Cross-Examination Fails to Safeguard Against Feedback Effects on Eyewitness Testimony

Abstract

The legal system relies heavily on eyewitness evidence to identify and prosecute criminal perpetrators, but wrongful convictions resulting from eyewitness misidentification have led many to conclude that eyewitness memory is unreliable. Advances in eyewitness identification research have produced a more nuanced understanding of eyewitness reliability, however. Whereas pristinely collected eyewitness identification evidence provides diagnostic information about a suspect’s guilt or innocence, numerous contaminants of eyewitness memory can undermine the reliability of eyewitness identification evidence. One such contaminant is confirming post-identification feedback—feedback given to or inferred by an eyewitness that communicates that their identification decision was correct. Confirming feedback is inevitable in real cases involving eyewitness identification and compromises the diagnostic value of eyewitness memory to such an extent that it undermines evaluators’ abilities to differentiate between accurate and mistaken eyewitnesses (Smalarz & Wells, 2014). The current research tested whether cross-examination, a fundamental legal safeguard for preventing wrongful conviction based on eyewitness misidentification, can help remedy the contaminating effects of feedback on eyewitness testimony. Evaluators (N = 128) viewed direct examination testimony or direct- and cross-examination testimony of accurate and mistaken eyewitnesses, some of whom had received confirming feedback following their identification. Although the majority of eyewitnesses admitted during cross-examination that some or all of their recollections may have been influenced by the feedback, viewing the cross-examination did not improve evaluators’ abilities to differentiate between accurate and mistaken eyewitness testimony. Cross-examination appears to be an insufficient safeguard for preventing wrongful convictions based on contaminated eyewitness evidence.