Male volunteers (N = 120) in small groups of 5 to 10 watched a staged theft involving live actors. Some (n = 47) were under the influence of alcohol (average blood alcohol level of .10) at the time. Some subjects (n = 58) were interviewed immediately after the event, and all were interviewed 1 week later. The delayed interview included the presentation of a photospread that either did or did not contain the picture of the "thief." Alcohol suppressed the amount recalled during the immediate interview and both the amount and accuracy of recall after the 1-week delay. Alcohol had no influence on the ability of witnesses to recognize the thief's picture. When the thief's picture was not present in the photospread, however, alcohol increased the rate of false identifications. An immediate interview substantially improved the amount of information subjects were able to recall 1 week later.

No abstract

Unconscious transference refers to an eyewitness's misidentification of an innocent bystander for a criminal perpetrator because of the witness's exposure to the bystander in another context. In a series of five field studies involving 330 retail store clerks and 340 students, five retention intervals from 2 hours to 2 weeks, seven bystander-perpetrator intervals from 2 minutes to 2 weeks, three line-up types, two levels of line-up similarity, four different bystanders and four different targets, with one exception no evidence was obtained that could be interpreted to demonstrate the phenomenon of unconscious transference.That is, theresultsrepeatedly failed to reveal more misidentifications of an innocent bystander by witnesses who had been previously exposed to the bystander than by control eyewitnesses who had not. To the contrary, the prior observation of the bystander often served to reduce the frequency of misidentification. In the final experiment the kind of misidentification error referred to as unconscious transference did occur, but only within a particular combination of bystander-perpetrator similarity and line-up construction: a combination that, in conjunction with the kind of event used, seems unlikely in real-world settings. Nevertheless, the inclusion of a familiar face in the line-ups often altered witnesses' choices in such a way that choosing someone was more likely when the lineup included a familiar face than when it did not. Finally, in contrast to the current explanations of unconscious transference, it is argued that it may not be a sense of familiarity with the bystander that is the basis of misidentifications; rather, it may also include incorrect inferences about the likelihood that the bystander might be the perpetrator.

One week after committing a simulated robbery while intoxicated or sober, each of 142 subjects recalled the event within a "cognitive interview." In an initial exploratory experiment, alcohol consumption reduced the accuracy of recall of a variety of types of information, in particular, information about persons. In the second experiment, person identification suffered following the consumption of alcohol, but only when arousal was low. Higher levels of arousal appeared instead to minimize the negative impact of alcohol upon encoding and recall. Second, whereas the recollections by subjects of what they saw during the crime were not impaired by alcohol consumption, their recollections of what they did were impaired. Both experiments examined the effects of arousal upon the subjects' recalls, and Experiment 2 tested the hypothesis that increased arousal serves to reduce attention to peripheral sources of information. This hypothesis was supported because the identification of persons central to the crime benefited from increased arousal but the identification of persons peripheral to the crime did not. A similar hypothesis about the effects of alcohol received only mixed support because the subjects' behaviors reflected "alcohol myopia" but their identifications of target persons did not. Finally, manipulations at the time of retrieval of the subjects' beliefs about how much alcohol had been consumed also altered accuracy of recall.

No abstract