EWT & Cognitive Interview

• 30 years of research suggest that eyewitness testimony is unreliable. Wells (1998) studied 40 people who were convicted and then later released as a result of DNA evidence. More than 90% of those convicted were convicted because of eyewitness testimony.
• Ronald Cotton spent 11 years in prison for two counts of rape which he didn't commit. He was picked out of a line up of two victims who misidentified him. Bobby Poole later confessed to the crimes.
• Johnson and Scott - weapon focus effect

  • Group 1 overheard a heated argument in the next room, a crash of equipment and then a man emerging with a paper knife
  • Group 2 overheard a disagreement about lab equipment and then a man left holding a pen.
• Yuille & Cutshall (1986) - 21 witnesses observed a shooting in which one person was killed and a second seriously wounded. Little change in eyewitness testimony after 5 months.
• Loftus et al (1978) - participants watched a video of an accident at a T-junction. They asked questions such as 'how fast was car A going when it hit car B'. The wording of this question led them to believe that there had been a collision. This resulted in higher estimates of speed compared to control group
• Lindsay & Johnson (1989) - participants read a passage describing a crime committed by either a tall or short suspect. Later they saw a taller/shorter suspect and identified them more accurately if their initial description matched the height of the suspect.
• Gabbert et al (2003) investigated the effect of post-event discussion on the accuracy of eyewitness testimony. 60 students from Uni of Aberdeen and 60 older adults from community. 71% of the witnesses in the post event discussion group recalled information that they had not seen.
• Fisher (1987) looked at real interviews by experienced police detectives in Florida for 4 months - and found that interviewees were bombarded with short, direct and constantly changing questions.
• Researchers showed 80 children between the age of 8 and 10 a video of a magic act. Those interviewed a day later with the cognitive interview recalled more details correctly, and were resistant to leading questions.
• COGNITIVE INTERVIEW

  • Recall everything
  • Context reinstatement
  • Reverse order
  • Change perspective
• Geiselman (1985)

  • Mean number of items correctly recalled with a standard interview: 35.58
  • Mean number of items correctly recalled with a cognitive interview: 41.67
  • Mean number of items incorrectly recalled with a standard interview: 8.61
  • Mean number of items incorrectly recalled with a cognitive interview: 8.57