Retrieval effort and accuracy in eyewitness testimony
Sammanfattning: For better or worse, eyewitness testimonies make up common evidence in criminal trials. This has the benefit that it allows for guilty offenders to be convicted even in the absence of physical evidence. However, the fallibility of memory also means that eyewitnesses may be mistaken in their recall, risking wrongful, innocent convictions. This dilemma has sparked a multitude of studies on situations and factors that distort memory and make it unreliable, and consequently, situations where a testimony should be reliable. It has also sparked research into methods that attempt to distinguish between correct and incorrect memories. It is this latter research that forms the foundation for this thesis.In this thesis, I have attempted to improve judgments of eyewitness accuracy by examining the relationship between expressions of effort in memory retrieval, and accuracy. The thesis builds upon earlier studies demonstrating a negative relationship between retrieval effort and accuracy (easily retrieved memories are more often correct), and explores this relationship further in eyewitness testimonies, while also addressing a number of limitations. Three empirical studies were undertaken, wherein participants saw a fictitious crime and were interviewed as eyewitnesses (Study I & II), or read testimonies and judged the accuracy of statements (Study III). Study I showed that four (para)verbal markers of retrieval effort (“retrieval-effort cues”) were more common in incorrect recall compared to correct recall in eyewitness testimonies. Study I also showed that the retrieval-effort cues mediated between confidence in a memory, and accuracy. Study II replicated Study I and moreover, showed a robustness of the findings, as retrieval-effort cues predicted accuracy both directly after viewing an event, as well as two weeks later, even when witnesses had actively engaged in memory repetition. Finally, Study III showed that fact finders could be instructed to use retrieval-effort cues to predict memory accuracy, and furthermore, results suggested that this method could be superior to fact finders’ spontaneous accuracy judgments. Taken together, this thesis provides evidence that brings us closer to a working method to judge eyewitness accuracy, and illuminates important next steps to be solved, moving forward.
KLICKA HÄR FÖR ATT SE AVHANDLINGEN I FULLTEXT. (PDF-format)