To wait or not to wait? Improving results when interviewing intoxicated witnesses to violence

Abstract

Witnesses to violent crimes are often alcohol intoxicated, but few studies have investigated the impact of alcohol on witness reports. This study investigated how alcohol intoxication and time of interview affected reports of intimate partner violence (IPV). One hundred thirty six healthy men (N = 66) and women (N = 70) were randomized to an alcohol group (0.8g/kg for men, 0.75g/kg for women) (N = 70) or control group (N = 66), given juice. Participants consumed drinks in a laboratory setting before they witnessed an IPV scenario. Fifty percent of the intoxicated and sober participants were interviewed ten minutes after viewing the film and all participants were interviewed one week later. For the analyses, participants in the alcohol group were divided into two groups (moderately/highly intoxicated) based on their BAC-level. Ten minutes after viewing the event, highly (BAC = 0.08-0.15) intoxicated witnesses gave shorter, but as accurate, reports as moderately intoxicated/sober witnesses. All witnesses gave shorter and less accurate reports one week later compared to immediately after. However, an immediate interview increased completeness one week later. In general, time and high intoxication made witnesses give less detailed accounts of actions and verbal information, but not of objects. Highly intoxicated witnesses reported less actions and verbal information in all interviews, while information regarding objects was reported to a similar extent. At the present BAC-level, it is beneficial to conduct an immediate free recall interview with intoxicated witnesses to obtain a maximum amount of correct information and minimize the negative effect of time.

Interviewing intoxicated witnesses: Memory performance in theory and practice

“…In Hildebrand Karlén et al . (), this reduction in completeness of the report was found among witnesses with BAC > 0.08 both in the immediate interview condition and in the one‐week delay interview condition. This reduction primarily consisted of less reporting of actions (i.e., presumably high salience details) between the involved parties in the scenario, while reporting of objects (i.e., low salience details) was not affected.…”

“…The same pattern has been found for sober participants in a study using a witness context (Pansky & Nemets, ). Studies of intoxicated witnesses’ immediate and delayed free recall have also found a decreased quantity but maintained accuracy of the reported information and used quantity‐accuracy trade‐off as a theoretical framework for these findings (Hildebrand Karlén et al ., ; see also Flowe et al ., for a discussion).…”

“…The appraisal disruption model has also been cited in Hildebrand Karlén et al . () to account for why intoxicated witnesses reported that they did not perceive witnessed violence as disturbing as sober witnesses did. This model could also explain why the “centralizing” or myopic effect of stress (i.e., weapon focus effect) is found to a lesser extent among intoxicated witnesses and among persons with impaired amygdala function, who comparatively focus more on peripheral information than sober persons/persons with maintained amygdala function (Adolphs et al ., ; Crossland et al ., ).…”

“…The seventh and eight studies (Hildebrand Karlén et al ., , ), showed that the dose 0.7g/kg generated higher BAC mean for women than for men, and that a BAC > 0.08 was required to reduce the completeness of the free recall report for intoxicated witnesses to intimate partner violence (including neutral/verbally aggressive/physically aggressive interaction between a man and woman) viewed on video. In Hildebrand Karlén et al .…”

“…However, rehearsal has been shown to increase the amount of information that intoxicated witnesses report about a crime one week later, possibly through enhancing the memory consolidation process of the witnessed information (e.g. Hildebrand Karlén, Roos af Hjelmsäter, Fahlke, Granhag & Söderpalm‐Gordh, ). Alcohol also seems to affect some cognitive functions more than others (Zoethout, Delgado, Ippel, Dahan & van Gerven, ).…”