Factoring the Role of Eyewitness Evidence in the Courtroom

• Author: Chad S. Dodson,Karen Kafadar,Joanne Yaffe,Alice Liu,Brandon L. Garrett
• Date: 01 September 2020
• DOI: http://doi.org/10.1111/jels.12259
• Published date: 01 September 2020

Journal of Empirical Legal Studies

Volume 17, Issue 3, 556–579, September 2020

Factoring the Role of Eyewitness Evidence

in the Courtroom

Brandon L. Garrett,*Alice Liu, Karen Kafadar, Joanne Yaffe, and

Chad S. Dodson

A pressing concern with the eyewitness testimony used in many criminal cases is that jurors

may be swayed by the high confidence of an eyewitness and, as a result, may disregard

other factors that provide more diagnostic information. Mock jurors were surveyed using a

large national sample of 1,684 laypeople, selected to be representative of the

U.S. population (age, race, gender, geographic region), using mock trial videos of eyewit-

ness testimony. To explore the relationship between courtroom confidence and other fac-

tors, we used a fractional factorial design, permitting examination of the relationships

among seven factors. Among these seven factors, we found that jurors gave most weight to

the confidence of eyewitnesses, especially that expressed in the courtroom, irrespective of

the eyewitness’s testimony about confidence (low or high) at the initial police lineup.

Jurors’ assessments were not sensitive to the other factors or their interactions in the

experiment: crime type (burglary or sexual assault), the race of the defendant and eyewit-

ness, or information provided in judicial instructions or by expert testimony. The dispro-

portionate importance of the eyewitness’s expressed confidence has implications for the

effectiveness of legal efforts to inform jurors about factors affecting eyewitness memory.

I. Introduction

In a well-known example of an eyewitness identification leading to a wrongful conviction,

Jennifer Thompson misidentified Ronald Cotton, who spent over 10 years in prison

before DNA exonerated him and identified the true culprit. Thompson expressed uncer-

tainty in her identifications at both a photo lineup and a live lineup, and told police that

she had a hard time deciding between two people in the lineup (saying only that Cotton

“looks the most like him”). However, Thompson’s confidence was boosted by police offi-

cers’ suggestive comments and by the fact that only Cotton was repeated from the photo

*Address correspondence to Brandon Garrett, L. Neil Williams, Jr. Professor of Law, Duke University School of

Law, 210 Science Dr., Durham, NC 27708-0373; email: bgarrett@law.duke.edu. Liu received her PhD at the Univer-

sity of Virginia; Kafadar is Commonwealth Professor of Statistics and Department Chair, Department of Statistics,

University of Virginia; Yaffe is Professor, College of Social Work, University of Utah; Dodson is Professor of Psychol-

ogy, Department of Psychology, University of Virginia.

This research was supported by a grant from Arnold Ventures. This article is an independent work product.

The views expressed are those of the authors and do not necessarily represent those of the funder. The authors

are extremely grateful to the anonymous reviewers, whose comments have greatly improved this article.

556

array to the live lineup. During the criminal trial some months later, Thompson’s confi-

dence shifted to “absolutely sure” and she confidently identified Cotton in the courtroom

(Garrett 2011; NRC 2015). That increase in confidence has been termed confidence

inflation (Wells et al. 1998). Such confidence inflation has been observed in other DNA

exoneration cases in the United States (Garrett 2011).

Eyewitness testimony is crucial in many criminal cases, and yet cases like Ronald Cot-

ton’s illuminatethe results of more than three decades of memory research: unreliable eye-

witness testimony can be extremely powerful to jurors. Many studies have shown that jurors

are not aware of the limitations of eyewitness testimony that memory researchers have stud-

ied (Benton et al. 2006; Cutler & Penrod 1995; Cutler et al. 1990a; Desmarais & Read 2011;

Kassin & Barndollar 1992; Simon & Chabris 2011): jurors place great weight on the confi-

dence of an eyewitness in the courtroom (Bradfield & McQuiston 2004; Brewer &

Burke 2002; Cutler et al. 1990a; Douglass et al. 2010; Lindsay et al. 1981).

In addition to courtroom confidence, some researchers have argued that jurors

should focus on a different factor: the confidence of an eyewitness in the initial out-of-

court identification. Jennifer Thompson’s uncertainty at the time of the police lineup

procedures should have been a red flag. Recent research describes how an eyewitness’s

confidence at the time of a prior lineup identification and the conditions under which it

took place can provide important information regarding eyewitness accuracy (Wixted &

Wells 2017).

In response, influential courts, such as the New Jersey and Massachusetts Supreme

Courts, have adopted legal reforms to promote better jury understanding of research on

eyewitness evidence, including mechanisms to focus jurors on factors such as the eyewit-

ness confidence at the time of the earlier lineup, rather than on courtroom confidence

(Commonwealth v. Gomes 2015; New Jersey v. Henderson 2011). However, some research sug-

gests that reformed judicial instructions are not effective at conveying such information

(Cutler et al. 1990b; Dillon et al. 2017; Jones et al. 2017; Papailiou et al. 2015).

As a result, a 2015 National Research Council Report (hereafter, NRC 2015), concluded

that more research was needed on how to better sensitize jurors about the strengths and weak-

nessesofeyewitnessevidence.NRC2015notedconcerns regarding the effectiveness of jury

instructions as a means to educate jurors, preferring instead the use of expert witnesses to

explain eyewitness memory research to the jury (NRC 2015:27–29). Further, the NRC 2015

noted that eyewitness research typically has involved small-scale studies that have not ade-

quately examined interactions among multiple variables (NRC 2015:2–3). Although the NRC

report cited research on eyewitness evidence and jury decision making, research on how eye-

witnessconfidenceaffectsjurors,andhowtobestaddressit,hasbeenslight.Nor,astheNRC

report emphasized, have interactions among multiple variables been thoroughly studied.

One purpose of this study is to examine whether an eyewitness’s expression of con-

fidence in the identification in the courtroom dominates other factors that could be

more relevant, and thus could or should influence jurors. Also, previous studies often

involved relatively small numbers of participants, typically undergraduate students, who

may not be representative of jury-eligible adults. This study aims to address those inadequa-

cies in the literature by conducting a study using an experimental design that included

multiple factors and with jury-eligible adults as participants. A second purpose of this study

Factoring Eyewitness Evidence 557