Improving the Accuracy of LTM

• how is EWT considered inaccurate?


  - numerous research studies have indicated that EWT lacks accuracy due to factors such as anxiety, leading questions and post-event contamination
• standard interview - Fisher (1987)


  - defined the standard interview from observations of police interviews in Florida

  - police asked quick, direct and closed questions; leading the recall with witnesses being unable to talk freely and being frequently interrupted
• cognitive interview - Fisher and Geiselman (1985)


  - context reinstatement: mentally returning to the crime scene to trigger environmental/emotional contextual cues

  - report everything: all details, even if the seem irrelevant should be mentioned

  - recall from a changed perspective: consider the perspective of another witnesses/the perpetrators to disrupt schema

  - recall in reverse order: switch to different chronology, to check the accuracy of recall and challenge expectations
• A03 - Fisher (standard vs. cognitive interview)


  - 7 detectives trained in CI were compared with 9 detecting using SI

  - found CI detectives produced 47% more information in real interviews after their training and 63% more information than SI detectives

  - suggests CI effectively enhances memory of witnesses in the real world
• A03 - Kohnken (meta-analysis of CI vs SI)


  - meta-analysis of 42 CI studies, including over 2500 interviews

  - found a significant increase in amount of correct information recalled; however, a significant increase in the amount of incorrect info recalled resulted in a similar accuracy rate: CI = 85%, Si = 82%
• A03 - criticism of CI

  - time-consuming, requiring more time than officers have available

  - also requires significant training and investment which diverts officers from their normal work