Multiple confidence estimates as indices of eyewitness memory

Eyewitness identification decisions are vulnerable to various influences on witnesses' decision criteria that contribute to false identifications of innocent suspects and failures to choose perpetrators. An alternative procedure using confidence estimates to assess the degree of match between novel and previously viewed faces was investigated. Classification algorithms were applied to participants' confidence data to determine when a confidence value or pattern of confidence values indicated a positive response. Experiment 1 compared confidence group classification accuracy with a binary decision control group's accuracy on a standard old-new face recognition task and found superior accuracy for the confidence group for target-absent trials but not for target-present trials. Experiment 2 used a face mini-lineup task and found reduced target-present accuracy offset by large gains in target-absent accuracy. Using a standard lineup paradigm, Experiments 3 and 4 also found improved classification accuracy for target-absent lineups and, with a more sophisticated algorithm, for target-present lineups. This demonstrates the accessibility of evidence for recognition memory decisions and points to a more sensitive index of memory quality than is afforded by binary decisions.     

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

Discrimination accuracy is usually higher for same- than for cross-race faces, a phenomenon known as the cross-race effect (CRE). According to prior research, the CRE occurs because memories for same- and cross-race faces rely on qualitatively different processes. However, according to a continuous dual-process model of recognition memory, memories that rely on qualitatively different processes do not differ in recognition accuracy when confidence is equated. Thus, although there are differences in overall same- and cross-race discrimination accuracy, confidence-specific accuracy (i.e., recognition accuracy at a particular level of confidence) may not differ. We analysed datasets from four recognition memory studies on same- and cross-race faces to test this hypothesis. Confidence ratings reliably predicted recognition accuracy when performance was above chance levels (Experiments 1, 2, and 3) but not when performance was at chance levels (Experiment 4). Furthermore, at each level of confidence, confidence-specific accuracy for same- and cross-race faces did not significantly differ when overall performance was above chance levels (Experiments 1, 2, and 3) but significantly differed when overall performance was at chance levels (Experiment 4). Thus, under certain conditions, high-confidence same-race and cross-race identifications may be equally reliable.     

Applying the attractor field model to social cognition: Perceptual discrimination is facilitated, but memory is impaired for faces displaying evaluatively congruent expressions

A new model of mental representation is applied to social cognition: the attractor field model. Using the model, the authors predicted and found a perceptual advantage but a memory disadvantage for faces displaying evaluatively congruent expressions. In Experiment 1, participants completed a same/different perceptual discrimination task involving morphed pairs of angry-to-happy Black and White faces. Pairs of faces displaying evaluatively incongruent expressions (i.e., happy Black, angry White) were more likely to be labeled as similar and were less likely to be accurately discriminated from one another than faces displaying evaluatively congruent expressions (i.e., angry Black, happy White). Experiment 2 replicated this finding and showed that objective discriminability of stimuli moderated the impact of attractor field effects on perceptual discrimination accuracy. In Experiment 3, participants completed a recognition task for angry and happy Black and White faces. Consistent with the attractor field model, memory accuracy was better for faces displaying evaluatively incongruent expressions. Theoretical and practical implications of these findings are discussed.     

Subjective experience guides betting decisions beyond accuracy: evidence from a metamemory illusion

The goal of this research was to test whether subjective memory experiences drive accuracy regulation decisions above and beyond objective memory indices. In four experiments (n?=?115) subjective recollection (i.e., reporting “Remember” in the Remember-Know task) was dissociated from memory accuracy by manipulating retrieval during a two-alternative forced-choice recognition task: in the Match condition the distracter was a novel exemplar of the target (e.g., a studied and an unstudied toaster) and in the Non-match condition the distracter was a novel exemplar of another studied but untested item (e.g., a studied toaster and an unstudied birdhouse). Participants were more accurate on Match trials, but reported subjective recollection more frequently on Non-match trials. Critically, participants also bet more often on Non-match trials to the detriment of their score (Experiment 1). This pattern persisted when participants were additionally required to retrieve details about items (Experiment 2) and when confidence assessments were collected (Experiment 3). Finally, participants bet more on Non-match trials even when subjective judgments were not elicited, suggesting that the decision process does not require reporting on subjective experience (Experiment 4). These results indicate that subjective memory experiences guide decision-making independent of objective accuracy and thus are critical to accuracy regulation.     

Choosing what to choose from: Preference for inclusion over exclusion when constructing consideration sets from large choice sets

Decision making is a two‐stage process, consisting of, first, consideration set construction and then final choice. Decision makers can form a consideration set from a choice set using one of two strategies: including the options they wish to further consider or excluding those they do not wish to further consider. The authors propose that decision makers have a relative preference for an inclusion (vs. exclusion) strategy when choosing from large choice sets and that this preference is driven primarily by a lay belief that inclusion requires less effort than exclusion, particularly in large choice sets. Study 1 demonstrates that decision makers prefer using an inclusion (vs. exclusion) strategy when faced with large choice sets. Study 2 replicates the effect of choice set size on preference for consideration set construction strategy and demonstrates that the belief that exclusion is more effortful mediates the relative preference for inclusion in large choice sets. Studies 3 and 4 further support the importance of perceived effort, demonstrating a greater preference for inclusion in large choice sets when decision makers are primed to think about effort (vs. accuracy; Study 3) and when the choice set is perceived as requiring more effort because of more information being presented about each alternative (vs. more alternatives in the choice set; Study 4). Finally, Study 5 manipulates consideration set construction strategy, showing that using inclusion (vs. exclusion) in large choice sets leads to smaller consideration sets, greater confidence in the decision process, and a higher quality consideration set.     

Comparative difficulty and the strategic regulation of accuracy: The impact of test-list context on monitoring and meta-metacognition

A growing body of research has shown that context manipulations can have little or no impact on accuracy performance, yet still significantly influence metacognitive performance. For example, participants in a test-list context paradigm study one list of words with a medium levels-of-processing task and a second word list with either a shallow or deep task: Recognition for medium words does not differ across conditions, however medium words are significantly more likely to be labeled as “remembered” (vs. merely familiar) if they had been studied with a shallow word list (Bodner & Lindsay, 2003). The goal of the current studies was to extend the test-list context paradigm to strategic regulation (report/withhold recognition test), and broaden it to incorporate different types of stimuli (i.e., face stimuli in place of a medium word list). The paradigm also was modified to include separate answer (studied/new) confidence and decision (report/withhold) confidence ratings at test. Results showed that context did not impact recognition accuracy for faces across the context conditions, however participants were more likely to report (i.e., volunteer) their face responses if they had studied the shallow word list. The results also demonstrated a difference between answer confidence and decision confidence, and the pattern of this difference depended on whether responses were reported or withheld (Experiment 1). Overall, the data are presented as support for the functional account of memory, which views memory states as inferential and attributional rather than static categories.     

Postidentification feedback effects on face recognition confidence: evidence for metacognitive influences

Information given to witnesses after an identification decision greatly alters their impressions of the original event and importantly, their identification confidence. Two experiments investigated the possibility that the effect of feedback on confidence may be altered according to the strength of the witness's cues to accuracy. Experiment 1 used a manipulation of exposure duration to alter recognition accuracy prior to the delivery of confirming, disconfirming or no feedback. While the feedback effect was not different across exposure duration conditions, decisions that were made more quickly were less likely to show large changes in confidence due to feedback. Experiment 2 manipulated the distinctiveness of faces and showed that the effects of feedback on confidence, and on the resolution of the confidence judgement, were more pronounced when disconfirming feedback was given for distinctive faces and when confirming feedback was given for typical faces. These studies showed that the impressions that participants formed of their likely accuracy might moderate the effects of feedback on decision confidence. Copyright © 2006 John Wiley & Sons, Ltd.     

Effects of divided attention and social categorization on the own-race bias in face recognition

Perceivers remember own-race faces more accurately than other-race faces (i.e., Own-Race Bias). In the current experiments, we manipulated participants' attentional resources and social group membership to explore their influence on own and other-race face recognition memory. In Experiment 1, Chinese participants viewed own-race and Caucasian faces, and between-subjects we manipulated whether participants attention was divided during face encoding. We found that divided attention eliminated the Own-Race Bias in memory due to a reduction of memory accuracy for own-race faces, implicating that attention allocation plays a role in creating the bias. In Experiment 2, Chinese participants completed an ostensible personality test. Some participants were informed that their personality traits were most commonly found in Caucasian (i.e., other-race) individuals, resulting in these participants sharing a group membership with other-race targets. In contrast, other participants were not told anything about the personality test, resulting in the default own-race group membership. The participants encoded the faces for a subsequent recognition memory test either with or without performing a concurrent arithmetic distracting task. Results showed that other-race group membership and reducing attention during encoding independently eliminated the typical Own-Race Bias in face memory. The implications of these findings on perceptual-expertise and social-categorization models are discussed.     

Effects of Attribute Scales on Process and Performance in Multiattribute Choice

The type of scale used to represent the attributes of decision alternatives has received little attention in previous research, despite the pervasive presence of this issue in decision making. In two process tracing experiments we investigated how the type of scale used affects choices and choice processes. We hypothesized that scales affect the ease of accomplishing various component processes in choice. Experiment 1 used scales that differed in two ways: (1) whether all attributes were on the same scale (i.e., commonality) and (2) whether the scales were in units that were meaningful in the problem domain (i.e., context-relevance). We also manipulated economic incentives to favor speed over accuracy or the converse. Participants made more accurate decisions with common, context-independent scales but faster decisions with context-relevant scales. In addition, choice processes mediated the effects of scaling on both speed and accuracy. Experiment 2 replicated the results of Experiment 1 with decision contexts that differed in familiarity. Both experiments demonstrate that the scales used to present attributes can influence both the speed and accuracy of decision making, despite the presence of explicit economic incentives. Designers of choice studies and information displays are therefore advised to exercise care in choosing attribute scales.     

Participants' free‐hand drawings of a target face can influence recognition accuracy and the confidence–accuracy correlation

Two experiments were conducted to examine the effects of participants' free‐hand drawings on recognition accuracy and confidence for targets presented in a standard recognition paradigm and a lineup identification task. For both experiments, drawing a target influenced recognition accuracy and the confidence–accuracy correlation. In Experiment 1, the confidence–accuracy correlation was higher for participants completing a drawing than controls. Experiment 2 examined the drawings in relation to participants' decision to choose from the lineups. Once choosing was statistically controlled, the confidence–accuracy correlation for drawers was not higher than controls. These results suggest that the drawing influences the confidence–accuracy correlation by decreasing the likelighood of erroneously rejecting a target‐present lineup. Copyright © 2002 John Wiley & Sons, Ltd.     

When false recognition is out of control: The case of facial conjunctions

In three experiments, a dual-process approach to face recognition memory is examined, with a specific focus on the idea that a recollection process can be used to retrieve configural information of a studied face. Subjects could avoid, with confidence, a recognition error to conjunction lure faces (each a reconfiguration of features from separate studied faces) or feature lure faces (each based on a set of old features and a set of new features) by recalling a studied configuration. In Experiment 1, study repetition (one vs. eight presentations) was manipulated, and in Experiments 2 and 3, retention interval over a short number of trials (0\2-20) was manipulated. Different measures converged on the conclusion that subjects were unable to use a recollection process to retrieve configural information in an effort to temper recognition errors for conjunction or feature lure faces. A single process, familiarity, appears to be the sole process underlying recognition of conjunction and feature faces, and familiarity contributes, perhaps in whole, to discrimination of old from conjunction faces.     

Categorical and dimensional perceptions in decoding emotional facial expressions

We investigated whether categorical perception and dimensional perception can co-occur while decoding emotional facial expressions. In Experiment 1, facial continua with endpoints consisting of four basic emotions (i.e., happiness–fear and anger–disgust) were created by a morphing technique. Participants rated each facial stimulus using a categorical strategy and a dimensional strategy. The results show that the happiness–fear continuum was divided into two clusters based on valence, even when using the dimensional strategy. Moreover, the faces were arrayed in order of the physical changes within each cluster. In Experiment 2, we found a category boundary within other continua (i.e., surprise–sadness and excitement–disgust) with regard to the arousal and valence dimensions. These findings indicate that categorical perception and dimensional perception co-occurred when emotional facial expressions were rated using a dimensional strategy, suggesting a hybrid theory of categorical and dimensional accounts.     

On interpreting the relationship between remember–know judgments and confidence: The role of instructions

Two experiments were designed to test the hypothesis that the nature of the remember–know instructions given to participants influences whether these responses reflect different memory states or different degrees of memory confidence. Participants studied words and nonwords, a variable that has been shown to dissociate confidence from remember–know judgments and were given a set of published remember–know instructions that either emphasized know judgments as highly confident (Experiment 1) or as less confident (Experiment 2) states of recognition. Experiment 1 replicated the standard finding showing that remembering and knowing were differently influenced by the word–nonword variable, whereas confidence responses were not. By contrast, Experiment 2 showed a similar pattern of data for remember–know and sure–unsure responses, thus demonstrating the importance of the instructions for interpreting the relationship between remembering and knowing and confidence.     

The own‐age face recognition bias is task dependent

The own‐age bias (OAB) in face recognition (more accurate recognition of own‐age than other‐age faces) is robust among young adults but not older adults. We investigated the OAB under two different task conditions. In Experiment 1 young and older adults (who reported more recent experience with own than other‐age faces) completed a match‐to‐sample task with young and older adult faces; only young adults showed an OAB. In Experiment 2 young and older adults completed an identity detection task in which we manipulated the identity strength of target and distracter identities by morphing each face with an average face in 20% steps. Accuracy increased with identity strength and facial age influenced older adults' (but not younger adults') strategy, but there was no evidence of an OAB. Collectively, these results suggest that the OAB depends on task demands and may be absent when searching for one identity.     

Asking an eyewitness to predict their later lineup performance could harm the confidence–accuracy relationship

Few studies have investigated eyewitnesses' ability to predict their later lineup performance, known as predecision confidence. We applied calibration analysis in two experiments comparing predecision confidence (immediately after encoding but prior to a lineup) to postdecision confidence (immediately after a lineup) to determine which produces a superior relationship with lineup decision accuracy. Experiment 1 (N = 177) featured a multiple-block lineup recognition paradigm featuring several targets and lineups; Experiment 2 featured an eyewitness identification paradigm with a mock-crime video and a single lineup for each participant (N = 855). Across both experiments, postdecision confidence discriminated well between correct and incorrect lineup decisions, but predecision confidence was a poor predictor of accuracy. Moreover, simply asking for predecision confidence weakened the postdecision confidence–accuracy relationship. This implies that police should exercise caution when interviewing eyewitnesses, as they should not be asked to predict their ability to make an accurate lineup decision.     

Using recall to reduce false recognition: diagnostic and disqualifying monitoring

Whether recall of studied words (e.g., parsley, rosemary, thyme) could reduce false recognition of related lures (e.g., basil) was investigated. Subjects studied words from several categories for a final recognition memory test. Half of the subjects were given standard test instructions, and half were instructed to use recall to reduce false recognition. Manipulation checks indicated that the latter instructions did elicit a recall-to-reject strategy. However, false recognition was selectively reduced only when all the words from a category could be recalled (Experiment 1). When longer categories were used, thereby minimizing exhaustive recall, a recall-to-reject strategy was ineffective at reducing false recognition (Experiment 2). It is suggested that exhaustively recalling a category allowed subjects to disqualify the lure as having occurred, analogous to recall-to-reject demonstrations in other tasks. In contrast, partially recalling a category did not help to diagnose the lure as nonstudied. These findings constrain theories of recall-based monitoring processes.     

More Inaccurate But Not More Biased: Anxiety During Encoding Impairs Face Recognition Accuracy But Does Not Moderate the Own‐Ethnicity Bias

Heightened state anxiety can have a deleterious impact on memory for faces. In this paper we investigated whether anxiety: (i) moderates the own‐ethnicity bias (OEB) and (ii) impairs face recognition accuracy at the encoding or retrieval phase of an OEB face‐recognition task. Using a typical OEB task, anxiety was induced during encoding and retrieval in Experiment 1, but only during retrieval in Experiment 2. An OEB was found in both experiments, but anxiety did not moderate the OEB in either experiment. In Experiment 1, anxious participants were poorer at face recognition for both own‐ and other‐ethnicity faces. In Experiment 2 anxiety did not impair face recognition. Together, these studies suggest that anxiety impaired participants' encoding, but not retrieval, of faces. The implications of these findings are discussed. Copyright © 2015 John Wiley & Sons, Ltd.     

No evidence that alcohol intoxication impairs judgments of learning in face recognition

Alcohol use is frequently involved in crime, making it crucial to understand the role of alcohol in facial recognition to maximize correct perpetrator identifications. Although the majority of the alcohol and face recognition research has investigated recognition with retrospective confidence judgments, we examined the effects of alcohol intoxication on face recognition with prospective metacognitive judgments. Participants (N = 54 university students without a history of hazardous alcohol/substance use) consumed either alcohol (mean breath alcohol concentration of 0.06 at pretest and 0.07 at post‐test) or a non‐alcoholic placebo drink. Participants then studied unfamiliar male and female faces and made judgments of learning (JOLs) for each face (i.e., predicted the likelihood of recognizing that face on a future memory test). After a brief distractor task, participants completed an old–new recognition test on which they attempted to distinguish the studied faces from new faces. It was found that the alcohol manipulation had minimal effect on face recognition performance or judgments of learning. Our results suggest that theory‐based cues about the effects of alcohol might play a greater role in retrospective judgments than prospective judgments. Although not a primary focus of the study, face recognition was better for male faces than female faces, and this occurred for both female and male participants. Limitations and implications of the research are discussed.     

个体自信度对双人决策的影响

群体决策是重要的社会现象, 个体自信度在群体决策中发挥了重要作用。本文开展了不同难度和信息交流方式下的双人决策实验, 通过分析自信度和个体决策以及决策调整行为的关系, 研究了个体自信度的交流对双人决策的影响。实验结果表明, 个体的自信度与选择的正确率高度正相关; 双人决策过程是个体根据对方的自信度和选择来不断调整自己的选择最终达成一致的过程, 并通过交互过程提高双人决策的正确率; 实验中双人决策的质量明显优于“自信度分享模型”和“更自信者主导决策模型”的预期结果, 表明群体决策不是通过分享自信度进行的贝叶斯优化整合过程, 也不是由更自信的个体完全主导的过程。     

Finding an unfamiliar face in a line-up: Viewing multiple images of the target is beneficial on target-present trials but costly on target-absent trials

When viewing unfamiliar faces, photographs of the same person often are perceived as belonging to different people and photographs of different people as belonging to the same person. Identity matching of unfamiliar faces is especially challenging when the photographs are of a person whose ethnicity differs from that of the observer. In contrast, matching is trivial when viewing familiar faces, regardless of race. Viewing multiple images of an own-race target identity improves accuracy on a line-up task when the target is known to be present (Dowsett et al., 2016, Q J Exp Psychol, 69, 1), suggesting that exposure to within-person variability in appearance is key to face learning. Across three experiments, we show that viewing multiple images of a target identity also improves accuracy for other-race faces on target-present trials. However, viewing multiple images decreases accuracy (i.e., increases false alarms) on target-absent trials for both own- and other-race faces. We discuss the implications of our findings for models of face recognition and for forensic settings.