Crime topic 2 = The collection and processing of forensic evidence (biological) Flashcards
Key research
Hall and Player - if introduction of an emotional context affect finger print analysis and decision making
Aims of Hall and Player
To see whether a written report of crime could affect the fingerprint identification of a poor quality print by experts
And whether the level of emotion in written report affects identification of poor quality prints
Sample of Hall and Player
70 volunteers working for the Met Police in London
Sample method by Hall and PLayer
Self selected sample as 70 fingerprint experts from Met Police responded to a request and volunteered to the study
Apparatus of Hall and Player
14 copies of a £50 note with a superimposed obscured finger print
Crime scene examiners report
Context of crime
10 print form of possible suspects
How did Hall and Player create a scenario to test fingerprint experts?
Make up a fake crime with fake reports, evidence, suspects etc
Either given a high emotional or low emotional context of crime
Asked participants to deduce suspect from fingerprint
Why were 14 prints made in Hall and Players study
To check for consistency that same results occur no matter the print
How did we know the print on £50 note was obscured in Hall and Players study
Other fingerprint experts confirmed the results were ambiguous so could therefore be open to bias
Hall and Player procedure: what were participants given?
Give every participant a crime scene report, £50 note with obscure print, suspect print form and written up context of crime (manipulated IV to either be high or low emotional context)
Independent variable of Hall and Player
The crime was high emotional context or low emotional context as written in the crime scene report given to participants
High emotional context of crime in h+P
Allegation of murder that suspect shot victim
Low emotional context of crime in H+P
Allegation of forgery by paying for goods with forged 50 pound note
NO VICTIM
Experimental design of H+P
Independent measures, half the participants (35) were given high emotional context crime and other half low emotional context then compared
What type of research is H+P?
Experiment because there is an IV
What type of experiment is H+P?
After participants in H+P were given the crime scene report etc…..
Asked to complete identification of fingerprint to suspect as if it were part of normal working day
Use a magnifying glass for fingerprints and comparator
Then report what they found
Why was H+P high ecological validity?
Because they completed study as if it were normal working day so could leave whenever they wanted and had no time limit
So accurate to normal working environment = naturalistic behaviour
How did H+P have participants report their identification of suspects?
By self report in a questionnaire
H+P questionnaire at end of the study
Demographic questions on who the P is + how many years they worked
Whether they thought a match could be made or not, or if not enough detail was available and justifications + observations of the identification process
How many participants in Hall and player had read the crime scene report?
57/70
So 13 didnt know the emotional context
H+P results: how much did participants believe the emotional context affected results?
52% in high emotional context believed the report of the crime biased/ affected decision, significantly higher than 6% in low emotional context condition
In hall and player, was a significant difference found in actual identification of suspect between participants in high/low emotional context ?
No, their judgment was not affected even though Ps in high emotional context thought it had
Conclusions of H+P
The emotional context of a crime did not affect ability of fingerprint experts to find identification
Additional research
Dror - top down biases
Dror - research with 27 students
LSU (Linear sequential unmasking)
Miller - 6 pack idea
Dror research - what are the 2 processes used in fingerprint matching?
Bottom up
Top down
Bottom up approaches in fingerprint matches
Data driven analysis of ridges and patterns to identify unique match (more objective)
Top down approaches in fingerprint matches
When the expert’s previous experience and knowledge is used to make AN ASSUMPTION about the identity of the fingerprints if the ridges are unclear and objective analysis cant be done
Problem Dror found with top down processes
Less objective identification + more open to mistakes: Expert’s biases and expectation will falsely influence analysis
2 biases that decrease objectivity of top down process
Conformity effect
Observer bias
Observer bias problem in top down process of fingerprint analysis
The expert will anticipate if there is a match or not based on previous knowledge of crime learnt from police so may interpret the data based on this expectation
So not accurate identification
Conformity effect problem in top down process of fingerprint analysis
If fingerprint expert is asked to confirm the identification of a fingerprint done by peers/superiors, they may be more biased to agree with their finding
Dror’s research on students sample
27 students
Dror’s research on students IV
Whether the fingerprint pair was part of a low (burglary) or high emotional context (murder)
Whether subliminal messages flashed on screen
Procedure of Drors research on 27 students
Show 96 pairs of fingerprints, half clear match, half more ambiguous
Show images of context of each crime before
Participant presses button if they think it matches or not
How did Dror manipulate the IV?
Either the fingerprint pair was from a crime with a high emotional context or low emotional one
Reinforced by flashing images on screen before each pair of what the crime scene looked like
Subliminal messages appearing saying guilty
DV of Dror
Number of matches participants made in high emotional contexts or low emotional contexts
Results of Dror
Participants were shown high emotional cases they were more likely to find a match between fingerprints (58%) compared to the low emotional cases (49%)
When the subliminal messages were added to the emotional cases, this rose to (66%)
Conclusions of Dror
Emotional context of a crime given in info before identification will affect analysis
In students, not necessarily true for experts
Circular approach in identification of finger prints
Analyst is given suspects print and evidence print from crime scene and looks between them to find identification in ridges etc
Changes can be made on whether there is a match or not etc
What is wrong with the circular approach in identification?
Bias because expert may only look for specific ridges that match and can be influenced = CONFIRMATION BIAS of looking for specific details to confirm experts expectation
How to overcome bias in circular approach in identification?
Use Linear sequential Unmasking (LSU)
Linear sequencing unmasking step 1
Look at the print from the crime scene on its own and make notes on unique details and features
Decide if enough detail is present for possible identification
Linear sequential unmasking step 2
Give small amounts of information about the case little by little in order of importance to identify if the info is consistent
#CANNOT GO BACK AND CHANGE NOTES SO THIS NEW INFO DOES NOT BIAS NOTES MADE BEFORE
What piece of evidence did Miller consider in his theory?
Hair strands not fingerprints
What type of bias did Miller say was present in police analysis?
Need determination bias
What is need determination bias?
When experts felt they had to find a match on crime scene hair to suspect hair so they claimed to find a match when not true to reality
Miller’s six pack idea
Expert given 6 strands of hair to compare against crime scene hair
1 is from suspect and other 5 is fillers so expert doesnt know which is suspect’s hair
Will be more careful than just blindly finding a match
applications for reducing bias in processing and collection of forensic evidence : 6 Pack idea
Use 5 filler samples and 1 from suspect when comparing to crime scene one - not know which sample is from suspect or filler - eliminate need confirmation bias - need to be more careful and objective in analysis - Miller’s six pack idea
applications for reducing bias in processing and collection of forensic evidence : Awareness of top down biases
Educate themselves on conformity effect + Observer bias - apply more bottom up approach of data driven analysis for unique details - Dror found inaccurate identifications were made as a result of top down processes where experts are influenced by biases and expectations fro what theyll find
Applications for reducing bias in processing and collection of forensic evidence: limit amount of information learnt of context
Only allow experts to view information relevant to comparison between prints - In lead up to analysis, avoid presenting info on nature of crime and victim etc - Dror found matches were more likely if participants saw high emotional context compared to low one
