People readily incorporate contextual information into their interpretations of witnesses’ verbal expressions of confidence and evaluations of accuracy. This is the bottom line of a recently published article in Law and Human Behavior. Below is a summary of the research and findings as well as a translation of this research into practice.
Featured Article | Law and Human Behavior | 2017, Vol. 41, No. 2, 180-190
Context Influences Interpretation of Eyewitness Confidence Statements
Daniella K. Cash, Louisiana State University
Sean M. Lane, Louisiana State University
When an eyewitness makes an identification from a lineup, he or she is asked to provide a confidence statement to help jurors assess credibility. However, these are verbal statements and people must rely on metacognitive processes to correctly interpret them. Recently, Dodson and Dobolyi (2015) argued that a person’s interpretation of a witness’s verbal confidence is influenced by the diagnosticity of the features used to justify his or her identification. We tested this hypothesis in 2 experiments. Experiment 1 found that, relative to a confidence-only control, participants reduced their ratings of confidence when statements were justified using a facial feature that was shared by lineup members, but not when the feature was unique to the member chosen from the lineup. In Experiment 2, we found that participants integrated across the dimensions of witness confidence and accuracy, even when asked to make separate ratings. Altogether, the results suggest that people readily incorporate contextual information into their interpretations of witnesses’ verbal expressions of confidence and evaluations of accuracy.
eyewitness confidence, lineup identification, juror interpretation, interpersonal metacognition
Summary of the Research
“The ability to evaluate one’s own memory processes and make decisions based on such evaluations—metacognition—plays an important role in allowing humans to adapt to a changing environment. When people assess the accuracy of their own memory, they are relying at least partially on the products of retrieval that come to mind. Metacognition also plays a role when people assess the veracity of other people’s memories (sometimes called social metacognition), although doing so involves making inferences based on other types of information, including the confidence that others express about their memory in words or demeanor. Indeed, research has shown that participants assign different strengths to a statement as a function of the vocabulary used by the information provider. For example, a person would be more likely to say that an individual who is “certain” an event occurred is more confident than a person who “supposes” the event occurred. The legal system presents a particularly consequential venue for these types of judgments. Judges and jurors are frequently called upon to evaluate the veracity of eyewitness statements as one piece of evidence of the defendant’s guilt in criminal cases. Although research has documented that mock jurors rely heavily on witness confidence in such evaluations, it is only recently that psycholegal researchers have examined how people interpret verbal expressions of confidence in an eyewitness identification. In the experiments that follow, we explore how this interpretation process is influenced by the context that accompanies witness’s verbal identification statements” (p. 180).
“Best practice guidelines for eyewitness identifications instruct lineup administrators to obtain statements of verbal confidence from the witness at the time of an identification. One obvious reason for doing so is that judges and jurors will be able to use such information when evaluating witness veracity at trial (e.g., by assessing the difference in confidence expressed at identification and trial). The reliance on confidence for making these decisions is both mandated through legal precedent and documented in the research literature. However, one potential issue that arises when using confidence statements is the assumption that an individual who is asked to interpret verbal expressions of confidence will assign the same meaning to it as the witness intended” (p. 180).
These potential issues have been explored and researchers have proposed an explanation coined perceived diagnosticity. Perceived diagnositicity indicates that individuals should assess features that are in verbal statements to examine how the respected feature differentiates the culprit from the other lineup members. For example, when an individual states that they recognize a culprit because of a specific feature, that persons nose should be compared to the other lineup members respected feature. If the identified feature is not distinctive, the reasoning for identifying that individual should be deemed less compelling, than if the feature had been distinctive. “This account predicts that in such situations participants should reduce their ratings of perceived confidence because they will find the reason for the witness’s confidence to be less compelling than if the feature were not verifiable or if there were no justification at all. Additionally, these nondiagnostic statements should produce more variability in participants’ numerical confidence ratings, relative to when they see confidence-only statements or statements accompanied by unverifiable details” (p. 181).
“In the current study, we further examined the perceived diagnosticity account with regard to participants’ interpretations of verbal witness confidence. One prediction of the perceived diagnosticity account is that the translation of verbal confidence into a confidence rating should be influenced by the degree to which a feature appears to be useful for discriminating the culprit from others in the lineup. If the feature described in the witness’s statement is salient and clearly differentiates the identified target from lineup members, participants’ interpretations of witness confidence should be similar to the confidence-only condition. However, if the described feature does not clearly differentiate the target from other lineup members, the perceived diagnosticity account predicts there should be a decrease in confidence ratings relative to a confidence-only statement” (p. 188).
“Our results are consistent with predictions of the perceived diagnosticity account proposed by Dodson and Dobolyi (2015). In this view, participants utilize information in the justification for its potential to accurately discriminate between lineup members. In essence, participants evaluate the adequacy of the justification when assigning numerical confidence to the witness’s statement” (p. 188).
Translating Research into Practice
“The National Academy of Science suggests that law enforcement personnel be required to obtain a confidence statement from eyewitnesses at the time of identification (National Research Council, 2014) so that it might be used by triers of fact at trial. If such information is provided to jurors, how can we expect it to be evaluated? We acknowledge that the real-world situation encountered by jurors is more complex than our procedure. During a trial, jurors listen to different types of evidence, discuss their impressions with their peers, and finally render a verdict at the end of the trial. Although the final decision is the result of a group process, this outcome is heavily influenced by predeliberation evaluations of individual jurors. Thus, although our procedure approximates individual juror evaluations of a single type of (important) evidence, we believe our results are relevant to jury decision-making. Another concern that could be raised is that the featural justification effect is observed in a situation that is unlikely to occur in actual eyewitness identifications (i.e., where the described facial feature does not distinguish between the suspect and other lineup members). For that reason, some might argue that its application to real-world scenarios would be fairly narrow. However, these external justifications may seem implausible to an observer, but might simply be a brief verbal reference to a richer, more detailed memory. In the laboratory, mock witnesses who viewed a crime video and identified a nondistinctive suspect from a lineup nevertheless justified their decision using facial features. Altogether, we believe that situations where seemingly implausible feature justifications are offered are unlikely to be rare, and thus may have an important impact on legal cases” (p. 189).
“Our results paint both a positive and a negative picture for juror evaluations of eyewitness evidence. First, across these studies, participants are able to consistently discriminate between verbal expressions of high and moderate confidence. This would suggest that jurors should be able to interpret verbal expressions of confidence systematically. To the extent that confidence at the time of identification provides a reasonable cue to accuracy, this would generally be an optimistic outcome for decision making. Second, the results of these studies suggest that jurors can take into account some types of information that should reasonably call into question the accuracy of the witness. For example, real life witnesses who justify their high confidence by claiming to have made their identification decision based on features that do not appear to discriminate between lineup members should perhaps be scrutinized by jurors. However, the implications are not all positive. The results of these studies also suggest that when witnesses express confidence that is not directly contradicted by verifiable information, jurors or judges likely simply assume he or she must be accurate. As described above, this assumption of relatively optimal witnessing conditions or memory states, in the absence of contradictory information, is clearly unwarranted. Furthermore, it suggests research might evaluate whether decisionmakers could appropriately integrate information about such features (e.g., how long he or she saw the perpetrator) into their evaluations of witness accuracy” (p. 189).
“The finding that participants conflate judgments of confidence and accuracy has implications not only for jurors’ evaluation of eyewitness testimony, but also other decision-making domains where evidence is provided and presented with varying levels of justification. Professionals such as intelligence analysts, weather forecasters, and climatologists often make these types of statements. For example, an intelligence analyst might rely on information about an upcoming attack provided by a source. That source might express strong confidence in the likelihood of an attack and justify that belief based on additional information about the group planning the attack. In cases where the justification seems untrustworthy, an analyst might reduce his or her assessment of the source’s expressed confidence and the likely accuracy of their prediction. However, the danger of conflating confidence and accuracy is that this produces a situation where the evaluator is losing (or simply fails to consider) information about the calibration of the source. Although calibration can be an important issue even when making judgments about single events, it becomes even more important to correctly identify (mis)calibration if the information provider is to be relied upon in the future. In the case of the intelligence analyst, calibration of a source might indeed be an important thing to consider in deciding whether to use a source again” (p. 189).
Other Interesting Tidbits for Researchers and Clinicians
“Our results and those of others suggest that people are sensitive to certain types of contextual information when assessing eyewitness identification statements. However, the task faced by jurors and judges when evaluating eyewitness accuracy from testimony is an even more daunting one. For instance, witness statements are given in the context of other case evidence. Furthermore, additional information relevant to the identification (e.g., viewing conditions at the time of the crime, postidentification feedback, and changes in confidence from identification to courtroom) must be integrated appropriately. Although more research is needed to appreciate the impact of this complexity, the general goal of this research is to better understand the decision-making processes used in this situation, and to provide a strong foundation for improving the ability of triers of fact to discriminate accurate witnesses from inaccurate ones” (p. 189-190).
Join the Discussion
As always, please join the discussion below if you have thoughts or comments to add!
Authored by Amanda Beltrani
Amanda Beltrani is a current graduate student in the Forensic Psychology Masters program at John Jay College of Criminal Justice in New York. Her professional interests include forensic assessments, specifically, criminal matter evaluations. Amanda plans to continue her studies in a doctoral program after completion of her Masters degree.