Evidence for a confidence–accuracy relationship in memory for same- and cross-race faces

Does presenting perpetrator and innocent suspect faces from different facial angles influence the susceptibility of eyewitness memory? An investigation into the misinformation effect and eyewitness misidentification

Introduction

This study investigated the effects of face angle congruency across stages of a misinformation paradigm on lineup discrimination accuracy.

Methods

In a between-subjects design, participants viewed a mock crime with the perpetrator's face from the front or profile angle. They then read a news report featuring an innocent suspect's image from the same or different angle as the perpetrator had been shown. A subsequent lineup manipulated perpetrator presence and viewing angle of the lineup members, who were all shown either from the front or in profile.

Results

No significant difference emerged in identification errors based on angle congruency between stages. However, accuracy was higher when faces were shown from the front angle, both during the initial event and the lineup, compared to the profile angle.

Discussion

The results of this research underscore the importance of considering viewing angles in the construction of lineups.

Unskilled, underperforming, or unaware? Testing three accounts of individual differences in metacognitive monitoring

Many studies show that competence (e.g., skill, expertise, natural ability) influences individuals' capabilities of monitoring their item-level performance. However, debate persists about how best to explain these individual differences in metacognition. The competence-based account ascribes differences in monitoring to individuals' objective ability level, arguing that the same skills necessary to perform a task are required to effectively monitor performance. The performance-based account attributes differences in monitoring to changes in overall task performance - no individual differences in competence required. Finally, the metacognitive awareness account proposes that alignment between an individuals' self-assessed and objective ability leads to differences in monitoring. In this study, 603 participants completed a self-assessment of face recognition ability, a lineup identification task, and an objective assessment of face recognition ability. We manipulated the number of encoding repetitions and delay between encoding and test to produce varying levels of task performance across objective face recognition ability. Following each lineup decision, participants provided both a numeric confidence rating and a written expression of verbal confidence. We transformed verbal confidence into a quantitative value using machine learning techniques. When matched on overall identification accuracy, objectively stronger face recognizers used numeric and verbal confidence that a) better discriminates between correct and filler lineup identifications than weaker recognizers, and b) shows better calibration to accuracy. Participants with greater self-assessed ability used higher levels of confidence, irrespective of trial accuracy. These results support the competence-based account.

An own-race bias in the categorisation and recall of associative information

ABSTRACTPeople tend to better remember same-race faces relative to other-race faces (an "own-race" bias). We examined whether the own-race bias extends to associative memory, particularly in the identification and recall of information paired with faces. In Experiment 1, we presented white participants with own- and other-race faces which either appeared alone or accompanied by a label indicating whether the face was a "criminal" or a "victim". Results revealed an own-race facial recognition advantage regardless of the presence of associative information. In Experiment 2, we again paired same- and other-race faces with either "criminal" or "victim" labels, but rather than a recognition test, participants were asked to identify whether each face had been presented as a criminal or a victim. White criminals were better categorised than Black criminals, but race did not influence the categorisation of victims. In Experiment 3, white participants were presented with same- and other-race faces and asked to remember where the person was from, their occupation, and a crime they committed. Results revealed a recall advantage for the associative information paired with same-race faces. Collectively, these findings suggest that the own-race bias extends to the categorisation and recall of information in associative memory.

Laypeople's beliefs about the effects of common estimator variables on memory

This research builds on James Ost's research investigating whether laypeople's beliefs align with those of experts. Recent studies that examined the relationship between high-confidence eyewitness identifications and accuracy proposed that the mechanism underlying this relationship may be based on a knowledge-conditional model. According to this model, the accuracy of a confidence judgment depends on knowledge about factors that affect memory accuracy. However, there has not been a comprehensive assessment of laypeople's knowledge about the effect on memory accuracy of many estimator variables known to influence the accuracy of eyewitnesses, specifically those relevant to research on the relationship between witness confidence and accuracy. This study consists of the development of a 30-item scale to assess laypeople's knowledge of the effect of 10 common estimator variables on memory accuracy from three points of view (POV): Self, Other, and Juror. Across MTurk and undergraduate samples, laypeople's beliefs about the effect of these estimator variables were generally consistent with research findings and did not differ as a function of POV. Additionally, for most estimator variables, participants' beliefs about memory were consistent with results in the confidence-accuracy literature; confidence and identification accuracy appear to be poorly calibrated for estimator variables that people know less about.

Postdictive confidence (but not predictive confidence) predicts eyewitness memory accuracy

If testing conditions are uncontaminated, confidence at test reliably predicts eyewitness memory accuracy. Unfortunately, information about eyewitness postdictive confidence (at the time of the identification test) is frequently unavailable or not well documented. In cases where postdictive confidence is unavailable, a useful indicator of eyewitness accuracy might be an eyewitness's predictive confidence made shortly after the event. How do the accuracy of predictive and postdictive confidence judgments compare; and do variables reported to affect memory (e.g. exposure duration, face race) affect the reliability of the confidence-accuracy relationship for predictive and postdictive judgments? In two experiments, we tested the accuracy of memory predictions (immediate and delayed judgments of learning [JOLs]) and postdictions (confidence) for same- and cross-race faces. Although delayed high JOLs were indicative of higher recognition memory accuracy than delayed low JOLs for both same- and cross-race faces, the accuracy of even high predictive JOLs was objectively low. Postdictive confidence was a far stronger indicator of memory accuracy than predictive JOLs; high postdictive confidence was indicative of high accuracy; and this was true for both same- and cross-race recognition memory.