3. COLLECTING EVIDENCE Flashcards
Recognising E-fit faces
Bruce aim
See if there was a difference in our ability to recognise internal and external features of facial composites
Facial composites of celebrities
Bruce method
Lab
Bruce sample
Self selected
30 staff and students from Stirling Uni (M+F)
Bruce design
Independent 3 conditions: 1. internal + external 2. internal 3. external Given ten photos of celebs (targets) Match 40 composite images (4 for each target)
Results from Bruce
No significant difference in accuracy between complete composite and external factors only (35% correct)
Internal factors not so accurate- 19.5% correct
What do Bruces findings suggest
When witnesses are recalling internal features, we may not be able to trust accuracy
Pros of Bruce
Lab= increased control Independent- reduces DC Standardised- replicable Quantitative Useful applications
Cons of Bruce
Small sample- reduced population validity
Reduced ecological validity- no consequences
Reduced task validity- using celebrities
False confession
Kassin aim
Whether psychologically vulnerable people are more likely to make a false confession
Kassin sample and design
79 undergraduates
Independent
Deceived into thinking the experiment was on reaction time
Voluntary confession
No external pressure
Coerced compliant confession
Had enough of questioning so confess
Coerced internalisation confession
Persuaded that you committed a crime
Kassin procedure
Ps tested individually with confederate
Told letters to type on computer- specifically told not to press ALT key, as it will ‘crash’ the computer
3 minted of pressing keys
Experimenter accuses P of pressing ALT key, confederate said they did when they didn’t (pressure)
4 conditions for Kassin
- high vulnerability- 67 letters per minute, accused of pressing ALT
- high vulnerability- 67 letters, not accused
- low vulnerability- slow, 43 letters per min, accused
- low vulnerability- 43/min, not accused
Kassin results
More typing errors in high vulnerability (stressful)
Compliance- least likely in condition 4, most likely in condition 1
Internalisation- 65% in condition 1 said they pressed it
Confabulation- main up details, 35% of condition 1 confabulated
What do findings from Kassin suggest
Two factors cause false confessions:
- stress
- incriminating evidence
Pros of Kassin
Large sample- 79 undergraduates Quantitative and qual data Independent- reduced DC Useful applications- remove stress High control Standardised
Cons of Kassin
Undergraduates- not generalisable
Unethical- deceived + high levels of stress
Reduced ecological validity- no consequence
Standard police interview
Free recall about event
Followed by specific questions on information gathered
Not standardised
Cognitive interview (Fisher)
Structured, standardised interview Four stages: 1. context reinstatement- setting scene 2.report everything 3. recalling in different temporal order 4. recall from different perspectives
Testing the CIT
Fisher, aim
Investigate usefulness of CIT compared with standard interview techniques
Fisher sample
16 detectives
Miami police department
Fisher method
Field experiment
Fisher phase 1
Detectives recorded number of interviews using standard interview techniques (burglary and robbery crimes)
16 split in half- one group trained in CIT (experimental group) other was not (control)
Fisher phase 2
Post training, more interviews were recorded for same crimes as P1
Interview analysed by a team (unaware of condition)
Identified how much info was obtained and match it to corroborative evidence
Fisher results
CIT trained officers gathered 47% more info
CIT produced 63% more information than control group
CIT gather more information and to a higher level of accuracy- therefore train police to use CIT method
Fisher pros
Increased ecological validity- real police
Useful applications- CIT
Cons of Fisher
Lacks generalisability- only experimented on petty crimes
Reduced control
PEACE
Prep + planning Engage + explain Account, clarification + challenge Closure Evaluation
Reviewing CIT
Memon and Higham
Effectiveness of CIT components
Context reinstatement most effective
Recalling in different temporal orders is not beneficial= fabrication
Most effective recall= forward order then reverse
M+H
Comparing interviews
Not possible to accurately compare
Not standardised
Compare CIT to structured interviews= only difference is lack of cognitive elements
M+H
Quality of training
Police sergeant should deliver training- ‘one of them’
Base line measures before and after- measure improvements
Training should be intensive- 2 days, followed by practice
M+H
Conclusions
More research needs to be done to see why CIT is better than standardised interviewing
