Improving Accuracy of EWT

• Fisher (1987)


  Assessing standard interviews from police interviews in Florida, observed that witnesses couldn't recall freely, with police leading recall with direct, closed questions
• Fisher + Geiselman (1992)


  Establishing the Cognitive Interview

  • Context Reinstallment: mentally return to crime scene, triggers environmental cues
  • Report Everything: record all details, even if irrelevant
  • Recall from Changed Perspective: consider perspective of other witnesses to reduce schemas
  • Recall in Reverse Order: switch chronology to check accuracy of recall and challenge expectations
• (+) A03: Fisher


  7 Detectives using Cognitive Interview, and 9 Detectives using Standard Interview

  • Found that Cognitive Interview produced 47% more new information, and 67% more total information than Standard Interview
• (+) A03: Milne + Bull


  Found aspects of Cognitive Interview produced similar amounts of recall, but Context Reinstallment and Report Everything produced the most correct recall, suggesting CI is cumulative
• (-) A03: Daily Crimes

  Cognitive Interview is not helpful in recognising the suspect, on smaller, daily crimes this isn't useful