Week 4: Eyewitness Recall & Interviewing Flashcards
What is an important thing to note regarding memory?
What people know and what they report isn’t always the same thing
What are the 3 key problems in interviewing settings?
- Witness expectations - problematic expectations re their role leads them to passive and only respond to questions asked (we want witness to volunteer the information they have)
- Interviewer expectations - problematic beliefs re the information that people will be able to provide and recall (memory function) - unrealistic expectations
- Lack of good standardised techniques
Why was the cognitive interview developed?
Was developed to provide a standardised procedure for effective interviewing
Based on core psychological principles of memory and effective communication
Is now the building block for what we know as common best practice
Recall memory is..
Reconstructive
How you ask questions affects the answer you get
What is the cognitive interview designed to do?
- Assist retrieval
- Optimise reporting of information
Both based on the understanding of memory functioning and social dynamics
Reinstatement of context (aiding retrieval)?
Aids retrieval
Scuba divers (Godden and Baddeley)
Might not want to go back to scene of crime, may be too traumatic but there may be ways to mentally reinstate context - think about environment, how they interacted with the environment. Were there smells etc…
Helps to cue and retrieve additional aspects of memory
Multiple and varied retrieval attempts (aiding retrieval)?
When report an experience, they do so in a chronological way - this is just one way to tell a story.
If we can get people to tell in eg. reverse chronological order, varies to way we try to access things from memory. - activate different nodes in memory network so we might activate additional things that we originally wouldn’t
Activating and strengthening multiple retrieval paths is a good way to aid retrieval
Limited mental resources (aiding retrieval)?
We only have so many cognitive resources to deploy at any given time across all of the tasks that require our attention.
If we want witnesses to be devoting majority of resources they have to retrieving, evaluating and optimally reporting information - we want to free up as much of their cognitive resources as possible for this task
- Reduce distractions: conduct interview in quiet room
- Avoid interruptions: external people and the interviewer - break train of thought or shift attention, adds to cognitive load
- Papers looking at if witnesses closing their eyes has an effect - reduce visual distractions. Vision is our primary source of information about environment - reduce distractions through eye closure
Active witness participation (optimising reporting)?
We need active participation - they know what happened, they experienced it so they have all of the information
The interviewer DOES NOT
- this is why the witness needs to be the one driving the interactions
- creating an environment for them to tell their story
BUT witnesses often wait to be prompted instead of being active - assume authoritative figure will drive interaction and their job is to solely answer questions their asked
Should encourage witness to tell their story and give them a chance to do so
Crime relevant information?
It is true that the interviewer has the best sense of what information is relevant to the information
Whilst permitting as much free reign as possible and encouraging the witness to tell their story in their own way - does need to guide to more relevant bits to the story (closed questions or prompts)
- keep interview on track to provide useful and relevant information
Promoting detailed responses?
Unwilling to provide vague information probably because they perceive it to be relatively uninformative and perhaps unhelpful
Interviewer has to recollaborate the witnesses expectations and let them know to give as much detail as can remember and that it will be useful (don’t edit responses)
Interviewer can then sort through all information and determine what is useful
Non-verbal responding?
Verbal isn’t the only way to convey information - especially information is difficult to put into words and articulate. Might benefit from giving information in non-verbal ways
E.g. Acting something out, drawing something
What are the two benefits of drawing to aid recall?
- Can get a lot of information in a very simple drawing - things that might not be articulated by someone providing a verbal account. A lot can be conveyed in a very simple drawing that might not pop up in a verbal description because your’e rushing to get all ideas out within a verbal code
- Can serve as a retrieval cue - as you’re drawing out the environment and you start to think of spatial relationships between things, it is possible that it will cue additional details
What are the 5 stages of the cognitive interview?
- Introduction/rapport building - make them feel comfortable, put them in a state of mind where they feel okay to provide information - also correct problematic assumptions - setting expectations so can provide…
- Open ended narrative
- Probing - for additional details once they have provided their initial account can start to dig deeper into things they have mentioned - not suggesting anything/trying to contaminate memory but trying to probe for additional details
- Review - need to correct witness assumptions and make sure we have a reliable account - go back over events as described by the witness but need to let witness know to correct them if they’re wrong or have misunderstood something - need to feel comfortable to correct if something is wrong
- Closing - make sure they feel comfortable leaving the environment and know that they were appreciated - feeling comfortable to get back into touch if they remember anything else
Is the CI a formula?
No it is a toolbox - collection of recall and reporting tools that can be employed as appropriate
What are the commonly seen experiments involving the CI?
Basic methodology
- participants are typically shown a non-violent video and with hours or days delays, they will get people back in to interview them
- Compare CI to control groups
- or to a structured interview - structure but do not include thinks that enhance reporting etc
What are the general findings of the CI in empirical evaluations?
Enhanced CI elicits more correct information than standard interview - increases the number of correct details by approx. 50-60%
Also leads to interviewers retaining more information than standard interviews - approx. 40%
Advantage holds for person, place, and event descriptions - broad memory benefits
How much more information do detectives that have CI training elicit (field studies)?
60% more information
Nearly all of this information (95%) can be determined to be accurate based on what we already know from scene of crime or if not certain, it is corroborated to be as it aligns with other things
What does the CI decrease?
Number of Q’s/leading Q’s asked
What does the CI increase?
The proportion of open questions
Number of pauses (give witness time to try and remember any additional details - gain in amount of info)
More details and accurate details!!
What are some limitations to the CI?
- Uncooperative witnesses: cannot make people talk - will help people who want to cooperate to produce the best info
- Only useful early on: once witness has provided an initial report and especially once they have rehearsed this, it will become their narrative - will then stay consistent
- Time consuming & hard work: takes a lot of cognitive effort - they need to tailor interview based on what witness reports - what have they said and what can they probe from them
When can’t we do the CI?
When there are many people around the scene of the crime and any of them can know something - CI administration takes training so cannot interview heaps of people at once
The self-administered interview (SAI)?
A booklet that witnesses can work through at their own pace - each page includes prompts that reflect the basic principles of the CI (eg. drawing something, reverse chronological order)
- can give out heaps at the scene
- cost effective
- can then determine which witnesses they want to call in for another interview
Does the SAI actually work?
When compared to free recall:
- SAI elicits more information with no cost to to accuracy
Delay and the SAI?
All witnesses at crime of scene, some completed SAI straight after the event, some didn’t. When came back in a week later, those who completed SAI recalled more info a week later than those who hadn’t
- Helps to preserve information for a later interview so by the time you can carry out a more structured interview, the witness is better at retrieving information from memory when done SAI
Are multiple perpetrator crimes (MPC’s) common?
Yes they are
What problem do MPC’s pose?
May admit being present at crime but dispute what they actually did - ‘i was there but i didn’t do it, they did’. This is a problem because to prosecute a perpetrator you need to be able to actually link them to specific actions - affects charges and sentences
What happens if memory errors occur in witnesses?
If other evidence such as CCTV contradict a witness, it may discredit them as a reliable witness
- Errors are a natural part of these stressful situations but they do pose problems
Should not be considered an unreliable witness
What is a timeline technique?
Physical timeline - eg. strip of cardboard on table
Can attach report cards (action cards and person description cards) to this timeline
Can start anywhere and rearrange as needed
Is the timeline technique effective?
- Yes
Increase in correct information recalled compared to other techniques (free recall included) - record card is similar to timeline technique and gave similar but not quite as strong results - Significantly reduced sequencing errors
How can the timeline technique be used for multiple perpetrator crimes?
Separating out who did what and when
What are the 3 distinct ways that we can conceptualise memory?
All related but are different
- Stored information
- Retrieved information
- Output on a memory test : what we report when asked a question
What is strategic regulation of memory outputs?
When asked a question we don’t just regurgitate an answer that contains everything that we know about a topic - there is a strategic filtration process between what we bring to mind and what we choose to tell someone
What are the 2 processes involved in the strategic regulation of memory outputs?
- Monitoring process: before we answer a question we are likely to asses the reliability of our answer
- Control processes: once we have done this we determine if we will report it or not and how much detail we will report
Monitoring and control processes (flow-chart thing)?
INPUT QUESTION - request for information
MEMORY SEARCH - engage in this to see if they can retrieve any of the requested information
Hopefully will retrieve the BEST CANDIDATE ANSWER - strongest memory they have
Then engage in a MONITORING PROCESS - develop a sense of confidence in the probable accuracy in this info (do this automatically)
We then compare this probable accuracy to a threshold level that determines if we are going to report this information or not.. where does this threshold come from?
Around the same time that all of this is happening, we have another process at play - we develop a response criteria - reflects context specific expectations and task demands and payoffs (instructions, incentives, beliefs influence)
We then volunteer this information or withhold it - if not, we go back to look for another best candidate answer
What affects confidence of own retrieved information?
Reliance on heuristics - if we take ages to think about what the perpetrator was wearing, we may have less confidence in this answer
Fluency or the ease in which something comes to mind - will be more confident
What are the 2 key control processes that ultimately determine what the witnesses report?
- Regulating report option - they have the ability to decide what to report - can choose to just withhold things (all or nothing)
- Regulating grain size (or precision) - if confident can give detailed report, if not confident or not as strong of a memory probably will give a vague response
Balance between being informative and being accurate
What are report option and grain size influenced by?
Confidence (heuristics)
Constraints imposed at test by interviewer (how we frame questions)
What grain size is the typical information given in eyewitness settings?
Tend not to give coarse grain info - perception that not useful
What are some key issues with interviewing characteristics and timing on recall?
Interview format (e.g. closed vs. open Q’s)
Number of interviews (what is omitted in first interview typically doesn’t resurface - grain size and report option vary across interviews eg. less detail as memory declines
Delay to first interview and intervals between interviews - effective early interview crucial for activating and maintaining trace strength
Closed interviews?
Increase reporting, decrease reliability
- suggestiveness, prompting, guessing
OPEN INTERVIEWS ARE RECOMMENDED PRACTICE - putting the witness in charge
Testimonial consistency?
Witnesses are usually interviewed over a number of occasions
- at scene, follow up, if charge prosecutor will have meeting with attorney, in court
Over these multiple interviews, testimonial inconsistencies are highly likely - contradict themselves, or they will give new details that weren’t initially mentioned
Most of us are not likely to be convinced by witnesses who do these things
Inconsistency is considered to be….
A hallmark of inaccuracy
What % of witnesses produce contradictions?
97% - almost all!!
But just because they may contradict themselves on one detail, doesn’t tell us about their global accuracy or reliability of their memory on other details
What might reminiscence be a reflection of and why?
Maybe reflecting external influences instead of memory
Seems to violate basic principles of memory as memory weakens overtime and from this we should recall less over time
How many witnesses display reminiscence?
98%
Why might reminiscence occur?
Throughout multiple interviews with different types of questions and probing etc, likely to cue additional detail
Studies have shown that these details can have high accuracy
