Can Laboratory Findings on Eyewitness Testimony Be Generalized to the Real World? An Archival Analysis of the Influence of Violence, Weapon Presence, and Age on Eyewitness Accuracy

Abstract

The authors conducted 2 studies to assess the effects of levels of violence, the presence of a weapon, and the age of the witness on the accuracy of eyewitness testimony in real-life crime situations. Descriptions of offenders were taken from eyewitnesses' statements obtained by the police and were compared with the actual details of the same offenders obtained on arrest. The results showed that eyewitnesses tended to recall the offenders' hairstyle and hair color most accurately. None of the effects for the level of violence, the presence of a weapon, or age approached statistical significance, with the exception that, in the 1st study, accuracy in describing hair color was better when associated with high levels of violence and in cases of rape. It is argued that caution must be exercised in generalizing from laboratory studies of eyewitness testimony to actual crime situations.

Sex trade workers' narratives of sexual violence: A field investigation

“…Between-participants comparisons for detailed and for less-detailed narratives of sexual assaults reveal no differences in the overall amount of details between weapon-present and weapon-absent narratives. This finding is in line with other field studies that have not supported the weapon focus effect (e.g., Hulse & Memon, 2006;Kuehn, 1974;Tollestrup et al, 1994;Wagstaff et al, 2003). Of course, such findings do not prove that the weapon focus effect does not exist.…”

“…However, in line with various other field studies (e.g., Cutshall & Yuille, 1989;Peace, Porter, & ten Brinke, 2007;Porter & Birt, 2001;Wagstaff et al, 2003;Woolnough & MacLeod, 2001), which suggest that eyewitnesses can actually recall their traumas in great detail, this assumption is not supported by the present study. Memory impairments may not be typical of trauma after all.…”

“…For instance, the completeness of witnesses' descriptions of crime perpetrators did not depend on the presence vs absence of a weapon (Kuehn, 1974). Similarly, Cooper, Kennedy, Hervé, and Yuille (2002), like Wagstaff et al (2003), did not find any differences in recall between events that did vs did not involve a weapon. Moreover, Hulse and Memon's (2006) study of simulated crimes revealed no differences in identification accuracy between shootings vs other crimes not involving any weapons.…”

“…In some studies the level of stress has been operationalised by the role of the witness: a bystander vs a victim, the former being less involved and presumably less ''stressed'' than the latter (e.g., Christianson & Hubinette, 1993;Yuille & Cutshall, 1986). Another operationalisation has been the type of event: violent vs less (or non-) violent (e.g., Tollestrup, Turtle, & Yuille, 1994;Wagstaff et al, 2003;Woolnough & MacLeod, 2001). The present study involved only one type of witness (victims) and one type of crime (sexual violence).…”

“…Due to the fact that trauma cannot be simulated in experimental studies, the laboratory research paradigm is of limited use to the investigation of memory for highly stressful (traumatic) events. Accordingly several researchers have cautioned against the application of laboratory findings in a forensic context, for example by providing expert testimony on eyewitness memory or by conducting credibility assessments (e.g., Arntzen, 2007;Wagstaff et al, 2003;Yuille & Wells, 1991). Indeed, field research provides fewer unanimous findings than laboratory research: Some studies emphasise their high level of detail, their consistency, vividness, and accuracy (e.g., Cutshall & Yuille, 1989;Peace & Porter, 2004;Pezdek, 2003;Porter & Birt, 2001;Wagenaar & Groeneweg, 1990).…”