Ruth Horry

Functions for traditional and multilevel approaches to signal detection theory

In the present article, functions written in the freeware R are presented that calculate several measures from traditional signal detection theory for each individual in a sample, along with summary statistics for the sample. Bias-corrected and accelerated bootstrap confidence intervals are also produced. Arguments are made for using an alternative approach—multilevel generalized linear models—and a function is presented for it. These functions are part of the R package sdtalt, which is available on the Comprehensive R Archive Network. Recent data from memory recognition studies are used to illustrate these functions.

Predictors of eyewitness identification decisions from video lineups in England: a field study

Eyewitness identification decisions from 1,039 real lineups in England were analysed. Identification procedures have undergone dramatic change in the United Kingdom over recent years. Video lineups are now standard procedure, in which each lineup member is seen sequentially. The whole lineup is seen twice before the witness can make a decision, and the witness can request additional viewings of the lineup. A key aim of this paper was to investigate the association between repeated viewing and eyewitness decisions. Repeated viewing was strongly associated with increased filler identification rates, suggesting that witnesses who requested additional viewings were more willing to guess. In addition, several other factors were associated with lineup outcomes, including the age difference between the suspect and the witness, the type of crime committed, and delay. Overall, the suspect identification rate was 39%, the filler identification rate was 26% and the lineup rejection rate was 35%.

Recognition and context memory for faces from own and other ethnic groups: A remember-know investigation

People are more accurate at recognizing faces from their own ethnic group than at recognizing faces from other ethnic groups. This other-ethnicity effect (OEE) in recognition may be produced by a deficit in recollective memory for other-ethnicity faces. In a single study, White and Black participants saw White and Black faces presented within several different visual contexts. The participants were then given an old/new recognition task. Old responses were followed by remember-know-guess judgments and context judgments. Own-ethnicity faces were recognized more accurately, were given more remember responses, and produced more accurate context judgments than did other-ethnicity faces. These results are discussed in a dual-process framework, and implications for eyewitness memory are considered.

I know your face but not where I saw you: Context memory is impaired for other-race faces

People are more likely to falsely identify a face of another race than a face of their own race. When witnesses make identifications, they often need to remember where they have previously encountered a face. Failure to remember the context of an encounter can result in unconscious transference and lead to misidentifications. Forty-five White participants were shown White and Black faces, each presented on one of five backgrounds. The participants had to identify these faces in an old/new recognition test. If participants stated that they had seen a face, they had to identify the context in which the face had originally appeared. Participants made more context errors with Black faces than with White faces. This shows that the own-race bias extends to context memory.

Pictorial versus structural representations of ingroup and outgroup faces

Mental representations of unfamiliar faces may rely, to a greater or lesser extent, on pictorial cues that are closely linked to the specific image studied, and structural cues that allow for recognition across various transformations. Here, we test participants’ recognition of unfamiliar ingroup and outgroup faces, manipulating image transformation (pose) and exposure duration. The results showed that changes in pose reduced accuracy for outgroup faces but not for ingroup faces. Overall, accuracy increased as exposure duration increased. Accurate responses were made more rapidly than inaccurate responses for both ingroup and outgroup faces, showing that response latency is a useful postdictor of accuracy even for cross-ethnic identifications. The results are discussed in the framework of Bruce and Youngs (1986) model of face recognition.

The grain-size lineup: A test of a novel eyewitness identification procedure.

When making a memorial judgment, respondents can regulate their accuracy by adjusting the precision, or grain size, of their responses. In many circumstances, coarse-grained responses are less informative, but more likely to be accurate, than fine-grained responses. This study describes a novel eyewitness identification procedure, the grain-size lineup, in which participants eliminated any number of individuals from the lineup, creating a choice set of variable size. A decision was considered to be fine-grained if no more than 1 individual was left in the choice set or coarse-grained if more than 1 individual was left in the choice set. Participants (N = 384) watched 2 high-quality or low-quality videotaped mock crimes and then completed 4 standard simultaneous lineups or 4 grain-size lineups (2 target-present and 2 target-absent). There was some evidence of strategic regulation of grain size, as the most difficult lineup was associated with a greater proportion of coarse-grained responses than the other lineups. However, the grain-size lineup did not outperform the standard simultaneous lineup. Fine-grained suspect identifications were no more diagnostic than suspect identifications from standard lineups, whereas coarse-grained suspect identifications carried little probative value. Participants were generally reluctant to provide coarse-grained responses, which may have hampered the utility of the procedure. For a grain-size approach to be useful, participants may need to be trained or instructed to use the coarse-grained option effectively.

Confidence–accuracy resolution in the misinformation paradigm is influenced by the availability of source cues

After witnessing an event, people often report having seen details that were merely suggested to them. Evidence is mixed regarding how well participants can use confidence judgments to discriminate between their correct and misled memory reports. We tested the prediction that the confidence-accuracy relationship for misled details depends upon the availability of source cues at retrieval. In Experiment 1, participants (N=77) viewed a videotaped staged crime before reading a misleading narrative. After seven minutes or one week, the participants completed a cued recall test for the details of the original event. Prior to completing the test, all participants were warned that the narrative contained misleading details to encourage source monitoring. The results showed that the strength of the confidence-accuracy relationship declined significantly over the delay. We interpret our results in the source monitoring framework. After an extended delay, fewer diagnostic source details were available to participants, increasing reliance on retrieval fluency as a basis for memory and metamemory decisions. We tested this interpretation in a second experiment, in which participants (N=42) completed a source monitoring test instead of a cued recall test. We observed a large effect of retention interval on source monitoring, and no significant effect on item memory. This research emphasizes the importance of securing eyewitness statements as soon as possible after an event, when witnesses are most able to discriminate between information that was personally seen and information obtained from secondary sources.

Are Portrait Artists Superior Face Recognizers? Limited Impact of Adult Experience on Face Recognition Ability.

Across 2 studies, the authors asked whether extensive experience in portrait art is associated with face recognition ability. In Study 1, 64 students completed a standardized face recognition test before and after completing a year-long art course that included substantial portraiture training. They found no evidence of an improvement in face recognition after training over and above what would be expected by practice alone. In Study 2, the authors investigated the possibility that more extensive experience might be needed for such advantages to emerge, by testing a cohort of expert portrait artists (N = 28), all of whom had many years of experience. In addition to memory for faces, they also explored memory for abstract art and for words in a paired-associate recognition test. The expert portrait artists performed similarly to a large, normative comparison sample on memory for faces and words but showed a small advantage for abstract art. Taken together, the results converge with existing literature to suggest that there is relatively little plasticity in face recognition in adulthood, at which point our substantial everyday experience with faces may have pushed us to the limits of our capabilities.

Comparative Legal Psychology: Eyewitness Identification

We focus in this chapter on one important area of legal psychology: eyewitness identification. Following a brief overview of the broader field of legal psychology and the history of eyewitness identification research, we outline the major contributions that psychological science has made to our understanding of eyewitness identification, and review the procedures used to collect and interpret identification evidence in the United States and England and Wales. We conclude with some thoughts about how one might improve the treatment of identification evidence in legal settings in the future.