Confidence-accuracy calibration in absolute and relative face recognition judgments

Greater target or lure variability? An exploration on the effects of stimulus types and memory paradigms

In recognition memory, the variance of the target distribution is almost universally found to be greater than that of the lure distribution. However, these estimates commonly come from long-term memory paradigms where words are used as stimuli. Two exceptions to this rule have found evidence for greater lure variability: a short-term memory task (Yotsumoto et al., Memory & Cognition, 36, 282-294 2008) and in an eyewitness memory paradigm (Wixted et al., Cognitive Psychology, 105, 81-114 2018). In the present work, we conducted a series of recognition memory experiments using different stimulus (faces vs. words) along with different paradigms (long-term vs. short-term paradigms) to evaluate whether either of these conditions would result in greater variability in lure items. Greater target variability was observed across stimulus types and memory paradigms. This suggests that factors other than stimuli and retention interval might be responsible for cases where variability is less for targets than lures.

Ventromedial Prefrontal Cortex Does Not Play a Selective Role in Pattern Separation

Humans have the capacity to form new memories of events that are, at times, highly similar to events experienced in the past, as well as the capacity to integrate and associate new information within existing knowledge structures. The former process relies on mnemonic discrimination and is believed to depend on hippocampal pattern separation, whereas the latter is believed to depend on generalization signals and conceptual categorization supported by the neocortex. Here, we examine whether and how the ventromedial prefrontal cortex (vMPFC) supports discrimination and generalization on a widely used task that was primarily designed to tax hippocampal processes. Ten individuals with lesions to the vMPFC and 46 neurotypical control participants were administered an adapted version of the mnemonic similarity task [Stark, S. M., Yassa, M. A., Lacy, J. W., & Stark, C. E. L. A task to assess behavioral pattern separation (BPS) in humans: Data from healthy aging and mild cognitive impairment. Neuropsychologia, 51, 2442-2449, 2013], which assesses the ability to distinguish previously learned images of everyday objects (targets) from unstudied, highly similar images (lures) and dissimilar images (foils). Relative to controls, vMPFC-lesioned individuals showed intact discrimination of lures from targets but a propensity to mistake studied targets and similar lures for dissimilar foils. This pattern was accompanied by inflated confidence despite low accuracy when responding to similar lures. These findings demonstrate a more general role of the vMPFC in memory retrieval, rather than a specific role in supporting pattern separation.

Facing up to others' emotions: No evidence of autism-related deficits in metacognitive awareness of emotion recognition

Emotion recognition difficulties are considered to contribute to social‐communicative problems for autistic individuals and awareness of such difficulties may be critical for the identification and pursuit of strategies that will mitigate their adverse effects. We examined metacognitive awareness of face emotion recognition responses in autistic (N = 63) and non‐autistic (N = 67) adults across (a) static, dynamic and social face emotion stimuli, (b) free‐ and forced‐report response formats, and (c) four different sets of the six “basic” and six “complex” emotions. Within‐individual relationships between recognition accuracy and post‐recognition confidence provided no indication that autistic individuals were poorer at discriminating correct from incorrect recognition responses than non‐autistic individuals, although both groups exhibited marked inter‐individual variability. Although the autistic group was less accurate and slower to recognize emotions, confidence‐accuracy calibration analyses provided no evidence of reduced sensitivity on their part to fluctuations in their emotion recognition performance. Across variations in stimulus type, response format and emotion, increases in accuracy were associated with progressively higher confidence, with similar calibration curves for both groups. Calibration curves for both groups were, however, characterized by overconfidence at the higher confidence levels (i.e., overall accuracy less than the average confidence level), with the non‐autistic group contributing more decisions with 90%–100% confidence. Comparisons of slow and fast responders provided no evidence of a “hard‐easy” effect—the tendency to exhibit overconfidence during hard tasks and underconfidence during easy tasks—suggesting that autistic individuals' slower recognition responding may reflect a strategic difference rather than a processing speed limitation.

Reexamining the sources of variance in recognition confidence: A reply to Kantner and Dobbins (2019)

Recently, Kantner and Dobbins (Psychonomic Bulletin & Review, 26(4), 1317-1324, 2019) reported a reanalysis of a series of previously published data sets in which they examined the sources of variation in recognition memory confidence ratings. Although between-subject differences in mean levels of confidence tended to account for the majority of variance in confidence ratings for both "old" and "new" decisions, the contribution of overall subject-level variation to confidence ratings for "new" decisions was noticeably and consistently larger. Here, I report a series of quantitative simulations along with a reanalysis of the original data to demonstrate that the relatively greater subject-level variation in mean confidence seen for "new" as compared with "old" recognition decisions largely reflects statistical constraints imposed by (a) the range limits of the ordinal scale used to measure confidence, and (b) the stronger relation between memory accuracy and confidence in "old" decisions. Therefore, any observed difference in the extent of subject-level variation in mean confidence between "old" and "new" recognition decisions need not imply a meaningful psychological distinction. These findings point out what in my view is an important statistical constraint that should be considered by researchers interested in understanding the bases of variation in memory confidence.

Forensic facial examiners versus super-recognizers: Evaluating behavior beyond accuracy

We evaluated the detailed, behavioral properties of face matching performance in two specialist groups: forensic facial examiners and super-recognizers. Both groups compare faces to determine identity with high accuracy and outperform the general population. Typically, facial examiners are highly trained; super-recognizers rely on natural ability. We found distinct behaviors between these two groups. Examiners used the full 7-point identity judgment scale (-3: "different"; +3: "same"). Super-recognizers' judgments clustered toward highly confident decisions. Examiners' judgments for same- and different-identities were symmetric across the scale midpoint (0); super-recognizers' judgments were not. Examiners showed higher identity judgment agreement than super-recognizers. Despite these qualitative differences, both groups showed insight into their own accuracy: more confident people and those who rated the task to be easier tended to be more accurate. Altogether, we show to better understand and interpret judgments according to the nature of someone's facial expertise, evaluations should assess more than accuracy.

A Quick Measure of Theory of Mind in Autistic Adults: Decision Accuracy, Latency and Self-Awareness

Autistic adults often experience difficulties in taking the perspective of others, potentially undermining their social interactions. We evaluated a quick, forced-choice version of the Adult Theory of Mind (A-ToM) test, which was designed to assess such difficulties and comprehensively evaluated by Brewer et al. (2017). The forced-choice version (the A-ToM-Q) demonstrated discriminant, concurrent, convergent and divergent validity using samples of autistic (N = 96) and non-autistic adults (N = 75). It can be administered in a few minutes and machine-scored, involves minimal training and facilitates large-scale, live, or web-based testing. It permits measurement of response latency and self-awareness, with response characteristics on both measures enhancing understanding of the nature and extent of perspective taking difficulties in autistic individuals.

Applying confidence accuracy characteristic plots to old/new recognition memory experiments

Confidence-accuracy characteristic (CAC) plots were developed for use in eyewitness identification experiments, and previous findings show that high confidence indicates high accuracy in all studies of adults with an unbiased lineup. We apply CAC plots to standard old/new recognition memory data by calculating response-based and item-based accuracy, one using false alarms and the other using misses. We use both methods to examine the confidence-accuracy relationship for both correct old responses (hits) and new responses (correct rejections). We reanalysed three sets of published data using these methods and show that the method chosen, as well as the relation of lures to targets, determines the confidence-accuracy relation. Using response-based accuracy for hits, high confidence yields quite high accuracy, and this is generally true with the other methods, especially when lures are unrelated to targets. However, when analyzing correct rejections, the relationship between confidence and accuracy is less pronounced. When lures are semantically related to targets, the various CAC plots show different confidence-accuracy relations. The different methods of calculating CAC plots provide a useful tool in analyzing standard old/new recognition experiments. The results generally accord with unequal-variance signal detection models of recognition memory.

Eyewitness identification performance is not affected by time-of-day optimality

The circadian rhythm regulates arousal levels throughout the day and determines optimal periods for engaging in mental activities. Individuals differ in the time of day at which they reach their peak: Morning-type individuals are at their best in the morning and evening types perform better in the evening. Performance in recall and recognition of non-facial stimuli is generally superior at an individual’s circadian peak. In two studies (Ns = 103 and 324), we tested the effect of time-of-testing optimality on eyewitness identification performance. Morning- and evening-type participants viewed stimulus films depicting staged crimes and made identification decisions from target-present and target-absent lineups either at their optimal or non-optimal time-of-day. We expected that participants would make more accurate identification decisions and that the confidence-accuracy and decision time-accuracy relationships would be stronger at optimal compared to non-optimal time of day. In Experiment 1, identification accuracy was unexpectedly superior at non-optimal compared to optimal time of day in target-present lineups. In Experiment 2, identification accuracy did not differ between the optimal and non-optimal time of day. Contrary to our expectations, confidence-accuracy relationship was generally stronger at non-optimal compared to optimal time of day. In line with our predictions, non-optimal testing eliminated decision-time-accuracy relationship in Experiment 1.

Developmental trends in lineup performance: Adolescents are more prone to innocent bystander misidentifications than children and adults

We tested developmental trends in eyewitness identification in biased and unbiased lineups. Our main interest was adolescent's lineup performance compared with children and adults. 7-10-year-olds, 11-13-year-olds, 14-16-year-olds, and adults (N = 431) watched a wallet-theft-video and subsequently identified the thief, victim, and witness from simultaneous target-present and target-absent six-person photo lineups. The thief-absent lineup included a bystander previously seen in thief proximity. Research on unconscious transference suggested a selection bias toward the bystander in adults and 11-13-year-olds, but not in younger children. Confirming our hypothesis, adolescents were more prone to bystander bias than all other age groups. This may be due to adolescents making more inferential errors than children, as predicted by fuzzy-trace theory and associative-activation theory, combined with lower inhibition control in adolescents compared with adults. We also replicated a clothing bias for all age groups and age-related performance differences in our unbiased lineups. Consistent with previous findings, participants were generally overconfident in their decisions, even though confidence was a better predictor of accuracy in older compared with younger participants. With this study, we show that adolescents have an increased tendency to misidentify an innocent bystander. Continued efforts are needed to disentangle how adolescents in comparison to other age groups perform in forensically relevant situations.

Diagnostic error increases mortality and length of hospital stay in patients presenting through the emergency room

Background

Diagnostic errors occur frequently, especially in the emergency room. Estimates about the consequences of diagnostic error vary widely and little is known about the factors predicting error. Our objectives thus was to determine the rate of discrepancy between diagnoses at hospital admission and discharge in patients presenting through the emergency room, the discrepancies’ consequences, and factors predicting them.

Methods

Prospective observational clinical study combined with a survey in a University-affiliated tertiary care hospital. Patients’ hospital discharge diagnosis was compared with the diagnosis at hospital admittance through the emergency room and classified as similar or discrepant according to a predefined scheme by two independent expert raters. Generalized linear mixed-effects models were used to estimate the effect of diagnostic discrepancy on mortality and length of hospital stay and to determine whether characteristics of patients, diagnosing physicians, and context predicted diagnostic discrepancy.

Results

755 consecutive patients (322 [42.7%] female; mean age 65.14 years) were included.

The discharge diagnosis differed substantially from the admittance diagnosis in 12.3% of cases. Diagnostic discrepancy was associated with a longer hospital stay (mean 10.29 vs. 6.90 days; Cohen’s d 0.47; 95% confidence interval 0.26 to 0.70; P = 0.002) and increased patient mortality (8 (8.60%) vs. 25(3.78%); OR 2.40; 95% CI 1.05 to 5.5 P = 0.038). A factor available at admittance that predicted diagnostic discrepancy was the diagnosing physician’s assessment that the patient presented atypically for the diagnosis assigned (OR 3.04; 95% CI 1.33–6.96; P = 0.009).

Conclusions

Diagnostic discrepancies are a relevant healthcare problem in patients admitted through the emergency room because they occur in every ninth patient and are associated with increased in-hospital mortality. Discrepancies are not readily predictable by fixed patient or physician characteristics; attention should focus on context.

Trial registration

https://bmjopen.bmj.com/content/6/5/e011585

Electronic supplementary material

The online version of this article (10.1186/s13049-019-0629-z) contains supplementary material, which is available to authorized users.

The importance of decision bias for predicting eyewitness lineup choices: toward a Lineup Skills Test

ᅟ

We report on research on individual-difference measures that could be used to assess the validity of eyewitness identification decisions.

Background

The predictive utility of face recognition tasks for eyewitness identification has received some attention from psychologists, but the previous research focused primarily on witnesses’ likelihood of correctly choosing the culprit when present in a lineup. Far less discussed has been individual differences in witnesses’ proclivity to choose from a lineup that does not contain the culprit. We designed a two-alternative non-forced-choice face recognition task (consisting of mini-lineup test pairs, half old/new and half new/new) to predict witnesses’ proclivity to choose for a set of culprit-absent lineups associated with earlier-viewed crime videos.

Results

In two studies involving a total of 402 participants, proclivity to choose on new/new pairs predicted mistaken identifications on culprit-absent lineups, with r values averaging .43. The likelihood of choosing correctly on old/new pairs (a measure of face recognition skill) was only weakly predictive of correct identifications in culprit-present lineups (mean r of .22).

Conclusions

Our findings could be the basis for further research aimed at developing a standardized measure of proclivity to choose that could be used, along with other measures, to weigh eyewitnesses’ lineup identification decisions.

Electronic supplementary material

The online version of this article (10.1186/s41235-018-0150-3) contains supplementary material, which is available to authorized users.

Face value: testing the utility of contextual face cues for face recognition

The presence of multiple faces during a crime may provide a naturally-occurring contextual cue to support eyewitness recognition for those faces later. Across two experiments, we sought to investigate mechanisms underlying previously-reported cued recognition effects, and to determine whether such effects extended to encoding conditions involving more than two faces. Participants studied sets of individual faces, pairs of faces, or groups of four faces. At test, participants in the single-face condition were tested only on those individual faces without cues. Participants in the two and four-face conditions were tested using no cues, correct cues (a face previously studied with the target test face), or incorrect cues (a never-before-seen face). In Experiment 2, associative encoding was promoted by a rating task. Neither hit rates nor false-alarm rates were significantly affected by cue type or face encoding condition in Experiment 1, but cuing of any kind (correct or incorrect) in Experiment 2 appeared to provide a protective buffer to reduce false-alarm rates through a less liberal response bias. Results provide some evidence that cued recognition techniques could be useful to reduce false recognition, but only when associative encoding is strong.

[Independence of old and new item processing in recognition judgment]

This study investigated the independence of old and new item processing in recognition judgment. Conventional recognition studies have focused on the “memory” aspects of recognition judgment, while new item processing has been treated as a “residual response” of old item processing. However, prior studies on confidence ratings and the brain research suggest the possibility that old and new items are independently processed. This study conducted recognition judgment tests with the number of learned items as independent variables, and response times for four answer categories (hit, miss, correct rejection, and false alarm) as dependent variables. Hit response times were consistently shorter than misses for all old items, while correct rejection (CR) and false alarm (FA) response times approached equivalence (CR = FA) for new items as the number of old (memorized) items increased. These results suggest that recognition judgment changes according to the number of old (memorized) items. We discuss the idea that new item and old item processing in recognition judgment occur independently.

Decision time and confidence predict choosers' identification performance in photographic showups

In vast contrast to the multitude of lineup studies that report on the link between decision time, confidence, and identification accuracy, only a few studies looked at these associations for showups, with results varying widely across studies. We therefore set out to test the individual and combined value of decision time and post-decision confidence for diagnosing the accuracy of positive showup decisions using confidence-accuracy characteristic curves and Bayesian analyses. Three-hundred-eighty-four participants viewed a stimulus event and were subsequently presented with two showups which could be target-present or target-absent. As expected, we found a negative decision time-accuracy and a positive post-decision confidence-accuracy correlation for showup selections. Confidence-accuracy characteristic curves demonstrated the expected additive effect of combining both postdictors. Likewise, Bayesian analyses, taking into account all possible target-presence base rate values showed that fast and confident identification decisions were more diagnostic than slow or less confident decisions, with the combination of both being most diagnostic for postdicting accurate and inaccurate decisions. The postdictive value of decision time and post-decision confidence was higher when the prior probability that the suspect is the perpetrator was high compared to when the prior probability that the suspect is the perpetrator was low. The frequent use of showups in practice emphasizes the importance of these findings for court proceedings. Overall, these findings support the idea that courts should have most trust in showup identifications that were made fast and confidently, and least in showup identifications that were made slowly and with low confidence.

The Relationship Between Eyewitness Confidence and Identification Accuracy: A New Synthesis

Summary

The U.S. legal system increasingly accepts the idea that the confidence expressed by an eyewitness who identified a suspect from a lineup provides little information as to the accuracy of that identification. There was a time when this pessimistic assessment was entirely reasonable because of the questionable eyewitness-identification procedures that police commonly employed. However, after more than 30 years of eyewitness-identification research, our understanding of how to properly conduct a lineup has evolved considerably, and the time seems ripe to ask how eyewitness confidence informs accuracy under more pristine testing conditions (e.g., initial, uncontaminated memory tests using fair lineups, with no lineup administrator influence, and with an immediate confidence statement). Under those conditions, mock-crime studies and police department field studies have consistently shown that, for adults, (a) confidence and accuracy are strongly related and (b) high-confidence suspect identifications are remarkably accurate. However, when certain non-pristine testing conditions prevail (e.g., when unfair lineups are used), the accuracy of even a high-confidence suspect ID is seriously compromised. Unfortunately, some jurisdictions have not yet made reforms that would create pristine testing conditions and, hence, our conclusions about the reliability of high-confidence identifications cannot yet be applied to those jurisdictions. However, understanding the information value of eyewitness confidence under pristine testing conditions can help the criminal justice system to simultaneously achieve both of its main objectives: to exonerate the innocent (by better appreciating that initial, low-confidence suspect identifications are error prone) and to convict the guilty (by better appreciating that initial, high-confidence suspect identifications are surprisingly accurate under proper testing conditions).

Knowing your heart and your mind: The relationships between metamemory and interoception

Humans experience a unified self that integrates our mental lives and physical bodies, but many studies focus on isolated domains of self-knowledge. We tested the hypothesis that knowledge of one's mind and body are related by examining metamemory and interoception. We evaluated two dimensions of metamemory and interoception: subjective beliefs and the accuracy of those beliefs compared to objective criteria. We first demonstrated, in two studies, that metamemory beliefs were positively correlated with interoceptive beliefs, and this was not due to domain-general confidence. Finally, we showed that individuals with better metamemory accuracy also had better interoceptive accuracy. Taken together, these findings suggest a common mechanism subserving knowledge of our cognitive and bodily states.

Factors predicting a change in diagnosis in patients hospitalised through the emergency room: a prospective observational study

INTRODUCTION

Emergency rooms (ERs) generally assign a preliminary diagnosis to patients, who are then hospitalised and may subsequently experience a change in their lead diagnosis (cDx). In ERs, the cDx rate varies from around 15% to more than 50%. Among the most frequent reasons for diagnostic errors are cognitive slips, which mostly result from faulty data synthesis. Furthermore, physicians have been repeatedly found to be poor self-assessors and to be overconfident in the quality of their diagnosis, which limits their ability to improve. Therefore, some of the clinically most relevant research questions concern how diagnostic decisions are made, what determines their quality and what can be done to improve them. Research that addresses these questions is, however, still rare. In particular, field studies that allow for generalising findings from controlled experimental settings are lacking. The ER, with its high throughput and its many simultaneous visits, is perfectly suited for the study of factors contributing to diagnostic error. With this study, we aim to identify factors that allow prediction of an ER's diagnostic performance. Knowledge of these factors as well as of their relative importance allows for the development of organisational, medical and educational strategies to improve the diagnostic performance of ERs.

METHODS AND ANALYSIS

We will conduct a field study by collecting diagnostic decision data, physician confidence and a number of influencing factors in a real-world setting to model real-world diagnostic decisions and investigate the adequacy, validity and informativeness of physician confidence in these decisions. We will specifically collect data on patient, physician and encounter factors as predictors of the dependent variables. Statistical methods will include analysis of variance and a linear mixed-effects model.

ETHICS AND DISSEMINATION

The Bern ethics committee approved the study under KEK Number 197/15. Results will be published in peer-reviewed scientific medical journals. Authorship will be determined according to ICMJE guidelines.

TRIAL REGISTRATION NUMBER

The study protocol Version 1.0 from 17 May 2015 is registered in the Inselspital Research Database Information System (IRDIS) and with the IRB ('Kantonale Ethikkomission') Bern under KEK Number 197/15.

Monitoring and meta-metacognition in the own-race bias

Although there is a great deal of research focused on identification issues related to own-versus other-race faces very few experiments have explored whether metacognitive monitoring contributes to the own-race bias. In the current experiment the typical own-race bias paradigm was modified so that type-2 signal detection measures (e.g. Higham & Arnold, 2007a,b) could be used to directly measure metacognitive monitoring at retrieval. A second goal of the experiment was to explore whether self-reported confidence ratings differed depending on whether they were directed at answer accuracy (e.g., judging a face as "studied") versus at decisions about that answer (e.g., volunteering vs. withholding that answer). Overall the results demonstrated that monitoring does contribute to the own-race bias, in that participants were better at monitoring their memory for own-race faces. Further, there was a significant difference between the two confidence measures, and the pattern of this difference depended on whether responses had been volunteered or withheld.

Exploring Qualitative Differences in Knowledge Sources

Focusing on knowledge management (KM) and strategic decision making in service businesses through the constructs of strategic capital and knowledge sharing, the study investigates qualitative differences in domain-specific knowledge of frontline employees and executives. The study draws on cognitive theory and investigates the extent to which the knowledge of these subject groups is correct with respect to incorporating intuitive judgments by various employee groups into forecasting and following strategic decision making. The authors carried out this investigation through an exploratory study of the subject groups’ confidence and accuracy (CA) performance in a constructed knowledge-based forecasting setting. The groups’ intuitive judgmental performances were examined when predicting uncertain business and industry-related outcomes. The authors surveyed 39 executives and 38 frontline employees in 12 hotels. The analysis is based on a between-participants design. The results from this setting do not fully confirm findings in earlier CA studies. Their results indicate that there are no significant differences in the accuracy of executives (as experts) and frontline employees (as novices). Although executives demonstrate overconfidence in their judgments and frontline employees demonstrate under confidence, in line with earlier CA theory of experts and novices, the differences we find are not significant. Similarly, the CA calibration performance difference between the two groups is not significant. They suggest, among other reasons, that our findings differ from earlier CA studies because of organizational politics and culture by power distance, social capital, misuse of knowledge and the size of the business.

Children’s metacognitive judgments in an eyewitness identification task

Two experiments examined children's metacognitive monitoring of recognition judgments within an eyewitness identification paradigm. A confidence-accuracy (CA) calibration approach was used to examine patterns of calibration, over-/underconfidence, and resolution. In Experiment 1, children (n=619, mean age=11 years 10 months) and adults (n=600) viewed a simulated crime and attempted two separate identifications from 8-person target-present or target-absent lineups given lineup instructions that manipulated witnesses choosing patterns by varying the degree of social pressure. For choosers, but not nonchoosers, meaningful CA relations were observed for adults but not for children. Experiment 2 tested a guided hypothesis disconfirmation manipulation designed to improve the realism of children's metacognitive judgments. Children (N=796, mean age=11 years 11 months) in experimental and control conditions viewed a crime and attempted two separate identifications. The manipulation had minimal impact on the CA relation for choosers and nonchoosers. In contrast to adults, children's identification confidence provides no useful guide for investigators about the likely guilt or innocence of a suspect. These experiments revealed limitations in children's metacognitive monitoring processes that have not been apparent in previous research on recall and recognition with younger children.

An application of the poisson race model to confidence calibration

In tasks as diverse as stock market predictions and jury deliberations, a person's feelings of confidence in the appropriateness of different choices often impact that person's final choice. The current study examines the mathematical modeling of confidence calibration in a simple dual-choice task. Experiments are motivated by an accumulator model, which proposes that information supporting each alternative accrues on separate counters. The observer responds in favor of whichever alternative's counter first hits a designated threshold. Confidence can then be scaled from the difference between the counters at the time that the observer makes a response. The authors examine the overconfidence result in general and present new findings dealing with the effect of response bias on confidence calibration.

Eyewitness Evidence: Improving Its Probative Value

The criminal justice system relies heavily on eyewitnesses to determine the facts surrounding criminal events. Eyewitnesses may identify culprits, recall conversations, or remember other details. An eyewitness who has no motive to lie is a powerful form of evidence for jurors, especially if the eyewitness appears to be highly confident about his or her recollection. In the absence of definitive proof to the contrary, the eyewitness's account is generally accepted by police, prosecutors, judges, and juries. However, the faith the legal system places in eyewitnesses has been shaken recently by the advent of forensic DNA testing. Given the right set of circumstances, forensic DNA testing can prove that a person who was convicted of a crime is, in fact, innocent. Analyses of DNA exoneration cases since 1992 reveal that mistaken eyewitness identification was involved in the vast majority of these convictions, accounting for more convictions of innocent people than all other factors combined. We review the latest figures on these DNA exonerations and explain why these cases can only be a small fraction of the mistaken identifications that are occurring. Decades before the advent of forensic DNA testing, psychologists were questioning the validity of eyewitness reports. Hugo Münsterberg's writings in the early part of the 20th century made a strong case for the involvement of psychological science in helping the legal system understand the vagaries of eyewitness testimony. But it was not until the mid- to late 1970s that psychologists began to conduct programmatic experiments aimed at understanding the extent of error and the variables that govern error when eyewitnesses give accounts of crimes they have witnessed. Many of the experiments conducted in the late 1970s and throughout the 1980s resulted in articles by psychologists that contained strong warnings to the legal system that eyewitness evidence was being overvalued by the justice system in the sense that its impact on triers of fact (e.g., juries) exceeded its probative (legal-proof) value. Another message of the research was that the validity of eyewitness reports depends a great deal on the procedures that are used to obtain those reports and that the legal system was not using the best procedures. Although defense attorneys seized on this nascent research as a tool for the defense, it was largely ignored or ridiculed by prosecutors, judges, and police until the mid 1990s, when forensic DNA testing began to uncover cases of convictions of innocent persons on the basis of mistaken eyewitness accounts. Recently, a number of jurisdictions in the United States have implemented procedural reforms based on psychological research, but psychological science has yet to have its fullest possible influence on how the justice system collects and interprets eyewitness evidence. The psychological processes leading to eyewitness error represent a confluence of memory and social-influence variables that interact in complex ways. These processes lend themselves to study using experimental methods. Psychological science is in a strong position to help the criminal justice system understand eyewitness accounts of criminal events and improve their accuracy. A subset of the variables that affect eyewitness accuracy fall into what researchers call system variables, which are variables that the criminal justice system has control over, such as how eyewitnesses are instructed before they view a lineup and methods of interviewing eyewitnesses. We review a number of system variables and describe how psychological scientists have translated them into procedures that can improve the probative value of eyewitness accounts. We also review estimator variables, variables that affect eyewitness accuracy but over which the system has no control, such as cross-race versus within-race identifications. We describe some concerns regarding external validity and generalization that naturally arise when moving from the laboratory to the real world. These include issues of base rates, multicollinearity, selection effects, subject populations, and psychological realism. For each of these concerns, we briefly note ways in which both theory and field data help make the case for generalization.

The confidence-accuracy relationship in eyewitness identification: effects of lineup instructions, foil similarity, and target-absent base rates

Discriminating accurate from mistaken eyewitness identifications is a major issue facing criminal justice systems. This study examined whether eyewitness confidence assists such decisions under a variety of conditions using a confidence-accuracy (CA) calibration approach. Participants (N = 1,200) viewed a simulated crime and attempted 2 separate identifications from 8-person target-present or target-absent lineups. Confidence and accuracy were calibrated for choosers (but not nonchoosers) for both targets under all conditions. Lower overconfidence was associated with higher diagnosticity, lower target-absent base rates, and shorter identification latencies. Although researchers agree that courtroom expressions of confidence are uninformative, our findings indicate that confidence assessments obtained immediately after a positive identification can provide a useful guide for investigators about the likely accuracy of an identification.

Eyewitness Identification Accuracy and Response Latency: The Unruly 10-12-Second Rule

Data are reported from 3,213 research eyewitnesses confirming that accurate eyewitness identifications from lineups are made faster than are inaccurate identifications. However, consistent with predictions from the recognition and search literatures, the authors did not find support for the "10-12-s rule" in which lineup identifications faster than 10-12 s maximally discriminate between accurate and inaccurate identifications (D. Dunning & S. Perretta, 2002). Instead, the time frame that proved most discriminating was highly variable across experiments, ranging from 5 s to 29 s, and the maximally discriminating time was often unimpressive in its ability to sort accurate from inaccurate identifications. The authors suggest several factors that are likely to moderate the 10-12-s rule.