Lecture 9 - Memory in forensic situations

Lecture 9 - Memory in forensic situations

1. Loftus et al 1978

   • Phase 1:Group A see slide A, Group B see slide B. Phase 2: Ss asked questions in phase 2: misinformation: “Did another car pass the Datsun when it was at the Yield sign". “Did another car pass the Datsun when it was at the Stop sign". Phase 3: Forced choice recognition (pairs of slides): 75% of consistent Ss were correct, 41% of inconsistent Ss were correct (i.e. 59% choosing what Q not what they saw). False memory been created. Also remember Lotus and Pickerell mall study - autobiographical memories can also be false.
     • What has happened to original memory? Cohen 1988 considered several hypotheses.
       • 1. Vacant slot - the original event was not remembered at all. Not replacing it, just filling it. McCloskey and Zaragoza 1985 - argued that hypothetically with no question, 40% of a group would notice sign, and therefore get it correct, whereas 60% wouldnt notice it (have a vacant slot) and so would have a 50% chance of guessing it correctly -- 70% correct. In a group that is misled with a question and doesnt notice the sign will get it right 25%, leading to 55% overall. Conclusion - by making reasonable assumptions, data which looks consistent with the vacant slot effect can be modelled. BUT at short term delays memory performane can be 90% or more, suggesting event are often encoded. Unlikely to be many vacant slots. So evidence not strong. ALSO a bias - even those who encoded it are biased towards wrong answer.
       • Demand Characteristics - both original and false memories exist and are equally accessible. Subjects recall misleading information because that is what the experiment demands. But, offering incentives for correct answer does not remove effect.
       • Substitution - the misinformation replaces the earlier information which is then irretrievably lost. This interpretation was favoured by Loftus. (integration in schema theory) But if information has been substituted, it no long exists and is not recoverable…

2. Can information be recovered?

   • Tversky and Tuchin 1989. Phase 1: Group A see slide A, Group B see slide B. Phase 2: Ss asked questions in phase 2: misinformation: “Did another car pass the Datsun when it was at the Yield sign. “Did another car pass the Datsun when it was at the Stop sign. Phase 3: Yes/no recognition test with original, misinformation and novel. All group rejected novel information – lures. Misinformed subjects showed misinformation effect by accepting slides showing misinformation. Misinformed subjects also accepted slides showing original information. Conclusion: both original and misinformation memory traces co-exist. SO new theory - Coexistence. Both exist as competing alternative memories. This is different to the Loftus study as forced choice measures which memory has highest strength, not memory availability.

3. Can original memory be reinstated?

   • Bekerian and Bowers 1983. Can changes in experimental procedure make original information more likely to be produced? Compared typical misinformation procedure (random order) with version where questions asked sequentially. WHY? Random order diminishes the temporal / thematic overlap between encoding and retrieval (i.e. encoding specificity). In Loftus expts, slides are presented for recognition in random order. Therefore temporal / thematic overlap between encoding and retrieval is not good - retrieval failure more likely. Lack of relevant retrieval (temporal/thematic) context leads to most recent memory being retrieved (misleading info). When asked sequentially, many fewer effects altogether and misinformation effect disappears.

4. Other ways to increase resistance to misleading information

   • Commitment to choice
     • Opportunity to commitment to choice before misleading information is presented reduces its influence (Loftus 1977).
   • Rehearsal
     • Does rehearsal/retrieval attempt strengthen the original memory trace? Yes. Or merely social effect? Subjects unwilling to contradict themselves?
   • Time
     • Time - if misinformation is presented immediately after viewing the event, around 50% of subjects are able to resist its influence. This figure drops to 20% one week later.
   • Misinformation most potent when original memory is weak

5. Apply to real life forensic situation

   • Interviewing in forensic situations needs memory reports to be accurate and complete. To maximise accuracy and completeness, interviewing should be based on principles of cognitive psychology. - a memory trace is composed of several features and the effectiveness of a retrieval cue is related to the amount of feature overlap with the encoded event. (encoding specificity) - that there may be several paths to the encoded event so information not available be one cue may be accessible by another. Geiselman 1985

6. The Cognitive Interview, Geiselman 1985

   • The original CI had four principles to use with witnesses:
     • Mentally reinstate the context of the event. Context includes physical and mental states, feelings etc.
       • Evidence: Encoding specificity, context-dependency effects, schema etc
     • Report every detail, even if it seems trivial or irrelevant
       • Evidence: e.g. Wagenaar 1986 on cues, encoding specificity
     • Report the event in different temporal orders, moving back and forwards in time.
       • Evidence: e.g. Bekerian & Bowers (1983); schema;
     • Describe the event from different points of view or perspectives. (e.g. from view of another bystander or perpetrator)
       • Evidence: e.g. schema, Andersen and Pichert (1978)
   • Geiselman 1985 - CI produced more correct info without increasing incorrect info. BUT later studies (Memon et al 1997) have suggested the CI generates both more correct and incorrect information
   • What makes the difference?
     • Witnesses instructed to recall event in one of four ways. Either: write down as much as they could remember, include everything, also reinstate context or all of above. Full produced best results (CI).
   • In the field
     • Fischer 1987: CI aided police officers in the field, but a recent study showed maybe not being used correctly (Dando et al 2008)
   • Enhance CI, Fisher and Geiselman 1992
     • Includes the 4 cognitive techniques but also strategies for improving interviewer-witness communication. - great rapport building between interviewer and witness - `transfer of control‘ from the interviewer to the witness.
   • Disadvantages: interviewer has to be trained, takes longer, is it accurate?, difficult on children