness's answer to question. 2 explanations: 1.) Response-bias factors. Different speed estimates occur as critical word influences or biases response. 2.) Memory representation is altered- critical word changes person's memory so their perception of accident is affected. Some critical words lead to perception of accident being more serious. Shown in experiment 2.
• Carmichael et al (1932):

  Evidence for effect of verbal labels on to-be-remembered forms. Verbal labels cause shift in way info is represented in memory in direction of being more similar to suggestion given by verbal label. Participants shown set of drawings & provided with verbal description. When asked to redraw image, resulting object was affected by verbal label.
• Why could a causal relationship be established?

  Controlled experiment- deliberately manipulated IV (verb used to describe impact) to see causal effect on DV (estimate of speed) & draw causal conclusion. Lab-confounding variables controlled so any change in DV is due to only IV & not other factors.
• Evaluate the sample:


  US college students-other groups of people may be more or less prone to being affected by misleading info e.g, age differences (may be a consequence of source monitoring). Eye-witness typically acquires info from 2 sources: observing event itself & from subsequent suggestions (misleading info).
• Schacter et al (1991):

  Compared to younger subjects, elderly people have difficulty remembering source of their info, even though their memory for the info itself is unimpaired. May become more prone to effect of misleading info when giving testimony.
• Evaluate the ecological validity:


  Watched film clips of accidents (not same as witnessing real accident). People don't take task seriously &/or they're not emotionally aroused in way they would be in real accident. Findings may not represent real-life (lack ecological validity).
• Foster et al (1994):


  Found if participants thought they were watching a real-life robbery & thought their responses would influence trial, their identification of robber was more accurate. (IRL, EWT may be more accurate).
• Yuille & Cutshall (1986):


  Found evidence of greater accuracy IRL. Witnesses to an armed robbery in Canada gave very accurate reports of the crime 4 months after event even though they had initially been given 2 misleading questions. Suggests misleading info may have less influence on real-life EWT.
• Buckout (1980):


  Conducted 'real life' study involving 2000 participants. Very short film (13 seconds) was shown on prime-time TV. Later, an identify parade was shown on TV & viewers invited to phone in their choice of suspect. Only 14% got it right.
• Braun et al (2002):


  College students asked to evaluate advertising material about Disneyland. Embedded in material was misleading info about Bugs bunny or Ariel (neither character could have been seen at Disneyland as Bugs not Disney & Ariel hadn't been introduced at time of their childhood). Participants assigned to Bugs, Ariel or control (no misleading info). All had visited Disneyland. Participants in Bugs or Ariel group more likely to report having shaken hands with these characters than control. (Misleading info created inaccurate memory).
• Evaluate consent in this study:

  Lack of valid consent (mild deception). If participants had been aware of aims of study their behaviour would have been affected-would have been aware of leading questions & thus more careful in their responses. Would not reflect EWT in everyday life. Researchers justify deception-importance of research lets us understand inaccuracy of EWT. Unlikely that knowing true purpose of study would have led to refusal of taking part. No harm from this.
• Evaluate psychological harm:

  Films clips of accident not as harmful as real accident. Films clips meant they may not have responded to task in way eye-witnesses would IRL. Could have used real accident but this would cause psychological harm not necessarily diffused by debriefing. Long-lasting emotional impact. Study avoided ethical issue of psychological harm.
• Social implications:


  Changed way witnesses questioned by police (no leading questions & free recall used instead), EWT alone from an individual not enough evidence in court- multiple pieces of evidence needed to prevent false convictions, cases where EWT is correct/important may be hindered & people may be less willing to come forward with testimonies as they feel that they won't be believed or doubt themselves.