1. 16TH International Conference of Investigative Psychology
15th-16th December 2015, London.
Investigating the effects of informational influence in an eyewitness setting
Dara Mojtahedi PhD Candidate | Maria Ioannou PhD | Laura Hammond PhD
International Research Centre for Investigative Psychology
Abstract
The majority of false convictions are said to be influenced by a false eyewitness statement (Cardozo, 2009). Previous research suggests that this could be due to individuals being prone to informational influence from other eyewitnesses, when faced with such situations (Gabbert, Memon, Allan, & Wright, 2004), however very little research has attempted to investigate the link between eyewitness statements and informational influence. The present study aimed to investigate the effects of social conformity on eyewitnesses after witnessing a crime.
Participants (N=90) watched a CCTV footage of a crime, in their groups. They were then asked to give statements on the event. Participants in the experimental groups were asked to discuss the footage amongst each other prior to giving their statements. The experiment was manipulated to identify if eyewitnesses would conform to one another. The findings were contrasted to that of the control group, where no group discussion was permitted.
Chi-square analysis identified that the inclusion of a group discussion had a significant effect on the eyewitness’s statements (p < 0.005).
The results indicated that social conformity can affect an eyewitness’s statement, through informational influence. The paper discusses the implications of the findings as well as direction for future research.
Problem
Method
Eyewitness testimonies influence approximately 75-84% of false convictions (Scheck et al.,2000), making them one of the least reliable forms of evidence used for conviction (Wells et al., 1998)
Yet there is still a strong tendency for police officers and jurors to believe that eyewitness statements are more reliable than research shows (Cutler & Penrod, 1995). One survey found that 87% police officers stated EW statements provide the major lead in most investigation (Kebbell & Milne, 1998).
Why are Eyewitness testimonies so unreliable? And should we reassess their validity as a safe source of evidence?
Participants (N=90) were randomly allocated to one of three conditions (A, B & C). In their groups, Participants would collectively watch a CCTV footage (Figure 1) of an offence (assault) occurring in a bar. Participants in condition A & B were then asked to discuss in their groups who they believed had started the fight. Afterwards participants would give their statements privately indicating who they believed had started the fight. Groups: Condition A: Each group in this condition consisted of 5 true participants + 1 confederate who was planted in the experiment to falsely argue that the wrong man had started the fight, during the group discussion. Condition B: Each group in this condition consisted of 5 true participant, no confederates were used. Group discussion was permitted Condition C: This was the control group. Each group consisted of 5 true participants, no confederates were used. No group discussion permitted.
Causes for false eyewitness statements
Inaccurate memory encoding (Cowan, 1988; Craik, Govoni, Naveh-Benjamin, & Anderson, 1996)
Memory decay (Ebbinghaus, 1985)
Poor interviewing techniques (Loftus, Miller, & Burns, 1978)
Pressure to perform (Wells et al., 2000)
If individuals can be influenced by investigators, then why not by other eyewitnesses?
Figure 2: Bar chart illustrating participants answers within each condition
The data was also analysed by looking at the total number of participants who identified the wrong man as the offender.
A Chi-squared test of independence found that there was a significant relationship between the presence of a confederate and false eyewitness identification x² (4, N = 90) = , p = .004.
Condition A: 73% identified the wrong man as the offender
Condition B: 40% identified the wrong man as the offender
Condition C: 32% identified the wrong man as the offender
Social Conformity
Research shows that in 38% of misidentification cases, there was more than one eyewitness giving the same false statement (Cardozo, 2009). Due to multiple eyewitnesses giving the same incorrect statement , it can be deduced that in these cases there must be another factor present during the event influencing the co-witnesses.
Social conformity is the act of changing your own views, attitude or behaviour to match that of a groups (Cialdini & Goldstien, 2004). Social conformity can be caused by informational influence, which is the process of obtaining information from a group and accepting it as accurate information about reality (Kaplan & Miller, 1987). Informational influence is at its strongest when an individual is uncertain about the task or topic at hand (Suls & Wheeler, 2000).
Therefore it can be hypothesised that if co-witnesses of a crime were to discus the event amongst each other, they would be at risk of changing their beliefs on the event and as a result give police investigators false information. Furthermore, previous literature suggests that the more uncertain an eyewitness is about the event, the more likely they are to conform the other eyewitnesses.
Research has found that social conformity can influence an individuals answers when recalling information (Gabbert, Memon, Allan, & Wright, 2004), however very little research has investigated this relationship through the context of witnessing a crime.
Discussion
Figure 1: Screenshot of CCTV footage of an assault, shown to all participants
The data indicates that eyewitnesses of a crime can be influenced by the views of other co-witnesses when recalling a crime. Therefor the false account of one eyewitness can have a negative impact on the identification accuracy of other eyewitnesses. The data suggests that eyewitnesses are at risk of taking on false information about an event as reality, therefor police investigators must uphold interventions to prevent eyewitnesses accounts from being contaminated by others (including the investigators). Finally, the findings also indicate that eyewitness testimonies are not always a reliable source of evidence and we must acknowledge this, especially within the court rooms. Directions for future research will aim to investigate individual differences in conformity rates amongst co-witnesses and attempt to identify the salient factors that can help predict an individuals risk of conforming to co-witnesses. As well as the individual, the group dynamics of the co-witness group will be observed to identify what social factors increase the level of informational influence exerted by a group.
Results
Group conformity was first analysed by looking at the similarities of statements within in each group.
Three levels were used as a measure for group conformity:
Unanimous conformity (when all five participants identified the same suspect as the offender)
Majority conformity (when four participants identified the same suspect as the offender)
No conformity (when less than four participants identified the same suspect as the offender)
Cross tabulation of the data found that when no group discussion (Group C) was permitted 100% of the groups showed no group conformity, with a significant standard residual of 1.9.
However in the conditions that permitted group discussion (Groups A & B), 76.9% of the groups showed group conformity with 30.8% if the groups displaying unanimous conformity.
The data suggests that discussing the event with other co-witnesses can impact an individuals final eyewitness statement
Aim & Objective
Aim To investigate the effects of informational influence on crime eyewitness statements Objective To identify if a group discussion amongst eyewitnesses prior to giving individual statements could influence their statements
Correspondence to: