James C Bartlett, PhD - US grants

The objective of this research is to understand deficits shown by elderly adults in visual long-term memory. Although these age- related deficits are generally small, they have important implications regarding eyewitness testimony by adults of different ages, and they reveal age differences in nonverbal memory mechanisms, which have been studied in other species and have been linked to specific brain structures. The focus of this project is memory for faces, an unusually pure case of nonverbal memory that is both well practiced and ecologically significant across all age groups and generational cohorts, and that is known to show deficits among elderly adults. A first specific aim of the project is to examine the age deficits in memory for faces from an information-processing perspective, comparing young and elderly subjects in the encoding, retention, and retrieval of faces, and clarifying some ambiguities in the prior research literature. A second specific aim is to examine the practical and theoretical implications of age-related increases in false recognitions of new faces. A third specific aim is to explore some hypotheses derived from research on amnesia for age-related deficits in face memory. These hypotheses include pronounced deficits on the part of older individuals in memory for the presentation context of faces, and age-invariance in some types of memory for emotional reactions to faces.

Music provides an ideal domain to test the generality of widely accepted principles of cognitive aging, and to explore how the factor of expertise can moderate age differences in cognition. We propose 12 studies that will examine how age and musical training are related to perception and memory of music. Our first specific aim is to examine the hypothesis that one's knowledge of scales and other musical structures does not decline and may even increase in old age, despite age-related deficits in on-line analysis of music. We will test this hypothesis in five different studies that assess the effects of tonal structure on (a) the perceived similarity of single tones and tone sequences, and (b) accuracy of short-term memory for melodies, by young and elderly adults with different levels of musical training. Our second specific aim is to determine if the principle of generalized slowing in old age extends to musical materials, and, if so, if this slowing is reduced among expert musicians. Avoiding the use of response-time measures, our approach will be to determine the fastest and slowest tempos at which melodies can be recognized, and to assess the accuracy of pitch-encoding in melodies presented at varying tempos. Our third specific aim is to examine age differences in several aspects of memory for melodies, including the maintenance of melodies in short-term or "working" memory, recognition of melodies, recollection of context of prior encounter with melodies, and reality-monitoring judgments of whether familiar melodies have been recently perceived or simply imagined. We will also test the principle that there are minimal age differences in "priming" tasks revealing memory for stimuli even in the absence of conscious recollection. Of interest throughout the proposed research will be possible interactions of age with expertise, and, in other cases, dissociative effects of these variables.

9515231 Bartlett As the elderly are disproportionately likely to be the victims of certain crimes, the legal system will have to rely more and more on elderly witnesses. There has as yet been little research with respect to the performance of the elderly in this role, and existing research has yielded few solid conclusions. The objective of this research is to examine systematically the performance of the elderly as witnesses. It will (1) investigate the conditions under which the elderly are likely to perform poorly and well; (2) test the rules and procedures within the legal system that may be manipulated to enhance the quality of information obtained from elderly witnesses, (3) evaluate the utility of psychometric measures for distinguishing accurate from inaccurate witnesses, and (4) explore the potential for using the relatively unimpaired "implicit memory" of the elderly to assess the accuracy of their memories or to elicit additional information. Experiments involving simulated crimes and other events will be used to compare the performance of young and elderly subjects in order to answer basic questions about age differences. Research has documented that jurors, attorneys, and others within the legal system tend to assume that the elderly witnesses have difficulty in performing accurately. Our goal is to obtain solid information as to what, if anything, should be considered when dealing with seniors who are witnesses to crimes. %%%% As the elderly are disproportionately likely to be the victims of certain crimes, the legal system will have to rely more and more on elderly witnesses. There has as yet been little research with respect to the performance of the elderly in this role, and existing research has yielded few solid conclusions. The objective of this research is to examine systematically the performance of the elderly as witnesses. It will (1) investigate the conditions under which the elderly are likely to perform poorly and well; (2) test the rules and procedures within th e legal system that may be manipulated to enhance the quality of information obtained from elderly witnesses, (3) evaluate the utility of psychometric measures for distinguishing accurate from inaccurate witnesses, and (4) explore the potential for using the relatively unimpaired "implicit memory" of the elderly to assess the accuracy of their memories or to elicit additional information. Experiments involving simulated crimes and other events will be used to compare the performance of young and elderly subjects in order to answer basic questions about age differences. Research has documented that jurors, attorneys, and others within the legal system tend to assume that the elderly witnesses have difficulty in performing accurately. Our goal is to obtain solid information as to what, if anything, should be considered when dealing with seniors who are witnesses to crimes. ****

Ten experiments are conducted that compare eyewitness testimony in young adulthood and old age, exploring the person- and task- characteristics that might affect performance of the older eyewitness. Parts I and 11 further examine findings that false identifications in lineups are increased in old age, and that exposing the eyewitness to a post- event narrative is a factor involved in producing this effect. Modifying the procedures of earlier researchers Experiment 1 addresses the possibility that verbal recall of the crime-event prior to the lineup task (a) increases rates of false choosing in elderly (and/or young) adults, and (b) magnifies the deleterious effects of previously exposed narrative. Extending recent work on the `verbal overshadowing effect`, Experiment 2 tests the possibility that high choosing rates among older eyewitnesses reflect a non-optimal mode of face processing (verbal encoding of facial features) that is reinforced by verbal recall of facial information, and can be prevented in at least some eyewitnesses by eliminating the task of verbal recall prior to a lineup. Considering prior evidence that context reinstatement prior to a lineup can either help or hurt lineup identification, in Experiment 3 participants are exposed to a highly salient event involving physical interaction with an actor, comparing the cognitive interview to both a standard structured interview and a control (non-structured) interview to assess its effects on both (a) completeness and accuracy of recall as well as (b) accuracy and choosing in lineups. Experiment 4_ is based on the `mugshot effect` of increased rates of false identifications of faces not seen at a crime but viewed in another context The hypothesis tested that the mugshot effect will be even stronger among elderly persons but that it can be reduced by instructions and testing materials that discourage eyewitnesses from basing their judgments on perceived familiarity. Experiment 5 addresses a similar hypothesis as applied to the `bystander effect` of increased rates of false identification due to the presence in a lineup of a bystander (instead of the perpetrator) in a prior a crime event). Although the bystander effect is known to be ephemeral it was found that older eyewitnesses show the effect more strongly. A replication of this potentially important finding is undertaken, examining whether older persons can overcome the effect if the instructions and materials discourage familiarity-based responding. The next three proposed studies (Part lil) depart from prior work in exploring how socio-cogntive variables of stereotypes and person-schemata affect recall of crime-events and lineup identification. Experiments 6 and 7 assess this possibility. In addition, both studies examine how positive and negative biographical information about the perpetrator and/or victim can moderate the effect. The key question is whether the tendency to pick a guilty-looking suspect reflects a simple response bias or a more complex appraisal of the appearance of the suspect relative to the story being told about the crime. Experiment 8 returns to the `bystander effect,` assessing whether the selection of a bystander is linked not merely to perceived familiarity, but to the congruence between the bystander's appearance (guilty vs. innocent) and the nature of the event (a crime vs. a charitable act). The final two experiments (Part IV) reflect a change of focus from lineup identification to verbal recall about a crime. Experiments 9 and 10 both address the well known observation that repeated questioning of witnesses can increase the frequency or erroneous intrusions and confabulations, particularly if the questioning entails presentation of misleading information. First, Experiment 9 examines the conditions under which repeated questioning can have positive, `hypermnesia` effects which heretofore have not been examined in old age. Experiment 10 examine the phenomenon that misleading information implanted in the questioning will be incorporated in later recalls, testing a prediction based on cognitive-aging theories that this effect will be increased in old age. An important component of all the experiments will be the inclusion of neuropsychological and/or personality tests to reveal individual-difference factors that might moderate age differences.