Biased lineup instructions and face identification from video images

Previous eyewitness memory research has shown that biased lineup instructions reduce identification accuracy, primarily by increasing false-positive identifications in target-absent lineups. Because some attempts at identification do not rely on a witness's memory of the perpetrator but instead involve matching photos to images on surveillance video, the authors investigated the effects of biased instructions on identification accuracy in a matching task. In Experiment 1, biased instructions did not affect the overall accuracy of participants who used video images as an identification aid, but nearly all correct decisions occurred with target-present photo spreads. Both biased and unbiased instructions resulted in high false-positive rates. In Experiment 2, which focused on video-photo matching accuracy with target-absent photo spreads, unbiased instructions led to more correct responses (i.e., fewer false positives). These findings suggest that investigators should not relax precautions against biased instructions when people attempt to match photos to an unfamiliar person recorded on video.     

Effects of postidentification feedback on eyewitness identification and nonidentification confidence

Two experiments investigated new dimensions of the effect of confirming feedback on eyewitness identification confidence using target-absent and target-present lineups and (previously unused) unbiased witness instructions (i.e., "offender not present" option highlighted). In Experiment 1, participants viewed a crime video and were later asked to try to identify the thief from an 8-person target-absent photo array. Feedback inflated witness confidence for both mistaken identifications and correct lineup rejections. With target-present lineups in Experiment 2, feedback inflated confidence for correct and mistaken identifications and lineup rejections. Although feedback had no influence on the confidence-accuracy correlation, it produced clear overconfidence. Confidence inflation varied with the confidence measure reference point (i.e., retrospective vs. current confidence) and identification response latency.     

The effect of biased lineup instructions on eyewitness identification confidence

Although it is well‐known that biased lineup instructions (i.e., those that do not inform witnesses the perpetrator may not be in the lineup) inflate false identifications, their effects on witness confidence are less well understood due to methodological limitations of past studies. We report two studies that use novel methodologies to obviate these limitations. Study 1 (N = 177) demonstrated that biased lineup instructions increased witnesses' average estimates of the likelihood that a lineup member is guilty. Study 2 (N = 137) introduces a novel debiasing paradigm that allows a parsing of choosers into those who made an identification only because of the biased instructions (induced choosers), and those who would have chosen despite the instructions (inherent choosers). Biased lineup instructions inflated confidence only among induced choosers, but not among inherent choosers. Contrary to legal reasoning, witness confidence is an insufficient metric to determine the suggestiveness of biased instructions.     

Do strict rules and moving images increase the reliability of sequential identification procedures?

Live identification procedures in England and Wales have been replaced by use of video, which provides a sequential presentation of facial images. Sequential presentation of photographs provides some protection to innocent suspects from mistaken identification when used with strict instructions designed to prevent relative judgements (Lindsay, Lea, & Fulford, 1991 ). However, the current procedure in England and Wales is incompatible with these strict instructions. The reported research investigated whether strict instructions would enhance the reliability of identification from video. The effect of using moving rather than still video clips was also investigated. Participants witnessed a live staged incident, and attempted to identify the culprit later from police video lineups, which were run double blind. Strict instructions produced a significantly lower rate of correct identifications in culprit present lineups, but did not significantly reduce the rate of mistaken identification in culprit absent lineups. Moving images yielded fewer mistaken identifications in culprit absent lineups. Copyright © 2006 John Wiley & Sons, Ltd.     

Influence of instructions on expectancy effects in Rorschach testing

Videotaped Rorschach instructions from a previous study in which two experimenters had biased subjects for high productivity and two experimenters had biased subjects for low productivity were played (audio or combined audio-visual channels) to matched subjects who then took the free association portion of the test. Biased results were obtained for one experimenter only. These did not interact with communication channels. It was concluded that experimenter-expectancy effects in Rorschach testing are not transmitted simply by exposing subjects to biased experimenters' instructions, but instead depend upon experimenter attributes which may or may not manifest themselves during the instructional phase. Limited evidence suggesting that expectancies are transmitted as effectively by audio as by a combination of audio and visual cues was obtained.     

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.     

Novel lineup methods for improving the performance of older eyewitnesses

Previous research shows that older witnesses demonstrate poorer lineup performance compared to younger adults. Two experiments are presented which investigated the effectiveness of pre‐identification procedures aimed to reduce false identifications made by older adults. In experiment one older adults' demonstrated poorer lineup performance compared to younger adults. However, older adults benefited from pre‐lineup questions and a practice lineup prior to viewing a target absent (TA) lineup. In a second experiment, participants in the practice lineup condition made significantly fewer false identifications and more correct rejections on two TA lineups compared to participants in the control condition. On both target present (TP) lineups there was no significant difference in lineup performance between the two conditions. The effect of pre‐identification procedure on self‐reported lineup decision strategy and memory for non‐biased lineup instructions are discussed, as well as future research directions. Copyright © 2009 John Wiley & Sons, Ltd.     

Why Do Eyewitnesses Make So Many Mistakes?

Previous research on eyewitness identification has demonstrated high rates of error. Subjects have frequently identified innocent targets as the "criminal" they had seen earlier (false identifications) or had falsely claimed that the criminal was not in the line-up (misses). The present study examines whether identification error rates are inflated by pressures in the typical experimental situation to "make a guess" regardless of one's confidence in the accuracy of the response. It was found that providing an explicit option for subjects to respond "don't know" significantly decreased false identifications and misses with no cost to the proportion of correct identifications. The addition of written and verbal instructions emphasizing the acceptability of the "don't know" option produced a marginally significant further decline in identification errors, again without cost to correct identifications. The discussion considered implications of the present results for experimental and actual police line-up procedures.     

Is the area measure a historical anomaly?

Green's well-known area theorem establishes an equivalence between the area under the yes-no ROC curve and the percent correct of an unbiased observer in a two-alternative forced-choice (2AFC) task with equivalent stimuli. In this article, we show that this conversion from yes-no detection data to hypothetical performance in a 2AFC task is unnecessary: The same yes-no detection data that are used to compute the area statistic can always be used to compute the percent correct of an unbiased observer in the yes-no detection task itself. We also show that the ROC curve may not be the ideal graphical device for many investigators. A more natural representation of the difficulty of a discrimination task is obtained by plotting the distribution of the posterior betting odds under equal base rates, which can be estimated from their distributions under unequal base rates. Finally, unlike the area measure and other traditional detection theory statistics, both the yes-no percent correct measure and the odds distributions generalize in an obvious and direct way to classification paradigms with more than two responses (e.g., identification).     

The existence and nature of racial bias in supervisory ratings

The purpose of this article is to facilitate a reconsideration of what the authors consider to be a mistaken belief among personnel psychologists--the belief that supervisory ratings of job performance are not biased on the basis of race. In this article, the authors reviewed the current literature, reexamined the data, and concluded that the research on which this mistaken belief is based clearly demonstrates that racial bias may indeed exist and is significant, both in statistical and practical terms.     

The Rorschach in the Context of Performance-Based Personality Assessment

Twenty-eight normal and 28 learning disabled children were given the Bender-Gestalt Test under instructional sets of low (standard) attention or high attention. Results failed to support the hypothesis that high attention instructions would lead to differential recall performance of the diagnostic groups. Normal children produced better designs during recall than learning disabled children regardless of instructional set. The evidence suggests that standard Bender instructions should not be altered in an attempt to elicit recall performance differences.     

Biased Lineup Instructions: Examining the Effect of Pressure on Children's and Adults' Eyewitness Identification Accuracy1

Two experiments were conducted to examine the effects of pressure on lineup decisions across child and adult eyewitnesses. In both experiments, participants were exposed to a target via videotape and then presented with a 6‐person, target‐absent lineup. Pressure was manipulated via lineup instructions (neutral vs. biased instructions). With neutral lineup instructions, children had a higher false‐positive rate than did adults. With biased lineup instructions, children's patterns of false‐positive responding paralleled that of adults; that is, an increase in false positives occurred. These results suggest that social pressure may not be the only factor driving children's higher false‐positive responding, compared to adults. Potential factors that may be affecting children's lineup decision processes are discussed.     

Detection and Selection Decisions with Conditional Feedback: Interaction of Task Uncertainty and Base Rate

Effects of task uncertainty on decision thresholds in a multiple cue decision task were examined under two types of feedback and three base rate conditions. In most such decision experiments, participants receive feedback after every trial (full feedback) with a single (usually .5) base rate. Our experiment explored conditional (decision‐contingent) feedback, in a task representing a detection problem (passenger screening) in which the decision maker receives no feedback unless the decision is positive (search the passenger). Increased uncertainty made all dependent measures worse. Task uncertainty had detrimental effects on both judgment and decision making, and interacted with effects of feedback and base rate. Decision performance was better with full feedback than with conditional feedback, but not by much. There may be no single unifying explanation for results of our base rate manipulation. Conditional feedback generally resulted in fewer positive decisions than full feedback, but not in the low (.1) base rate condition. Results provide partial support for constructivist encoding and for accuracy maximization with moderate and high base rates, but not with a low base rate. Our results indicated lower selection rates with conditional feedback compared with full feedback in moderate and high base rate conditions, and a more exploratory strategy with higher selection rates with conditional feedback compared with full feedback when base rate was low. Copyright © 2018 John Wiley & Sons, Ltd.     

A psychophysical theory of intensity proportions,joint presentations,and matches

Empirically testable assumptions relate 3 psychophysical primitives: presentations of pairs of physical intensities (e.g., pure tones of the same frequency and phase to the 2 ears or 2 successive tones to both ears); a respondent's ordering of such signal pairs by perceived intensity (e.g., loudness); and judgments about 2 pairs of stimuli being related as some proportion (numerical factor, as in magnitude production). Explicit behavioral assumptions lead to 2 families of psychophysical functions, one corresponding to unbiased joint presentations and the other to biased ones. Under an invariance assumption, the psychophysical functions in the unbiased case are approximate power functions, and those in the biased case are exact power functions. A number of testable predictions are made. The mathematics involved draws from publications in utility theory and mathematics but with a reinterpretation of the primitives.     

Transfer of location-specific control to untrained locations

Recent research highlights a seemingly flexible and automatic form of cognitive control that is triggered by potent contextual cues, as exemplified by the location-specific proportion congruence effect--reduced compatibility effects in locations associated with a high as compared to low likelihood of conflict. We investigated just how flexible location-specific control is by examining whether novel locations effectively cue control for congruency-unbiased stimuli. In two experiments, biased (mostly compatible or mostly incompatible) training stimuli appeared in distinct locations. During a final block, unbiased (50% compatible) stimuli appeared in novel untrained locations spatially linked to biased locations. The flanker compatibly effect was reduced for unbiased stimuli in novel locations linked to a mostly incompatible compared to a mostly compatible location, indicating transfer. Transfer was observed when stimuli appeared along a linear function (Experiment 1) or in rings of a bullseye (Experiment 2). The novel transfer effects imply that location-specific control is more flexible than previously reported and further counter the complex stimulus–response learning account of location-specific proportion congruence effects. We propose that the representation and retrieval of control settings in untrained locations may depend on environmental support and the presentation of stimuli in novel locations that fall within the same categories of space as trained locations.     

Extracting thresholds from noisy psychophysical data.

Psychophysical studies with infants or with patients often are unable to use pilot data, training, or large numbers of trials. To evaluate threshold estimates under these conditions, computer simulations of experiments with small numbers of trials were performed by using psychometric functions based on a model of two types of noise: stimulus-related noise (affecting slope) and extraneous noise (affecting upper asymptote). Threshold estimates were biased and imprecise when extraneous noise was high, as were the estimates of extraneous noise. Strategies were developed for rejecting data sets as too noisy for unbiased and precise threshold estimation; these strategies were most successful when extraneous noise was low for most of the data sets. An analysis of 1,026 data sets from visual function tests of infants and toddlers showed that extraneous noise is often considerable, that experimental paradigms can be developed that minimize extraneous noise, and that data analysis that does not consider the effects of extraneous noise may underestimate test-retest reliability and overestimate interocular differences.     

The role of compliance in hypnotic and nonhypnotic analgesia

We modified the standard procedure for administering finger pressure pain in order to assess the extent to which subjects biased their reports of pain reduction when exposed to compliance-inducing instructions. Experiment 1 used only highly hypnotizable subjects and found that compliance-induced reductions in reported pain were about half as large as the reported reductions produced by hypnotic analgesia suggestions. Experiment 2 used low as well as high hypnotizables and administered hypnotic or nonhypnotic analgesia instructions and compliance instructions to the same subjects on separate pain trials. Reported pain reductions produced by hypnotic and nonhypnotic analgesia suggestions and those produced by compliance instructions were substantially and significantly correlated. Furthermore, hypnotizability correlated with compliance-induced reported pain reductions as highly as it correlated with the reported reductions produced by hypnotic analgesia suggestions. Among high hypnotizables (but not low hypnotizables) compliant responding was predicted by social desirability. The findings of both studies indicate that compliance strongly influences the reports of hypnotic and nonhypnotic analgesia proffered by high hypnotizables. The role of compliance in the pain reductions reported by low hypnotizables is less clear-cut.     

A threat in the elder: the impact of task‐instructions,self‐efficacy and performance expectations on memory performance in the elderly

Widely used explicit memory tasks seem to overestimate age‐related differences in memory performance. Social and personal factors may buffer or undermine the effect of age on memory performance. In two studies, the performance of older adults was compared with the performance of younger adults. Tasks were presented either as memory tasks or non‐memory tasks. Older adults' performance on a memory task improved when the task‐instructions did not explicitly emphasize the memory component of the task. In the first study, results revealed that memory self‐efficacy beliefs play a moderator role on the impact of task‐instruction on memory performance, so that lower levels of memory self‐efficacy correlate with lower performance in the memory emphasizing task condition but not in the orientation emphasizing task condition. In a second study actual performance expectations were measured. For older participants only, expectations were sensitive to task‐instructions and mediated the relation between tasks‐instructions and performances. These findings suggest that observed age‐related differences in memory performance may be significantly exaggerated by the testing situation and by a low memory self‐efficacy and low memory performance expectancies prevalent among older adults. Copyright © 2005 John Wiley & Sons, Ltd.     

Extracting thresholds from noisy psychophysical data

Psychophysical studies with infants or with patients often are unable to use pilot data, training, or large numbers of trials. To evaluate threshold estimates under these conditions, computer simulations of experiments with small numbers of trials were performed by using psychometric functions based on a model of two types of noise:stimulus-related noise (affecting slope) andextraneous noise (affecting upper asymptote). Threshold estimates were biased and imprecise when extraneous noise was high, as were the estimates of extraneous noise. Strategies were developed for rejecting data sets as too noisy for unbiased and precise threshold estimation; these strategies were most successful when extraneous noise was low for most of the data sets. An analysis of 1,026 data sets from visual function tests of infants and toddlers showed that extraneous noise is often considerable, that experimental paradigms can be developed that minimize extraneous noise, and that data analysis that does not consider the effects of extraneous noise may underestimate test-retest reliability and overestimate interocular differences.     

The role of instruction preference in analogy learning: Brain activity and motor performance

This study examined the role of verbal instruction preference when learning motor skills by analogy. During skill learning, analogies are a useful tool for providing knowledge about how to move. It has been argued that analogy instructions reduce reliance on verbal information processes during motor planning, compared to traditional forms of instruction (i.e., explicit rules about how to move). This may be reflected by reduced verbal activity in the brain, measured by EEG alpha power at the temporal region, as well as reduced verbal-motor cross-communication (EEG T7-Fz coherence) during the preparation phase of a movement. Preference for using verbal or visual instructions is likely to influence the efficacy of analogy instructions. This study investigated whether preference for verbal instructions was related to a) changes in performance and b) changes in verbal-cognitive information processing during performance of an adapted basketball task after instruction by analogy. Basketball novices with a high preference for verbal instructions (n = 15) showed significantly decreased activation of verbal brain regions when they used the analogy (high-alpha power), but their performance remained stable. Novices with a low preference for verbal instructions (n = 13) did not show a significant decrease in activation of verbal regions, and their performance deteriorated significantly after introduction of the analogy instruction. It is likely that both cognitive and performance changes after analogy instruction depend on personal aspects of information processing, such as verbal preference.