How do they know what they know? [Elektronische Ressource] : the case of eyewitnesses / vorgelegt von Melanie Sauerland

Sujets

• Social work

Informations

• Publié par: justus-liebig-universitat_giessen
• Publié le: 01 janvier 2007
• Nombre de lectures: 56
• Langue: Deutsch

Extrait

How Do They Know What They Know?
The Case of Eyewitnesses
Inaugural-Dissertation
zur
Erlangung des Doktorgrades
der Philosophie des Fachbereichs 06
der Justus-Liebig-Universität Gießen
vorgelegt von
Melanie Sauerland
aus
Bonn
2007Dekan: Prof. Dr. Joachim C. Brunstein
1. Berichterstatter: Prof. Siegfried L. Sporer, Ph.D.
2. Berichterstatter: Prof. Dr. Joachim Stiensmeier-Pelster
Tag der Disputation: 05. Juli 20074
Dank
Während der Erstellung dieser Arbeit haben mich eine ganze Reihe von Personen
unterstützt, denen ich an dieser Stelle ganz herzlich danken möchte.
Mein Dank gilt meinem Betreuer Prof. Siegfried L. Sporer, der es mir ermöglicht
hat, eines der interessantesten Themen der Rechtspsychologie zu bearbeiten. Bedanken
möchte ich mich auch bei meinem Zweitgutachter Prof. Dr. Joachim Stiensmeier-Pelster.
Für die gute Zusammenarbeit, viele lange Gespräche, die unermüdliche Beratung
und Unterstützung danke ich meinen KollegInnen Maike Breuer, Jürgen Gehrke und Tanja
Stucke.
Mein besonderer Dank gilt Gary Wells, der mich während eines Besuchs seines
Labors in Iowa im März 2006 so herzlich empfangen und bei sich aufgenommen hat. Er
nahm sich sehr viel Zeit, um mich bezüglich der Auswertung des vierten Experimentes zu
beraten.
Hervorzuheben ist Christine Ernst, die sich eigentlich inzwischen mit ganz anderen
Sachen beschäftigt, für ihr stetes Interesse an der Thematik und ihr immer offenes Ohr.
Franziska Holub, Elisabeth Baumgartner, Sabine Bloss, Sarah Blum, Verena Bohn,
Kerstin Bullmann, Maike Davids, Claudia Dill, Lisa German, Sarah Jasmin Günthert,
Miriam Herder, Markus Jonitz, Morten Kaletsch, Miriam Kirschner, Norman Koch, Alana
Krix, Silke Leifheit, Nadine Lipke, Isabelle Sauerbier, Alexandra Schmoranzer, Patricia
Waschk und Jürgen Weber sei für die Unterstützung bei der Datenerhebung gedankt.
Weiterhin möchte ich Tanja Basseler, Anne Brinkmann, Cornelia Glagla, Merle
Kurpjuweit, Andrea Reinke, Birgitt Sauerland, Tanya Sauerland und Volker Sauerland
danken, die die vorliegende Arbeit in sehr unterschiedlicher, aber immer wesentlicher
Hinsicht beratend und/oder unterstützend begleitet haben.
Gießen, im Juli 2007 Melanie Sauerland5
Abstract
ABSTRACT
This dissertation reports 4 Experiments that are concerned with the evaluation of eyewitness
identification testimony. In Experiment 1, three target description groups were assessed in
order to test different theoretical accounts regarding the relationship between identification
performance and target description (verbal overshadowing): non-describers, describers, and
rereaders (describers with rereading of the description before the identification task). Reread-
ers less frequently chose somebody from the lineup than the other two groups, lending sup-
port to the decision criterion shift approach (Clare & Lewandowsky, 2004). In Experiment 2,
post-decision confidence, decision time, and self-reported decision processes were used as
postdictors of identification accuracy. Using a decision rule including highly confident and
fast choosers led to more correct classifications than either variable alone. Unexpectedly, self-
reported decision processes were neither associated with identification accuracy for choosers
nor for nonchoosers. In Experiment 3, combinations of post-decision confidence, decision
time, and Remember-Know-Familiar (RKF) judgments were evaluated as postdictors of iden-
tification accuracy in a field experiment with ten targets and a very large sample. Fast and
confident choosers were highly accurate. Including the RKF judgment did not lead to higher
correct classification rates. Participants' self-reported (estimated) decision times also proved
to be a postdictor of identification accuracy. Postdicting nonchoosers' identification perform-
ance by forming three homogeneous groups of nonchoosers failed, although there were differ-
ences with regard to confidence measures and decision times. Experiment 4 tested the useful-
ness of multiple lineup decisions (portrait face, body, bag, and profile face) for the assess-
ment of identification testimony for nine different targets. Performance in the four different
lineup types was not associated with each other, lending support to the idea that multiple
lineups can serve as independent sources of evidence. Compared to foil choices and lineup
rejections, target/suspect choices were most diagnostic of guilt. The portrait face lineup alone
and its combination with the body lineup were most diagnostic for target/suspect choices.6
Abstract
To conclude, the present studies suggest that both decision times and post-decision
confidence should be collected at the time of identification and be combined in order to assess
identification accuracy. Investigators need to be aware though, that there is no postdictive
value of nonchooser's estimates. Furthermore, there seem to be no negative effects of target
descriptions on identification accuracy when there is a sufficient interval between description
and identification, as there is in real cases. Finally, the data speak for the application of mul-
tiple lineups with regard to suspect/target choices as a procedure to avoid false identifications,
whereas the benefit of multiple lineups for lineup rejections and foil choices seems to be lim-
ited. Future studies should address how many and which specific lineup types should be
used. In real cases, the results for the assessment (decision times, confidence, decision proc-
esses) and control variables (target description, multiple lineups) examined in the present dis-
sertation may vary from those that we obtain in laboratory or field studies. Reasons could be,
for example, awareness of the severe consequences of false identifications and false rejections
or the stress level at encoding or recognition. It would be interesting to collect data on these
issues in real cases so they can be compared to the data obtained in laboratory/field studies.
Undoubtedly, it would be a great contribution to the field of identification if data were col-
lected even where DNA samples exist, so that identification accuracy could actually be as-
sessed in real cases.Contents
Table of Contents
INTRODUCTION 9
EXPERIMENT 1
Person Descriptions and Person Identifications: Verbal
Overshadowing or Recognition Criterion Shift? 17
Verbal Overshadowing 17
Context Reinstatement 25
Relationship between Quantity and Quality of Descriptions and Identification Accuracy 26
Method 29
Results 34
Discussion 47
References 52
Appendix 58
EXPERIMENT 2
Post-Decision Confidence, Decision Time, and Self-Reported
Decision Processes as Postdictors of Identification Accuracy 59
Response Latencies as Markers of Accuracy 60
Confidence-Accuracy Relationship 62
Self-Reported Decision Processes 63
Method 67
Results 70
Discussion 81
References 86Contents
EXPERIMENT 3
Fast and Confident: Postdicting Eyewitness Identification
Accuracy in a Field Study 89
Decision Time-Accuracy Relationship 91
Confidence-Accuracy Relationship 93
Remember-Know(-Familiar) Judgments 95
Nonchoosers’ Decision Processes 98
Method 101
Results 104
Discussion 126
References 134
Appendix 143
EXPERIMENT 4
The Application of Multiple Lineups in a Field Study 146
Multiple Lineups 147
Designation of an Innocent Suspect 149
Previous Research on Multiple Lineups 149
Method 152
Results 156
Discussion 168
References 173
Appendix 178
DISCUSSION 189
DEUTSCHE ZUSAMMENFASSUNG 199Introduction 9
INTRODUCTION
When a person witnesses a crime or becomes the victim of a crime, he or she will be
asked to give a description of the perpetrator and later will be presented with a live lineup or
a photospread. On the one side, eyewitnesses are capable of giving valuable testimony, on the
other hand, history has taught us that eyewitness identification evidence is not always reli-
able. A famous case of misidentification is the one of Jennifer Thompson. In 1984, the 22-
year old college student was assaulted and raped by a man who had broken into her apart-
ment. In order to get a better view of him, she lured him into the illuminated parts of the
apartment. She was determined to do everything that would enable her to later give a good
description of the man and to identify him so he could be convicted and pay for what he had
done to her. In the identification procedure she identified a man named Cotton as the offender.
Only little later, police was given a hint by a prison inmate who reported that a person named
Poole had confessed the offence while he had served time with him. Consequently, Jennifer
Thompson was presented with the lineup one more time but she declared that she had never
seen Poole before and that she was absolutely confident that Cotton was the man. The police
believed Jennifer and it was not until 1995 that Cotton was exonerated by DNA analysis. The
analyses also provided evidence that Poole committed the offence. Even though Jennifer had
been very confident about her decision and in spite of the fact that she had been presented
with the actual offender, she erred (for coverage on the case, see
www.truthinjustice.org/positiv