forensic evidence?

bottom up approaches to analyzing forensic evidence?
Less biased as they build theories from the data (evidence) alone but may not be very helpful to solve crime when there is not much data
1 of 56
Top-down approaches to analysing forensic evidence?
More biased as they use existing theories about crime or theories developed from personal experience to fill in the gaps in the data (evidence), but it can help narrow down suspects when evidence is limited.
2 of 56
Types of bias in processing forensic evidence, Confirmation bias?
: The analyst only looks for information supporting existing theories about the crime, i.e. if they are told that a match has already been found, or that the suspect whose fingerprints they are analysing has no alibi
3 of 56
conformation bias? 2
they are likely to search for matching (identifying) features rather than using falsification (trying to disprove their prediction of a match- guilt- by searching for differences and only accepting an identification if no difference can be found).
4 of 56
Types of bias in processing forensic evidence, Selective attention?
The analysts, because of expectations created by contextual information stated in the example above, will not process contradictory information. They will unconsciously only pay attention to matching features (features supporting an identification)
5 of 56
selective attention? 2
if they expect one, leading to a bias towards finding an identification and not spotting differences.
6 of 56
Types of bias in processing forensic evidence,Over-confidence bias?
The analyst will not change their decision if later, contradictory evidence is provided. They will ignore the evidence and not objectively re-examine evidence.
7 of 56
Types of bias in processing forensic evidence, Conformity effect?
Analysts will not challenge other experts’ decisions when asked to re-evaluate evidence, because of concerns about censure from others in their profession. This may even be an unconscious motivation.
8 of 56
Types of bias in processing forensic evidence, Need-determination perception?
If the crime is high profile and emotionally arousing, like murder, the individual will feel more pressure to help solve it quickly to get justice for the victim and put away the dangerous perpetrator.
9 of 56
Need-determination perception? 2
rush their analysis, failing to notice details which contradict their identification (match).
10 of 56
dror 2005 sample?
27 uni students
11 of 56
dror method?
lab experiment
12 of 56
dror procedure?
96 pairs of fingerprints were selected for the participants to identify as either a match or no match. Need-determination perception was manipulated by giving participants crime reports of emotionally arousing crimes, containing crime scene photos
13 of 56
dror procedure 2?
Some high emotional arousal trials were also preceded by subliminal priming: the words ‘guilt’ or ‘same’ were flashed for too short a period for participants to consciously notice them, before the fingerprint pairs were shown.
14 of 56
dror procedure 3?
difficulty of matching task was also manipulated, so that half the fingerprints were complete and half were partial or unclear, as they would be if collected from a crime scene:
15 of 56
what was procedure 3 used to investigate?
to investigate whether top-down information (from the crime scene report/ subliminal priming) was more likely to be used when there were gaps in the data.
16 of 56
findings and conclusions dror?
Participants only showed bias in their decision making when the print was unclear, suggesting that top-down information is only used when analysts need to fill in gaps.
17 of 56
findings and conlusions 2?
For unclear fingerprints, participants made significantly more matches when the context was emotionally arousing than when it wasn’t (58% compared to 49%), and this was enhanced when subliminal priming was used (found a match for 66% of fingerprint p
18 of 56
what does finding two suggest?
emotionally arousing context information about the crime causes a need-determination bias to influence identification when the evidence is ambiguous, and selective attention may affect what participants process about the evidence when they primed
19 of 56
dror 2006? sample
5 volunteer fingerprint experts who consented to being studied in the following 12 months as part of their everyday work with a mean number of years’ experience of 17 years. Unfamiliar with the fingerprint of Mayfield
20 of 56
who was mayfeild?
had been wrongly identified by the FBI as the Madrid bomber
21 of 56
The study allowed the experts to be studied without knowing when they would be presented with the test material.
22 of 56
procedure 2?
Each of the 5 participants were approached by a colleague and asked to examine a pair of fingerprints. They were given a latent print (ambiguous) from the crime scene and a print exemplar (a print from a suspect).
23 of 56
procedure 3?
5 years previously they had analysed these fingerprints and found a match, which was verified by two independent analysts.
24 of 56
procedure 4?
Context was manipulated by telling the participants that the print exemplar was from the suspect wrongly identified by the FBI as the Madrid bomber, the participants were given the expectation that these prints would not be a match to see if CB occur
25 of 56
results and conslusions?
3/5 completely contradicted their original decision, declaring the prints a definite non-match. The failure of experts to give reliable decisions provides evidence that they had been affected by contextual information- confirmation bias
26 of 56
key research, hall and player 2008?
Bias and motivating factors in fingerprint analysis
27 of 56
To study whether a crime report, provided with the fingerprints to be analysed, containing highly emotionally arousing information, biases fingerprint analysts decisions when making an identification.
28 of 56
70 volunteer fingerprint analysts working for the Metropolitan Police: the majority active (working as fingerprint analysts rather than in management), with a range of experience from 3 months to 30 years.
29 of 56
Independent measures design field experiment
30 of 56
procedure 1?
During their working day, they were presented with one of 14 fingerprints which had been matched for difficulty (ambiguity) by independent experts, printed on a £50 note.
31 of 56
procedure 2?
They were given a fingerprint form containing the suspect’s fingerprints, and a crime report containing information about the crime
32 of 56
procedure 3?
This was manipulated for emotional arousal to see if need-determination perception influences participants’ decisions when making an identification.
33 of 56
procedure 4?
The high emotional arousal group received a report stating the crime was a murder, and the offender had shot at the cashier twice, and the low emotional arousal group received a crime report stating the criminal had tried to use a forged £50
34 of 56
procedure 5?
They analysed it in their work environment at New Scotland Yard. Whether they found an identification (match) was measured and a follow-up questionnaire was conducted asking them if they used the crime report when making decision, and if it influence
35 of 56
results and conclusions? 1
Some analysts didn’t read the crime scene report when analysing the data (57/70): individual differences in bias may affect applicability of findings that bias influences processing of forensic evidence.
36 of 56
results and concllusions 2?
These participants may be highly trained as they work for a specialist unit of the Met. Police and therefore be aware of potential biases and remain more objective.
37 of 56
results and conlusions 3?
The majority of those that did read the crime report, stated it biased their decision when it was emotionally arousing (52%) but not when it wasn’t (6%) suggesting need-determination perception influences the analysis of forensic evidence
38 of 56
results and conclusions 4?
This was contradicted by their behaviour: they showed no significant difference in identification of fingerprints when the emotional context was arousing (6 matches reported) or when it was not (7 matches reported)
39 of 56
results and conclusions 5?
Most reported that there was insufficient evidence from the print to make a comparison, or that whilst some points matched there was insufficient detail to make an identification. This shows need-determination perception did not bias them
40 of 56
results and conclusions 6?
However, participants may have chosen not to make a decision of an identification or non-identification because they were aware they were being tested and wanted to seem objective and professional (social desirability bias).
41 of 56
results and conclusions 7?
. For a real case, they would be under pressure to find a result to either eliminate a suspect or provide evidence so they can be charged.
42 of 56
results and conclusions 8?
This pressure would not occur in a simulated case and therefore need-determination perception bias would be reduced. These conclusions therefore may not apply to real life fingerprint analysis.
43 of 56
Use background research to suggest strategies for reducing bias.
44 of 56
what could you be asked in the application?
You could also be asked to design a procedure to investigate how bias could influence the analysis of forensic evidence, in which case you would select and justify methods from Dror (2005/6) and Hall & Player (2008).
45 of 56
dror 2005?
suggests that need-determination bias can be avoided when analysing forensic evidence by excluding type of crime and crime scene photos from crime reports provided with the fingerprints.
46 of 56
dror 2005?
His findings also suggest analysts must not be primed to find a match, as this can cause selective attention processes to bias them towards finding information from the evidence which confirms existing beliefs about the suspect’s guilt
47 of 56
dror 2005?
Therefore, the language of the crime report must be objective, not implying that there is a strong case for the guilt of the suspect or that a match has previously been found by any other analysts.
48 of 56
dror 2005?
. Analysts could be given training sessions by Psychologists where the findings of Dror (2005) are outlined for them to persuade them not to refer to the crime report before making their analysis.
49 of 56
stregnths dror 2005?
Advantageous in that defence lawyers cannot claim that analysis of the evidence was biased, and have the evidence disallowed. It is not costly to remove details from crime scene reports which makes it practical
50 of 56
weaknesses dror 2005?
Based on evidence low in population validity meaning this type of bias may be unlikely to affect real analysts in the same way and therefore that the strategy is targeted at the wrong source and will be ineffective
51 of 56
dror (2006)
suggests that confirmation bias can affect decision making when analysing forensic evidence.
52 of 56
dror 2006?
Therefore, if asked to re-evaluate fingerprints another analyst has already identified, or that other evidence has already suggested the suspects’ guilt or innocence, analysts should be kept blind to this.
53 of 56
dror 2006?
Fingerprints could be anonymised by giving them identifying numbers so analysts do not know whose prints they are working on and cannot be influenced by any contextual information from police, judiciary, other colleagues or the media.
54 of 56
stregnths dror 2006?
Based on study with high ecological validity meaning this type of bias is likely to affect real analysts in the same way and therefore that the strategy is targeted at the right source and will be effective.
55 of 56
weaknesses dror?
Setting up a database anonymising fingerprints may raise technical issues.
56 of 56

Other cards in this set

Card 2


Top-down approaches to analysing forensic evidence?


More biased as they use existing theories about crime or theories developed from personal experience to fill in the gaps in the data (evidence), but it can help narrow down suspects when evidence is limited.

Card 3


Types of bias in processing forensic evidence, Confirmation bias?


Preview of the front of card 3

Card 4


conformation bias? 2


Preview of the front of card 4

Card 5


Types of bias in processing forensic evidence, Selective attention?


Preview of the front of card 5
View more cards


No comments have yet been made

Similar Psychology resources:

See all Psychology resources »See all crime resources »