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. (PsycINFO Database Record (c) 2010 APA, all rights reserved)     

Eyewitness identification accuracy, confidence, and decision times in simultaneous and sequential lineups.

Eyewitness identification accuracy was investigated in simultaneous and sequential lineups. 72 Ss watched a film of a robbery in a public park under incidental learning conditions and returned to the laboratory the following day to answer questions about the film. Sequential lineup procedures led to significantly fewer false identifications than the simultaneous lineup mode, with comparable performance in detecting the perpetrator in target-present conditions. Alternative methods for analyzing confidence and decision times in sequential lineups are presented which allow for more fine-grained analyses of the relationships between accuracy, confidence, and decision times both between and within Ss. Distinguishing between choosers and nonchoosers, these analyses show the predictive utility of decision times and confidence as assessment variables. (PsycINFO Database Record (c) 2010 APA, all rights reserved)     

Forensically relevant moderators of the relation between eyewitness identification accuracy and confidence.

Meta-analyses of 9 studies were conducted that all assess the association between pre-lineup confidence and identification accuracy; the association between post-lineup confidence produces a stronger correlation with identification accuracy than does pre-lineup confidence. In 5 of these studies the difference between correlations is statistically significant. Furthermore, correlations between pre-lineup confidence and accuracy were trivial in magnitude, ranging from .00 to .20. The conclusion is drawn that a witness's confidence in his or her ability to make an identification should not be given much weight in determining whether or not to have the witness attempt an identification. Also, pre-lineup confidence should not be used to evaluate the accuracy of an identification. (PsycINFO Database Record (c) 2010 APA, all rights reserved)     

Fast and confident: Postdicting eyewitness identification accuracy in a field study.

The combined postdictive value of postdecision confidence, decision time, and Remember-Know-Familiar (RKF) judgments as markers of identification accuracy was evaluated with 10 targets and 720 participants. In a pedestrian area, passers-by were asked for directions. Identifications were made from target-absent or target-present lineups. Fast (optimum time boundary at 6 seconds) and confident (optimum confidence boundary at 90%) witnesses were highly accurate, slow and nonconfident witnesses highly inaccurate. Although this combination of postdictors was clearly superior to using either postdictor by itself these combinations refer only to a subsample of choosers. Know answers were associated with higher identification performance than Familiar answers, with no difference between Remember and Know answers. The results of participants' post hoc decision time estimates paralleled those with measured decision times. To explore decision strategies of nonchoosers, three subgroups were formed according to their reasons given for rejecting the lineup. Nonchoosers indicating that the target had simply been absent made faster and more confident decisions than nonchoosers stating lack of confidence or lack of memory. There were no significant differences with regard to identification performance across nonchooser groups. (PsycINFO Database Record (c) 2010 APA, all rights reserved)     

Postdictors of eyewitness errors: Can false identifications be diagnosed?

Eyewitness researchers have "postdicted" identification accuracy using witness confidence (S. L. Sporer, S. Penrod, D. Read, & B. Cutler, 1995), response latency (S. L. Sporer, 1993, 1994), and endorsement of statements consistent with using relative versus absolute judgment strategies (D. Dunning & L. B. Stem, 1994; R. C. L. Lindsay & K. Bellinger, 1999). All of these measures were collected from 321 introductory psychology students who had viewed a staged crime and completed a lineup identification task. Some participants received feedback after identification (G. L. Wells & A. L. Bradfield, 1998). Lineup fairness was also used as a postdictor of eyewitness accuracy. Discriminant function analysis indicated that 75.2% of choosers and 63.0% of nonchoosers were correctly classified. Decision time and lineup fairness were the best postdictors of accuracy. The implications for postdicting real eyewitness decisions are discussed. (PsycINFO Database Record (c) 2010 APA, all rights reserved)     

Recall memory, recognition memory, and the eyewitness confidence–accuracy correlation.

Research on the eyewitness confidence-accuracy correlation assesses the degree of insight that eyewitnesses have into the accuracy of their memories. Recently, researchers have begun to consider some of the variables that may facilitate or hinder such insight. In the present study with 205 college undergraduates, a previously unexamined influence on this correlation in the context of memory for details of a simulated crime was investigated. The authors hypothesized and found that recall memory conditions (in which no alternative answers were provided) were characterized by a higher eyewitness confidence–accuracy correlation than recognition memory conditions (in which alternative answers were provided). These findings are explained as a function of the availability of an ease-of-retrieval cue in recall memory. (PsycINFO Database Record (c) 2010 APA, all rights reserved)     

Eyewitness identification: Retrospective self-awareness and the accuracy–confidence correlation.

Conducted 4 experiments to test the hypothesis that retrospective self-awareness (RSA) would increase the correlation between eyewitness identification accuracy and confidence. Undergraduate students (48, 33, 75, and 72 in Exps I–IV, respectively) participated as Ss. In all studies, Ss watched a staged crime; immediately afterwards, they were asked to identify the culprit from a photospread and to indicate their confidence in that judgment. In an RSA condition, Ss also viewed a videotape of their performance before rating their confidence. Collectively, the results show an average correlation of .04 in the control groups and .48 in the RSA condition. The data tentatively support a self-perception hypothesis that this manipulation is effective because it alerts Ss to valid but previously unobserved aspects of their own overt behavior (e.g., response latency). Findings are discussed for their theoretical and forensic implications. (50 ref) (PsycINFO Database Record (c) 2010 APA, all rights reserved)     

Predicting field of job entry from expressed vocational choice and certainty level.

Examined the predictability of actual jobs from expressed vocational choices and confidence in those choices. Ss were 1,994 members of a nationwide study of high school student career development who were working more than half time 3 yrs later. Both choices and occupations were categorized into Holland-type job clusters; analyses involved the use of weighted hit rates and coefficient kappa. Expressed vocational choices correctly predicted actual occupations 2 yrs after high school for 38% of the sample. Hit rates for males and females were 40 and 35%, respectively. When certainty level was added to expressed choice, hit rates were 43% for sure choosers, 38% for fairly sure choosers, and 28% for choosers who were not sure. Each choice level was significantly different from the others. (18 ref) (PsycINFO Database Record (c) 2010 APA, all rights reserved)     

Prediction and perception of social motives.

In 2 experiments 70 male and female and 89 female undergraduates were required to observe and predict the behavior of a hypothetical "chooser" who made choices for him- or herself and for a hypothetical other in a series of decomposed games. The preference for outcomes, or social motivational orientation, of the chooser was preprogramed and varied across conditions. Ss were more readily able to detect the outcome preferences of choosers who made choices according to individualistic or competitive choice rules than of choosers who behaved in a prosocial or negatively self-interested manner. Furthermore, the prediction data from Exp II reveal that Ss tended to perceive choosers' own gain as an important component of most of the choosers' secondary motivation. Evidence from Ss' ratings of the choosers' personality attributes and estimates of the relative weights the choosers attached to their own and the other's gain (Exp II) indicated that Ss formed distinctive impressions of the choosers despite differences in predictive accuracy across conditions. Exp III with 64 undergraduates was performed to investigate the relationship between predictive accuracy and the mathematical complexity of the choosers' various choice rules; no evidence was found that mathematical complexity influenced Ss' performance on the prediction task. (24 ref) (PsycINFO Database Record (c) 2010 APA, all rights reserved)     

Correlation of eyewitness accuracy and confidence: Optimality hypothesis revisited.

Inasmuch as a completely satisfactory estimate of effect size for the eyewitness accuracy-confidence relation does not exist, we conducted a meta-analysis of 35 staged-event studies. Estimated r?=?.25 (d?=?.52), with a 95% confidence interval of .08 to .42. Sampling error accounted for 52% of the variation in r, leaving room for measurement error and possibly moderator variables to account for the remaining variation. Further analysis identified duration of target face exposure as a moderator variable, providing support for Deffenbacher's (1980) optimality hypothesis. When corrected for the attenuating effect of sampling error in the accuracy-confidence correlations, the correlation of exposure duration and the accuracy-confidence correlation was .51: Longer exposures allowed for greater predictability of accuracy from confidence. Even through correlation for unreliability in the confidence measure produces a higher estimate of the population correlation of accuracy and confidence, .34, one must be cautious in assessing the utility of confidence for predicting accuracy in actual cases. (PsycINFO Database Record (c) 2010 APA, all rights reserved)     

Reaction time and assessments of cognitive effort as predictors of eyewitness memory accuracy and confidence.

The authors investigate reaction time (RT), subjective assessments of memory processing, and confidence as predictors of memory for the details of a crime. The authors also examine the mediation of a previously identified difference between recognition tasks and recall tasks in the correlation between confidence and accuracy. College undergraduates (n?=?111) answered either recognition or recall questions. RT and subjective assessments of cognitive effort were both negatively related to confidence and accuracy. Subjective assessments, however, were superior predictors of confidence, whereas RT was a unique predictor of accuracy. The RT–confidence and RT–accuracy correlations were stronger under recall conditions than under recognition conditions. Multiple regression results suggested a possible explanation for the superior insight of recall participants into memory accuracy. (PsycINFO Database Record (c) 2010 APA, all rights reserved)     

Victimization as a determinant of eyewitness accuracy.

Eyewitness identification research frequently uses a simulated crime paradigm, yet the witnesses in these studies are seldom themselves the victims of the crime. To the degree a witness is more personally involved (as a victim), eyewitness identification accuracy and confidence might be modified. 66 undergraduates were randomly assigned to 3 eyewitness conditions: no-theft control, impersonal calculator theft, or personalized watch theft. Whereas the victims in the watch-theft condition were more accurate than Ss in the other 2 treatments, they were not significantly more accurate than witnesses to the less personal calculator theft. Eyewitnesses' accuracy was unrelated to their confidence in their identification. (12 ref) (PsycINFO Database Record (c) 2010 APA, all rights reserved)     

Recognition instructions and recognition practice can alter the confidence–response time relationship.

Three face-recognition experiments examined how instructions for a recognition test (e.g., emphasize speed or emphasize accuracy) can impact the confidence–response time relationship for episodic memory reports. In all 3 experiments, the confidence–response time correlation was smaller when participants were told to speed up their responding rate, which suggests that participants in these conditions relied less on the artificially compressed response times in forming their confidence judgments than they would under "normal" circumstances. Also, recognition practice before the final memory test eliminated the effect of the recognition instruction manipulation. These results support J. S. Shaw's (1996) suggestion that witnesses rely in part on the fluency of their memory reports when generating confidence judgments, and these findings have important implications for understanding the relationships among witness confidence, accuracy, and response time. (PsycINFO Database Record (c) 2010 APA, all rights reserved)     

The Psychological Pleasure and Pain of Choosing: When People Prefer Choosing at the Cost of Subsequent Outcome Satisfaction.

This empirical investigation tested the hypothesis that the benefits of personal choosing are restricted to choices made from among attractive alternatives. Findings from vignette and laboratory studies show that contrary to people's self-predictions prior to actually choosing, choosers only proved more satisfied than nonchoosers when selecting from among more preferred alternatives. When selecting from among less preferred alternatives, nonchoosers proved more satisfied with the decision outcome than choosers. Subsequent analyses revealed that differences in outcome satisfaction between choosers and nonchoosers emerge even before the decision outcome is experienced and that interventions during the decision-making process can serve to attenuate these differences. Theoretical and practical implications are discussed. (PsycINFO Database Record (c) 2010 APA, all rights reserved)     

Changing photos of faces: Effects of exposure duration and photo similarity on recognition and the accuracy-confidence relationship.

With high school students' photo pairs taken 2 yrs apart, it was demonstrated that the effects of exposure duration on face recognition depend upon the similarity between the viewing and test photographs. When similarity between the 2 photos of a face was high, the expected positive relationship between exposure time and recognition accuracy was obtained. However, when the 2 photographs were of low or medium similarity to each other, increased study time had either no effect, increased performance, or in some cases significantly reduced recognition performance. One implication of these data is that in some circumstances gains in specific memory acquired through increased exposure time may serve to reduce rather than enhance recognition performance. Analyses of the relationship between Ss' accuracy and confidence judgments revealed that as both exposure duration and photo similarity increased, the magnitude of the accuracy–confidence correlation increased. Increasing Ss' prospective awareness of the changes that may be made to the photos increased the magnitude of the accuracy–confidence correlation but without corresponding increases in recognition performance. (PsycINFO Database Record (c) 2010 APA, all rights reserved)     

Eyewitness memory and aging: Predictors of accuracy in recall and person recognition.

Examined adult age differences in identification accuracy, confidence, and verbal recall of crime films. A total of 120 Ss in 3 adult age groups watched 2 crime films that contained characters varying in role and visibility. Results suggested a modest negative relationship between age and identification accuracy but no age differences in recall of main points or details. Education and self-reported health positively predicted verbal recall, but higher education was associated with lower identification accuracy. No age differences were found in confidence or in the relationship between confidence and accuracy. (PsycINFO Database Record (c) 2010 APA, all rights reserved)     

Evidence of limited capacity and noise reduction with single-element displays in the location-cuing paradigm.

Precuing a location facilitates accuracy of identification of a target at that location and reduces noise from other nontarget characters in a multicharacter field. In 5 experiments, evidence for facilitation included higher accuracy with long precue–target intervals than with short intervals and higher accuracy when a precue indicated the correct location than when it indicated the wrong location. These results were found for each target–mask condition used (1 target with 1 mask, 1 target with 4 masks, or 1 target and 3 nontargets with 4 masks) in experienced and inexperienced observers. Evidence for noise reduction was found because accuracy was higher in the 1 target–1 mask condition than in the other conditions on correctly cued trials with short cue–target intervals and on incorrectly cued trials. Data are related to methodological factors that are important to obtaining these effects and to capacity and noise reduction models. (PsycINFO Database Record (c) 2010 APA, all rights reserved)