Making a Case Flashcards
Bruce et al - Interviewing Witnesses
Importance of internal / external features in facial recognition.
> AIM To investigate the relative recognisability of internal and external features of a facial composite.
> SAMPLE - Exp 1: 30 staff + students = 15m, 15f.
- Exp 2: 48 undergrads - 21m, 27f.
> PROCEDURE
- Exp 1: Independant measures design.
- 10 celebs and 40 composite images produced by E-FIT, PRO-FIT, Sketch and EVO-FIT.
- Each face clean shaven + glasses avoided.
- 3 sets of composites used, a ‘complete’ set, an internal features set and external features set.
- Asked to place each composite in front of each celeb (in their own time)
- Exp 2:
- Photo array - distractor faces or foils making the task more difficult.
- Faces / foils were made easy (all different) or hard (all similar) to identify.
- P’s picked out which composite matched celebs face.
> RESULTS
- E1: Whole composite and external features sorted similarly (35% correct)
- Internal features only 19.5% correct.
- E2: External - 42%
- Internal: 24%
> CONCLUSION
In E1 and E2 p’s performed just above chance with internal features.
- P’s performed equally well with external features of whole face.
- Something about the internal features of a face that doesn’t work well when trying to create a reconstruction.
> EVALUATION
Low eco. valid - in real life faces move and display emotion.
- Small sample - all from one place - potentially similar type of person (uni staff / students) - less generalisable if they study same subject etc.
Loftus + Messo - Interviewing Witnesses
Weapon focus.
> AIM To provide support for the ‘weapon’ focus effect when witnessing a crime.
> SAMPLE - 36 students at Washington Uni - 18 - 31 yrs.
> PROCEDURE - P’s told study aim - about ‘proactive interference’
- 2 sets of 35 mm slides shown.
- 18 slides of queue at restaurant ‘Taco Time’.
- Control group: Person B hands cashier a cheque.
- Experimental group: Person B pulls out a gun.
- Each slide shown 1.5 seconds.
- DV = 20 item multiple choice questionnaire.
- P’s shown 12 head and shoulder photos and asked how confident they were that they had identified Person B correctly (scale 1-6 - 6 = v sure)
> RESULTS
- Answers to questionnaires showed no significant difference between 2 control groups.
- Chance performance was 8.5%
- Control condition, 38.9% identified person B correctly.
- Exp group - 11.1% correct
- No difference in confidence levels.
- Eye fixation showed 3.72 on gun and 2.44 on cheque.
> CONCLUSION
- P’s spent longer looking at weapon - harder to pick suspect.
- A second experiment done (80 students) - same results
- Influence may be further enhanced in a real world situation when a witness will be more aroused.
> EVALUATION
Biased sample - half paid / half gain credit at Uni.
- Demand Characteristics
- Low eco. valid - lab experiment.
Fisher et al - Interviewing Witnesses
Cognitive Interview Technique
> AIM To compare the performance of experienced detectives pre and post training in CIT and to compare their performance post training with a control group.
> SAMPLE 16 detectives in Florida - 7 trained in CIT PROCEDURE - 88 interviews recorded 0 each detective 5-7 interviews over 4 months (Mainly victims of commercial robbery or purse snatching.)
- 7 detectives - four 1 hour sessions of training.
- 47 interviews recorded over 7 months - lots of original sample unable to continue.
- Interviews transcribed and scored by independent judges.
- Number of relevant, factual and objective statements recorded.
> RESULTS Comparison pre / post training:
- 47% more info recorded in post training.
- 6/7 detectives in this group did better post training.
- Comparison of trained group and control group :
- 63% more infor recorded by trained detectives.
> CONCLUSION
- Effectiveness of CI’s:
- More info obtained from witnesses to real events with no loss of accuracy and a minimal increase in the time taken to interview them.
- CI has been applied to clinical settings, using therapists to develop medical histories.
- CI used by police forces around UK.
- CI developed for use with children, social work and therapy.
> EVALUATION
- Small sample - 16
High eco valid - field exp.
Mann et al - Interviewing Suspects
Detecting lies.
> AIM To test police officers’ ability to distinguish truth and lies during police interviews and suspects.
> SAMPLE - 99 Kent police officers - 24f, 75m. (34 mean age)
> PROCEDURE P’s judge the truthfulness of people in real-life police interviews.
- Saw video clips: 14 suspects - head and torso - expression + movement.
- Clips backed up by evidence which established lies or truth at any point.
- 54 video clips in total - varies between 6 - 145 seconds.
- Filled in questionnaire about detecting lies in past.
- After video clip: indicated whether it was a lie or truth, how confident they were, asked to list cues they used to detect the liars.
> RESULTS - Mean lie accuracy - 66.2%
- Mean truth accuracy - 63.6%
- Significantly greater than chance (50%)
- Mentioned cues: gaze, movements, vagueness, contradictions in their stories, fidgeting.
> CONCLUSION
Highest level of accuract for an ordinary group of police officers.
- The more experience the officer has, the better they are at detecting lies.
- Good detectors rely on more than story cues than physical cues.
- Police officers performed at above chance level.
> EVALUATION
- No control group.
- Ethnocentric bias - kent - lacking diversity - crime difference in inner city.
Gudjohnsson - Interviewing Suspects
False confessions.
> AIM To document a case of the false confession of a youth who was distressed and susceptible to interrogative pressure.
> SAMPLE - Case study - 1 person.
> PROCEDURE
Interview: Self-report
- Repeatedly accused of lying
- Questioning; leading, accusatory
- Police suggesting he was sexually impotent.
- After 14 hours he falsely confessed.
- Retracted next day only to confess again under pressure
- Failure to have successful relationships.
Psychiatric tests in prison:
- No evidence to suggest mentally ill.
- IQ: 94 (just below average)
- 10 on Gudjohnsson Suggestibility Scale.
1 year later released as another person pleaded guilty.
> CONCLUSION Clear case of ‘coerced compliant false confession’
- Gave into pressure to escape from interview situation.
> EVALUATION Small sample
- Very useful, shows that under correct conditions, FC is very possible. Needs to be prevented because prison and the situation can cause grave psychological harm.
> BACKGROUND
- Accused of murdering 2 women - battered to death in their home - savings missing - sexually assaulted.
- Arrested due to inconsistencies in his account of his movements and spending more money than usual.
- NO FORENSIC EVIDENCE AGAINST HIM.
Inabu et al - Interviewing Suspects
The Reid ‘9 Steps to Interrogation’.
> 9 STEPS
- DIRECT CONFRONTATION
- SHIFTING THE BLAME
- INTERRUPTING DENIAL
- IGNORING THE SUSPECT
- REINFORCING SINCERITY
- ALLOWING SUSPECT TO BECOME QUIET AND LISTEN
- POSING THE ALTERNATIVE QUESTION
- RESULT = GETTING SUSPECT TO ADMIT GUILT IN FRONT OF WITNESS.
- DOCUMENT THEIR ADMISSION.
> EVALUATION
- Western (more American) model.
- May not be appropriate in parts of the world.
- Even in England (similar society) this is not used.
Canter - Creating A Profile
TOP-DOWN TYPOLOGY
Organised / disorganised theory.
> AIM To test the reliability of organised / disorganised theory.
> SAMPLE 100 cases
> PROCEDURE Content-analysis
- Using psychometric method of multidimensional scaling has applied to find out if the organised or disorganised hypothesised features were consistent and distinctive.
- Cases were from published accounts of serial killings - cross checked with court reports and officers.
- 3rd crime committed by each serial killer in each series was analysed.
- The Crime Classification Manual was used to classify crimes as organised and disorganised.
> RESULTS 2x as many disorganised crime-scene action.
- Only sex / rape (vaginal) acts occur in more than 2/3 of disorganised crimes.
- Only 2 crime scene behaviours co-occurred in the organised typology in a level above chance:
- The body was concealed in 70% of cases
- Sexual activity occurred in 75% of cases.
> CONCLUSION
Disorganised offenders are more common or easier to identify.
- All crimes have an organised element to them.
- Look at individual differences (personality) between offenders.
> EVALUATION
- Published accounts of serial killings - cross checked w/ court reports and offices = RELIABLE
- 100 cases - large sample - more generalisable findings.
Canter - Creating A Profile
BOTTOM-UP APPROACHES
Developments in offender profiling.
> AIM To identify a behaviour pattern from similarities between offenders.
> SAMPLE Content-analysis - 27 offenders.
> PROCEDURE Content-analysis - 66 offences from various police forces.
- Committed by 27 offenders.
- Conducted to find 33 offence variables that were clearly linked to a potential behavioural characteristic.
- Possible to say ‘yes’ or ‘no’ to each variable.
> RESULTS
Following variables found to be central to the 66 cases of sexual assult:
- Vaginal intercourse
- No reaction to the victim
- Impersonal language
- Surprise attack
- Pattern of behaviour where the attack is impersonal and sudden.
- Important research: Attempted intimacy, impersonal interaction and criminal behaviour intentional.
> CONCLUSION
- All 5 aspects have now been shown to contribute to all sexual offences, but in different patterns for different individuals.
- Further understanding of how an offender’s behaviour changes over a series of offences.
- Useful in establishing whether two or more offences were committed by the same person.
> EVALUATION
Content-analysis - secondhand data
- Allows a range of study cases that have already occurred to be reviewed w/o having to carry out an experiment or manipulate any variables.
Canter - Creating A Profile
JOHN DUFFY
John Duffy Case Study.
> PROFILE John Duffy
- 3 murders, 24 sexual assults.
> CANTER people influence each others actions, peoples behaviour changes over time.
> SOCIAL PSYCHOLOGY
Type of relationship with others:
-JD tried to form a relationship with victims, asked them about their personal lives and showed consideration.
Degree of domination he exercised over victims:
- Only used enough control to rape - therefore not a very powerful / secure individual.
Familiar with police procedures:
- On one occasion he combed the pubic hair off his victim to remove his own hairs - Implied he had been questioned by police.
> MAIN POINT OF CANTERS PROFILE
- Lived near crimes
- Lives with women
- Aged mid-late 20’s
- Right-handed
- Semi-skilled / skilled job - relatively isolating
- Knowledge of railways
- Previous criminal record
> CHARACTERISTICS OF OFFENDER
- Lived in suggested area
- Recently separated from wife
- Aged late 20’s
- Right handed - yes
- Job : Travelling Carpenter
- Worked for British Rail
- Raped his wife at knife point
> EVALUATION
- Eye witness testimony - not accurate - rape victims unable to accurately describe attacker but accurate at dexcribing attack (“Weapon focus”)
- Supports ‘Bottom-Up Approach’
- Geographical profile of the attackers movements and predict their home address
- Canter’s CRIMINAL CONSISTENCY hypothesis suppoerted case.
