Loftus and Palmer decription

• Methodology - experiment, lab, opportunity sampling, students, independent group design, ex1 - iv verb, dv mean speed, exp2 - iv verb, dv yes or no to broken glass.
• Procedure exp1 - 45 ppts, 7 film traffic accidents, questionnaire, an account of accidents just seen, specific q's randomised, and one critical question, about how fast the cars were going when they, smashed, hit, collided, contacted, bumped. speed estimates recorded and mean calculated. 9 in each verb.
• Procedure ep2 - 150 ppts, one clip multiple crashes, asked to describe accidents and asked about the speed, 50 asked smashed, 50 asked hit, 50 asked none as control group, 1 week later asked about clip did you see any broken glass, no glass in clip.
• Findings - exp1 smashed 40.8, collided 39.3, bumped 38.1, hit 34.0, contacted 31.8.
• Findings - exp2 broken glass in smashed yes 16 no 34, hit yes 7 no 43, control yes 6 no 44.
• Conclusions - concluded that the form of a question can affect witness' answer due to - response bias factors, verb influenced response. memory representation is altered, verb in Q affects memory so change incident in mind. Exp 2 shows what the eyewitness memory is formed by what they see and hear at the time and post event information to both form a new memory.
• M + P - p - high control over ev's scientific, controlled, lab, controlled clips, increases internal validity, replicable study reliable and controlled, controlled ev's questionnaires randomised reduces demand characteristics.
• M + P - n - issues with the sample biased sample made up of students less experiences drivers more influenced by leading q's. artificial study lack ecological and external validity.
• F + C - p - high internal validity, cause and effect can be established, high reliability, lab location, standardised procedures can check for consistent results. Support loftus and palmers study, watched a car accident no broken headlight asked if they saw the/a broken headlight leading q 17% recalled wrongly.
• F + C - n - low external validity different to witnessing real life, difficult to generalise findings to real life, issues with the sample made up of students not representative lack external validity. Contradict loftus and palmers study eyewitnesses for an armed robbery still remembered it accurately 4 months later.
• E + S implications - p - ethical, car accidents at moderate speed not a real incident protected from experiencing stress/anxiety. Social significant changes in police interviews from leading q to cognitive interview to eliminate leading q's, improving eye witness testimony reduce the number of wrongful convictions.
• E + S implications - n - ethical, lack of valid consent, did not reveal the true aims of the study so ppts didnt give valid consent ppts may not have participated if they knew the true aims. Social, eyewitness mistakes used in criminal trials, if wrong offenders could be set free but a criminal high risk to society.