Weapon focus refers to the decreased ability to give an accurate description of the perpetrator of a crime by an eyewitness because of attention to a weapon present during that crime. In the first experiment, subjects viewed a mock crime scene in which a weapon was either highly visible or mostly hidden from view. Subjects in the highly visible weapon group recalled significantly less feature information. Overall, memory accuracy scores were negatively correlated with self-reported arousal. The second series of experiments tested the weapon focus effect in a nonemotional situation in which the "time in view" of both the weapon and the individual's face were manipulated. A series of six slides were used in which either the weapon or the face was not in view for specific intervals within the sequence. The weapon focus effect was found to occur within a nonarousing, environmentally stark setting and was dependent on the percentage of time the weapon was visible.
141 students were made witnesses to a simulated crime and tested for immediate recall. Seven weeks later the witnesses were tested for recognition of the perpetrator in a 2 by 2 design which varied the degr~e of bias in instructions and in the layout of a six-person photospread. Main effects of bias were significant, with a strong interaction effect leading to the highest number of identifications of the perpetrator in the biased photo/biased instruction condition. Suggestivity in photospreads confounds the attribution of a positive identification to the witness's original perception. Would•be "criminals" have been running into psychology classes for decades , committing "crimes" and creating eyewitnesses, eyewitnesses who later prove to be unreliable and inaccurate. Hugo Munsterberg articulated this application of perception research in the salad days of functionalism in his book On the Witness Stand in 1908. But, as compelling as this demonstration of poor recall is, eyewitness testimony continues to be overrated in the courtroom and is the source of many convictions of innocent people (Borchard, 1961). It is a clash of common sense with "expert" judgments, confounded by the fact that the experts know the limits of human perception so well from laboratory research that little well-controlled empirical research on the witness to real or simulated crime exists in the literature. Buckhout, Alper, Chern, Silverberg, and Slomovits (I 974) did research on witnesses to staged crimes, testing both immediate recall and the witness's ability to recognize the culprit in a lineup. Witnesses tended to make a large number of recall errors, to underestimate the weight of the culprit, and to rely on probable height, weight, and age descriptions which resembled published norms rather than the suspect. The relatively few witnesses who were successful in identifying the culprit on a Jineu p (14%) had made Significantly fewer errors of commission during recalL The lineup, on videotape, was difficult and as unbiased as any test can be. In practice, police lineups frequently are hastily assembled and suffer from biasing factors and suggestion which can lead to mistaken identifications. As Levine and Tapp (1973) point out, " ... the criminal identification process necessarily involves an interaction between sensory and social inputs.
Forty students served as subjects in two groups of witnesses to which a surprise filmed crime was shown. All subjects gave free recall statements about the crime. Following either a directed thinking interval or a diversionary task to block rehearsal, witnesses gave two additional statements. Hypermnesia for hits and memory intrusions was observed, but d' scores did not show hypermnesia. The witnesses fell into two groups: "good" witnesses who had ascertained the correct schema and "bad" witnesses who were obviously on the wrong track. While most witnesses showed short-term hypermnesia for correct responses (hits), the good witnesses showed a significant increase in memory intrusions (false alarms) during the third recall trial (p < .01). Only with correct-schema witnesses did d' scores above the guessing level occur. Witnesses who had ascertained the correct schema but who were interrogated too often tended to make things up to suit the demands for detail, with a resultant increase in memory intrusions and lower d'.
A simulated crime was staged before 64 witnesses in order to generate eyewitness reports to be analyzed for accuracy. Forty-eight of the witnesses attempted to pick out the suspect from two videotaped lineups-one with and one without the suspect. Successful witnesses showed significantly fewer errors of commission than those who picked the suspect but impeached their identification with another choice. There were 13.5% positive identifications, 13.5% impeached, 40.3% mistaken identifications, and 19.2% nonidentifications. An analysis was made of the differences between successful and unsuccessful eyewitnesses.Staging a simulated "crime" in a classroom has become a standard teaching device in psychology and law school classes to demonstrate the fallability of the eyewitness (Buckhout et aI, 1973). The demonstration works so well that we rarely see hard data in print, a deficiency which has hampered the application of well established psychological theories of human perception in the handling of eyewitness testimony in the courts (Marshall, 1969). Observers of the legal setting, such as Wall (1966), effectively generalize the voluminous technical literature on memory and perception to the unique situation of crime, eyewitness reports, mug shot and lineup recognition, and court testimony, but stress the need for more specific research to demonstrate the role of classic research findings in a more relevant experimental design.In previous research, Buckhout et al (1973) demonstrated the usual poor recall of 141 witnesses and showed that relatively few witnesses were able to pick out a suspect they had previously seen from a photo spread, except under conditions in which the photo spread was biased. We endeavored in the present study to create a simulated crime, test for recall, and administer lineups to witnesses under controlled conditions. It was predicted that the number of witnesses making a correct positive identification would be small and that successful witnesses would be those who demonstrated better recall immediately after the "crime." METHOD SubjectsThe witnesses were 64 Brooklyn College undergraduate students enrolled in two separate introductory psychology courses; 52 witnesses were tested in the lineups. MaterialsA 40-item questionnaire and two videotaped lineups were *Reprints and additional data may be obtained from the authors at the Center for ResPonsive Psychology, Brooklyn College, Brooklyn, New York 11210.employed. The questionnaire measured the Ss' recall of the details of a purse-snatching incident, including a description of the suspect, the events, the amount of stress experienced, and the degree of confidence in their answers. From the list of correct answers, errors of omission, errors of commission, and estimates of incident time were computed.The lineups were videotaped for the exepriment. Each lineup included five male college students selected on the basis of their resemblance to the suspect. The blank lineup (B) did not show the suspect, but rather a close look-a-like. The valid lineu...
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.
hi@scite.ai
10624 S. Eastern Ave., Ste. A-614
Henderson, NV 89052, USA
Copyright © 2024 scite LLC. All rights reserved.
Made with 💙 for researchers
Part of the Research Solutions Family.