February 2009 Archives

Sporer, S. L., Penrod, S., Read, D., & Cutler, B. (1995). Choosing, confidence, and accuracy: A meta-analysis of the confidence-accuracy relation in eyewitness identification studies. Psychological Bulletin, 118, 315-327.

            Sporer et al., (1995) present an in depth meta-analysis conducted using results and data from previous studies examining the relation between confidence-accuracy.  Thirty studies were used in the meta-analysis which used some kind of mock crime event in their methodology, used both target present and target absent lineups, had collected both confidence and accuracy measures from participants, and still had raw data to use for further analyses. The main addition to the existing research on the CA relation in the meta-analysis is the examination of the moderator variable of whether or not people identify someone out of a lineup or do not make an identification.

Clearly the processes would be different for those who choose compared to those who do not choose someone from a lineup. The authors cite evidence from Bem's (1972) self-perception theory regarding the notion that the act of making some kind of decision, rather than indecisiveness, should result in greater confidence in the ability to have made a worthwhile decision (i.e., in the case of eyewitness identification, an accurate one). Thus, stemming from the notion of these inherent differences between choosers and non-choosers, Sporer et al., (1995) ran separate analyses for choosers (those who chose either a correct or incorrect person from the lineup) versus non-choosers (those who rejected the lineup). The effect sizes for choosers were much higher than for non-choosers, even for weighted effect size estimates. The real interesting thing is that the effect sizes, weighted, were higher than the overall effect size with both choosers and non-choosers included. Additionally, those who chose were much more confident compared to those who did not make an identification from the lineup. This indicates the strong role of the moderator variable of being a chooser or a non-chooser when faced with making a lineup identification decision.

Being a meta-analysis, the article provides a variety of information from the various studies examined, including whether or not the mock crime event was live or recorded to video, gender of the target, scale range used for confidence statements, proportion of correct identifications, and correct decisions broken down for choosers and non-choosers. This is an excellent review of the CA relation by some of the biggest and well known researchers in the field of eyewitness identification. If you only had time to read one article about the relation between eyewitness confidence and accuracy, I would recommend you read Sporer et al., (1995).

--By DJP 

Luus, C. A. E., & Wells, G. L. (1994). The malleability of eyewitness confidence: Co-witness and perseverance effects. Journal of Applied Psychology, 79, 714-723.

            Luus and Wells (1994) provide an excellent study regarding the extent to which confidence of eyewitnesses can change and vary based on information they receive from various sources, especially regarding information about others decision making processes for lineups. Their study incorporates a very social psychological design with a traditional eyewitness identification approach in which participants are shown a crime video, asked to identify the person they saw commit the crime from a lineup, and then indicate how confident they are in their decision. The twist with this study is they then told participants, assigned to various conditions, information about other people's decisions on the lineup (i.e., same, different, etc.). Thus, participants receiving information indicating that another person picked the same face out of a lineup that they did, exhibited an increase in confidence. Those hearing from the confederates of the experiment regarding people who disagreed with their decision or rejected the lineup, indicated much lower confidence statements relative to other conditions of the experiment in which agreement occurred.

            In their next experiment, the researchers showed the videotaped lineup administrations complete with the decision, confidence, and induction of false information recorded from Experiment 1, to a new sample of participants to rate how accurate they perceived the participants in the various conditions in Experiment 1 to be regarding the decision made. They found that when viewing a recording of the inflated confidence conditions, participants predicted higher accuracy to result based on their confidence statements, compared to recordings of the confidence deflated conditions.

            This study provides a nice review of the malleability of confidence judgments and also tells a cautionary tale of how law enforcement officials and others who may influence eyewitness confidence with information or positive feedback regarding decisions made, should consider the potential harmful impact of the information they provide to witnesses.

--By DJP

Deffenbacher, K. A. (1980). Accuracy and confidence: Can we infer anything about their relationship? Law and Human Behavior, 4, 243-260.

            Deffenbacher (1980) provides a review of many of the earlier studies conducted on the relationship between eyewitness confidence and accuracy. As can be seen, the relationship between confidence and accuracy is not a reliable one and a variety of methods have been used to conduct research ranging from finding either significant positive correlations or no correlation at all. In addition to providing somewhat of a review of research literature examining this important topic, Deffenbacher provides a theory regarding the disparity in results regarding findings of various correlations between confidence and accuracy with reliability coefficients of an unimpressive magnitude. He proposes an optimal processing hypothesis attempting to explain why certain studies find high positive correlations between confidence and accuracy and why others find near zero or negative correlations.

            Under optimal conditions, the relationship between confidence and accuracy is thought to be high, yet under less than optimal conditions the relationship will decrease to chance levels or produce evidence of a negative relationship between the two variables. There is evidence to support the optimality hypothesis, with results indicating significant positive correlations based on methodologies in which the procedural phases of the experiment (i.e., encoding, retrieval) were carried out under conditions sufficient for optimal processing of information. Additionally, studies in which participants encoded and attempted to retrieve information regarding an event were conducted within low or poor processing conditions (i.e., noise, outdoors, low luminance) no significant positive correlations, and in some instances negative correlations, were found.

On the other hand, certain studies have found results which run counter to the proposals stemming from the optimality hypothesis. For example, certain studies using procedures conducive to optimal processing of information have found non-significant or negative correlations for the confidence and accuracy relationship. Thus, we researchers and those in the area of law enforcement need to be especially careful regarding attempts to predict accuracy from eyewitness confidence statements. Jurors may also be affected, because they are likely to be influenced by confidence statements (especially when high) made by witnesses, and infer accuracy from them (Wells, Lindsay, & Ferguson, 1979).  It is important that all those using or examining confidence statements made by witnesses are extremely careful when interpreting or estimating how well they provide information relate to accuracy of a identification.

--By DJP

Summary to be provided by Ian

Eyewitness Seminar

1.      Cara Wilkerson:  Cross Racial Identification

2.      Ian: Wrongful Convictions

3.      Dwight: Confidence Accuracy Relationship

4.      Matt Sullivan:  Interviewing

5.      Elizabeth Mendez:  Mug Shot Books

6.      Staci Devera:  Child/Elderly Eyewitness

7.      Sean Babinat:  Verbal Overshadowing

8.      Alex Esquirel:  Post Event Information

9.      Heather Caspers:  Ecological Validity

10.  Amanda Doloson:  Juror Knowledge

11.  Kelli Case:  Stress/Arousal

12.  Steve Berg:  Lineup Construction

13.  Laura Meyer:  Weapon Focus

Welcome to my home page, which is also a resource for those interested in eyewitness memory issues.
Scroll down first, then use your cursor to find links and download anything that interests you.

http://www.uni.edu/~maclino/ew/Unconscious_Transference.ppt

by DP

http://www.uni.edu/~maclino/ew/Lineup_Administration.ppt

By HC

Ross, D. F., Ceci, S. J., Dunning, D., & Toglia, M. P. (1994). Unconscious transference and mistaken identity: When a witness misidentifies a familiar but innocent person. Journal of Applied Psychology, 79, 918-930.

Ross et al., (1994) provide excellent background information on unconscious transference, including various theoretical perspectives that have been proposed as explanations for the phenomenon. The theoretical position that Ross and colleagues espouse relates to the notion of conscious inference, in which a witness to a crime actually thinks that the perpetrator and an innocent familiar person are one in the same. The authors propose, contrary to others (see Read et al., (1990)), that conscious inference can occur at encoding such that the perpetrator and the innocent person are inferred to be one and the same from the initial presentation of both faces.

Hypotheses include the pronounced occurrence of a transference effect depending on various lineups including the bystander present, the perpetrator present, or both present. Thus, when viewing a bystander present lineup, participants in the transference group compared to the control group were expected to misidentify the innocent bystander more often. In the perpetrator present lineup, participants, regardless of condition, should be equally accurate in correctly identifying the perpetrator from the lineup. For lineups including both the bystander and the perpetrator, participants in the control group should correctly choose the perpetrator to a greater extent than those in the transference group due to the presence of the bystander in the lineup. Transference participants should still choose the perpetrator to a greater extent compared to choosing the bystander, but some will still erroneously choose the bystander thinking he is the same person as the perpetrator.

Experiment 1 found support for the aforementioned hypotheses. In Experiment 2 researchers informed the participants in the transference condition that the perpetrator and the bystander in the video were two different people, expecting this to eliminate transference effects. They were correct, and concluded that knowing that the perpetrator and a familiar innocent bystander encoded around the same time during a mock-crime video are two different people will not allow for any inferences to be made regarding the identity of each person, because information is provided allowing them to differentiate between the two faces. Experiments 3 and 4 provide evidence that conscious inference occurs at the encoding stage. Experiment 3 requested context information from participants about the video right after encoding and also at the presentation of the lineup (retrieval). They found that during the video the participants made the conscious inference, stating that the innocent bystander was the same person as the perpetrator when giving details about the perpetrators subsequent activities during the video after the occurrence of the mock crime. Experiment 4 required participants to stop the tape while viewing it when they saw the perpetrator after the crime. Participants stopped the tape at scenes in which the innocent bystander was portrayed, thus indicating the temporal sequencing of conscious inference, evident as occurring as early as encoding.

The authors suggest a new title for this phenomenon given that witnesses do remember encountering a bystander, because they have to recognize the face in order to misidentify them from a lineup, which is involves conscious processing despite remembering the incorrect context or role that the bystander actually played (i.e., an innocent role versus the role of the perpetrator during instances in which both are viewed in the same context).

By DP

Read, J. D., Tollestrup, P., Hammersley, R., McFadzen, E., & Christensen, A. (1990). The unconscious transference effect: Are innocent bystanders ever misidentified? Applied Cognitive Psychology, 4, 3-31.

                Read et al., (1990) provide a field approach to studying unconscious transference. Three out of five of the experiments within this article were conducted at various locations external to a typical campus setting (e.g., shopping malls, convenience stores). The authors provide a review of the evidence in support of and the evidence lacking to find support for the unconscious transference effect. Evidence claiming to have found support for the UT effect is examined and critiques are made regarding the validity of the evidence based on the methodology (e.g., photos used instead of live face to face interaction) used and the utility of the results for applications to real world settings.

                Experiments 1 through 3 examine the unconscious transference effect in a field setting (e.g. shopping malls, convenience markets). They each examine and modify duration between initial encoding of the bystander and the perpetrator (who makes a strange change exchange request). Thus, the bystander enters the store either before or after the perpetrator (depending on the experiment) and then asks the clerk (i.e., the participant) about a woman and child (ostensibly his wife and child) being in the store earlier. Lineups, either TA (target absent) or both present (bystander and perpetrator both in the lineup) , were then given to the participants (i.e., clerks) who were in either a control condition (i.e., no bystander interaction) or the transference condition (i.e., bystander interaction).  No evidence of a transference effect was found in Experiments 1 through 3. Results of these experiments are discussed in terms of cautioning that the field of eyewitness identification does not yet have reliable evidence regarding the occurrence of this phenomenon to make suggestions or policy recommendations, nor to speak to the reliability of the effect in court for those who often testify as expert witnesses.

                Experiments 4 and 5 were conducted in a more traditional large classroom setting on a university campus involving a college student sample.  Both examined the effect of having a bystander enter the classroom and also having a "perpetrator" (electrician) entering the classroom. Lineups were then given to participants in a control group (no bystander) and participants in the transference group. No transference effect was found in Experiment 4 as no one in the transference condition chose the innocent bystander from a target absent lineup. If a transference effect were to be found, it would typically be found via a large proportion of those in the transference condition choosing (i.e., misidentifying) the innocent bystander from a target absent lineup.

Experiment 5 was the sole experiment to find a transference effect. This experiment made use of schedules of students concurrently enrolled in an introductory psychology class and other introductory classes from other fields of study. As such, the researchers were able to have bystanders enter some classrooms (for those in the 3 transference conditions; bystander before perpetrator, bystander at same time as perpetrator, and bystander viewed after the perpetrator) but not other classes (for those in the control condition). There was a significant effect for those in the transference conditions compared to the control condition in which those in the transference condition misidentified the innocent bystander to a greater extent than those in the control condition when viewing a target absent lineup with the face of the innocent bystander included.

The authors discuss these results in terms of the usefulness of information regarding instances in which the transference effect was not found and instances/circumstances in which it was found. Not finding an effect can provide useful information regarding phenomena such as the unconscious transference effect. Additionally, the authors speak to the notion that a conscious inference, in which the participants in the transference condition actually think that the perpetrator and the bystander are the same person, may occur at retrieval when viewing the lineups. Thus, these people may think that the bystander and the perpetrator are the same people and thus if viewing a target absent lineup in which the innocent bystander is present, but not the perpetrator, those who have experienced conscious inference erroneously think the bystander is the perpetrator and subsequently misidentify the bystander from the lineup (false alarm) instead of rejecting the lineup (correct rejection).

By DP

Davis, D., Loftus, E. F., Vanous, S., & Cucciare, M. (2008). Unconscious transference can be an instance of change blindness. Applied Cognitive Psychology, 22, 605-623.

            Davis et al., (2008) provide a new conceptual take on unconscious transference, postulating that this phenomenon may, under certain circumstances, result from change blindness. Unconscious transference, in which identifying an innocent person who is familiar instead of a perpetrator, has been received little to moderate empirical support. The literature in the field of eyewitness identification research has shown support for the existence of this phenomenon, yet the question of its reliable observance via empirical investigation is immersed in controversy.

            This study incorporates a concept from the research areas of perception and attention, called change blindness, to view unconscious transference as a type of perceptual illusion carried out by our own brains. Change blindness refers to the modification of some aspect of a visual scene, after which there is either another component added to the scene or removed from the scene. Typically, if a visual scene is occluded and a change occurs while the scene is occluded from our visual field, humans have a difficult time noticing the change (Levin & Simon, 2000). In reference to unconscious transference, if one person, initially detected by another person in the visual field, is removed and replaced with another, typically in close temporal proximity, the person who detected the first person may think that he and the replacement second are one in the same person. Understandably, this can cause problems regarding eyewitness identifications in which innocent bystanders are seen in close temporal proximity to the perpetrator of a crime.

            To examine this experimentally, Davis et al., (2008) conducted a field study at a supermarket in which they had a person (continuous innocent, CI) walking down a liquor aisle in the store, then go behind a stack of boxes, while simultaneously another person (perpetrator) emerged to pocket a bottle of wine. This tape also had a person (discontinuous innocent, DI) in a different aisle picking out a piece of fruit. Participants, after viewing this video engaged in a filler task, were shown a target absent lineup with including both the CI and the DI, and were asked to identify the person who stole the bottle of wine from the liquor aisle. The authors hypothesized that due to an illusion of continuity stemming from the change blindness, witnesses would perceive the CI and the perpetrator to be the same person, and thus anticipated a greater proportion of misidentification for the CI compared to the DI (also viewed in the mock-crime video).

In their first experiment, they found that the CI was erroneously chosen at a greater rate compared to the DI by participants who did not notice the change from the CI to the perpetrator. Two other experiments using the same procedure but with variations in the members included in the lineup presented were conducted and by in large replicated the results obtained in the first study. Clearly, viewing the phenomenon of unconscious transference as an instance of change blindness and failures in scene perception is a beneficial manner of exploring this rather unreliable effect.

By DP

By HC

By HC

By HC

By HC