![Dominic Allen](https://www.lancaster.ac.uk/2021ugconference/wp-content/themes/lancaster2021/images/profile.png)
Dominic Allen
Used to collect veracious eyewitness testimony, investigative interviews (e.g. the Cognitive Interview; CI) are often time- and labour-intensive. Self-report interviews (e.g. the Self-Administered Interview©; SAI©) have helped ease these constraints. This research examines the efficacy of online-based adaptions of these interviews (A-CI and A-SAI) and a novel combination of both. The effect of interviewer and perpetrator ethnicity on eyewitness recall are also examined. In an online study, 214 participants viewed a mock crime video depicting either a South Asian or Caucasian perpetrator. Participants then completed one of the three interviews, conducted by an avatar, representing a South Asian or Caucasian police officer. The Combined Interview elicited the most correct information, followed by the A-SAI©, then the A-CI. The Combined Interview and A-SAI© generated more inaccurate recall than the A-CI. However, no differences in accuracy rate were found between the three interviews. Thus, the Combined Interview represented the most efficacious self-report investigative interview, eliciting a greater quantity of eyewitness information without sacrificing overall accuracy. Neither perpetrator nor interviewer ethnicity affected recall, suggesting these factors do not influence online self-report investigative interviews. These findings have implications for policing in England and Wales, as the Combined Interview surpassed currently utilised investigative interviewing techniques.