Confidence malleability in the interviewing setting, and its effect on subsequent memory monitoring and regulation

Eyewitness evidence is often essential for the outcome of an investigation. However, research has shown that memory is not perfect, and eyewitnesses can make mistakes. In the past forty years, researchers have developed evidence-based techniques to interview eyewitnesses effectively and to maximise the amount and accuracy of the information elicited in an interview. Despite the wealth of research on best practice interviewing techniques, little is known about how these might affect eyewitness confidence. This is important because confidence plays a pivotal role in the regulation of memory output. For example, research in metamemory showed that confidence judgements underpin eyewitnesses’ decisions to report or withhold information. This PhD aims to fill this gap by investigating confidence changes within the context of an investigative interview and testing the hypothesis that confidence shifts following an interview might affect subsequent memory regulation. Study 1 showed that memory confidence can change after an interview. Study 2 built on this finding and showed that when the interview promotes free and undirected retrieval, confidence remains stable. On the contrary, when the interview promotes a directed retrieval via presenting Cued Recall questions, confidence decreases. A further investigation of the metacognitive processes that underpin these results (Study 3a and 3b) showed that different types of question lead to confidence shifts depending on the difficulty experienced when answering them. Finally, drawing upon these results, Study 4 investigated the conditions of an interview likely to lead to confidence shifts and those likely to promote confidence stability. Across the studies, no evidence was found that changes in confidence following an interview impact subsequent memory regulation. Overall, theresults confirm and further the existing evidence in support of good practice in eyewitnesses interviewing. As such, evidence-based techniques are compatible with confidence stability, while deviations from them are likely to lead to decreased confidence.