Eliciting information from cooperative sources about single & repeated multi-actor events

Abstract

Successful investigations in forensic and security contexts depend on eliciting reliable and detailed information from sources. Although research has developed effective interviewing protocols to improve recall of witnessed events in criminal investigations, there is only a small body of research on information elicitation tools for intelligence gathering. The overarching aim of this thesis is to contribute to the development of interviewing techniques for use with cooperative sources in security settings. Specifically, we examined the effectiveness of mnemonics, reporting formats and prompts to facilitate recall for multi-actor events witnessed on both single and repeated occasions. In Experiment 1, we introduced a new mnemonic to the timeline technique. Participants witnessed a multi-actor crime-event under full or divided attention and provided an account using self-generated cues, other-generated cues or no additional cues across timeline reporting conditions. The results showed that use of the self-generated cues increased the reporting of correct details (cf. other-generated and no cues) under full but not under divided attention conditions. In Experiments 2 and 3, we examined the efficacy of open-ended questions to follow-up on an initial report. In Experiment 2, participants witnessed a multi-actor crime-event and used the timeline or a free recall format to provide an initial report. In Experiment 3, participants used the timeline to provide their recall of a video depicting a group planning and carrying out an attack. Before being asked follow-up questions, half of the participants were instructed to avoid guessing, to feel free to withhold an answer and to consider the precision of their answers (i.e. provide general or specific details). The results showed that follow-up questions elicited new information across conditions. However, the accuracy of the responses to the follow-up questions was not as high as the initially reported information (Expt. 2 & 3), even after participants were instructed to monitor the accuracy of their responses (Expt. 3). In Experiment 4, we tested the effectiveness of self-generated cues, the timeline technique and follow-up open-ended questions, as part of a Multi-Method Interviewing Format (MMIF) to facilitate the retrieval and particularization of repeated events. Over the course of a week, participants witnessed four videos of a group planning and carrying out an attack, where either all four videos were highly similar, or where three videos were highly similar, and one video included a new and a changed critical detail to introduce a deviation to the script. After a week, participants provided an account using the MMIF, the timeline technique or a free recall format. The results showed that use of the MMIF elicited more correct information and increased particularization of specific instances of the repeated events (cf., timeline and free recall format). There was no additional benefit for recall when deviations were present in a specific instance. This set of experiments successfully extended the timeline technique into a format that can be used flexibly for the reporting of complex and repeated events. In the discussion of our results, we suggest avenues for future research focused on retrieval techniques for use in applied information elicitation contexts.

Date of Award	Sept 2018
Original language	English
Awarding Institution	University of Portsmouth
Supervisor	Lorraine Hope (Supervisor), Paul Taylor (Supervisor) & Aldert Vrij (Supervisor)