Can authority influence misinformation?: the relationship between misinformation and eyewitness accuracy

Mistaken identification by eyewitnesses was involved in 72% of the cases where people were later exonerated by DNA testing (Smalarz & Wells, 2015). There has been a large movement in support of eyewitness being unreliable. Misinformation and the pitfalls of memory have been widely studied; however, an interesting avenue of thought is how might the presence of authority interact with these phenomena. While previous research studied how authority impacted memory, few studies have examined the role that individual differences in response to authority impacted memory. This study examined how authority and response to authority impacted eyewitness memory. One hundred twelve participants were randomly assigned to four conditions (high authority-leading questions, low authority-leading questions, high authority-nonleading questions, and low authority-nonleading questions). Participants watched a short video in which a bank robbery was committed. Participants were interviewed and asked leading or nonleading questions by a person with high or low authority. After, participants were asked to make a statement about any details they could remember. For interview accuracy, there was a main effect of question type. Participants demonstrated better recall for nonleading than leading questions. The finding replicates previous research that shows leading questions can create more misinformation that neutral questions. However, no other effects were found for the interview accuracy or the witness statement accuracy. Thus, it appears that authority and response to authority did not impact eyewitness memory. This could be due to a weak authority manipulation and that participants may need additional cues, such as uniform, to legitimize a persona as an authority figure. Thus, it could be that when examining characteristics of the interview (e.g., response to authority), there may be other systems impacting recall other than authority. Limitations and future directions are discussed