Social influences on the meta-cognitive regulation of eyewitness’ memory reports

Abstract

Eyewitness testimony is important to legal procedures. However, eyewitnesses are sometimes exposed to post-event factors that can distort their memory reports. The aim of this thesis was to explore the effects of social influences on the meta-cognitive monitoring and control processes that regulate memory reporting. In five experiments, we exposed participants to social comparative information (Experiments 1, 2 and 3) and misinformation from a co-witness (Experiments 4 and 5) and examined the effects of these manipulations on i) participants’ subjective confidence in the accuracy of their recall, ii) the precision of the details they volunteered, and iii) their tendency to withhold responses. In Experiment 1, participants (N = 87) were given negative, positive, or no information about a co-witness’s performance on a cued recall task comprised of questions about a mock crime. Participants then independently answered cued recall questions about the event. Participants exposed to information about a co-witness’ performance (negative or positive) reported more precise (fine-grain) details than those in the no information control group. Selection of fine-grain responses positively correlated with participants’ confidence in the accuracy of these responses. However, confidence in fine-grain responses did not differ significantly between participants in the control and experimental groups. In Experiment 2, participants (N = 90) watched a video of a mock crime event and then completed a practice task in which they answered questions about the event. Participants in the experimental groups received either positive or negative feedback about their accuracy on the practice task, which compared their performance to that of others. Control participants received no feedback. Receiving feedback did not significantly affect participants’ confidence, accuracy, or their grain size selection in comparison to the control group. In Experiment 3, participants (N = 92) watched a video of a mock crime event and completed cued recall questions about one of the characters in the video. Participants in the experimental groups then took turns verbally reporting their answers with a confederate who either confidently agreed (confirming condition) or disagreed (disconfirming condition) with the majority of their answers. Participants in the control condition did not report their answers verbally. Participants then completed another set of cued recall questions about the event and provided confidence ratings. Participants in the disconfirming condition included fewer fine-grain details in their memory reports (cf. those in the confirming and control conditions). In Experiment 4, participants (N = 66) watched one of two versions of a video depicting a mock crime event. Video versions differed with respect to two critical items. Participants in the discussion condition then discussed the event with a co-witness who had seen a different version of the video; participants in the control condition did not discuss the event. Participants then completed a cued recall task comprised of questions about the crime. Participants who discussed the event with a co-witness reported more incorrect details (mentioned by the co-witness) in the cued recall task than those in the control condition. Co-witness discussion did not significantly affect the meta-cognitive regulation of participants’ subsequent memory reports. In Experiment 5, participants (N = 60) watched a video of a mock crime event and then read one of two versions of a bogus co-witness report, each of which contained three different items of misinformation. Participants then answered cued recall questions about the event. Participants were significantly less accurate when answering questions about critical items (cf. non-critical items), but confidence, fine-grain volunteering, and response withholding were not significantly different for critical item questions (cf. non-critical item questions). Overall, the results of the present thesis demonstrate that social conditions can affect meta-cognitive regulation of the content of individual memory reports. The five experiments that comprise this thesis represent the first programme of research to examine social influence effects on the meta-memorial monitoring and control processes that govern memory reporting. Avenues for further research on this topic are discussed in light of the present findings.

Date of Award	2016
Original language	English
Awarding Institution	University of Portsmouth
Supervisor	Lorraine Hope (Supervisor) & James Ost (Supervisor)