The Effect of Blinded Experts on Juror Verdicts

• DOI: http://doi.org/10.1111/j.1740-1461.2012.01273.x
• Author: David V. Yokum,Christopher T. Robertson
• Date: 01 December 2012
• Published date: 01 December 2012

The Effect of Blinded Experts on

Juror Verdicts

Christopher T. Robertson and David V. Yokum*

“Blind expertise” has been proposed as an institutional solution to the problem of bias in

expert witness testimony in litigation (Robertson 2010). At the request of a litigant, an

intermediary selects a qualified expert and pays the expert to review a case without knowing

which side requested the opinion. This article reports an experiment that tests the hypoth-

esis that, compared to traditional experts, such “blinded experts” will be more persuasive to

jurors. A national sample of mock jurors (N=275) watched an online video of a staged

medical malpractice trial, including testimony from two medical experts, one of whom (or

neither, in the control condition) was randomly assigned to be a blind expert. We also

manipulated whether the judge provided a special jury instruction explaining the blinding

concept. Descriptively, the data suggest juror reluctance to impose liability. Despite an

experimental design that included negligent medical care, only 46 percent of the jurors

found negligence in the control condition, which represents the status quo. Blind experts,

testifying on either side, were perceived as significantly more credible, and were more

highly persuasive, in that they doubled (or halved) the odds of a favorable verdict, and

increased (or decreased) simulated damages awards by over $100,000. The increased

damages award appears to be due to jurors hedging their damages awards, which interacted

with the blind expert as a driver of certainty. Use of a blind expert may be a rational strategy

for litigants, even without judicial intervention in the form of special jury instructions or

otherwise.

I. Background

The U.S. legal system tasks judges and jurors—both laypersons as to science—with resolving

highly technical questions. These laypersons are asked, for example, to evaluate DNA

evidence to determine whether it inculpates a particular defendant, to determine the

standard of care for lumbar radiculopathy, to interpret epidemiological data to determine

whether a given chemical causes an observed disease, and to ascertain the state of the art in

a patent suit for computer software. Thus, in both civil and criminal litigation, expert

*Address correspondence to Christopher T. Robertson, James E. Rogers College of Law, the University of Arizona, PO

Box 210176, Tucson, AZ 85721; email: chris.robertson@law.arizona.edu. Yokum is J.D./Ph.D. candidate, James E.

Rogers College of Law, Department of Psychology, The University of Arizona.

Thanks to Gregory Schwartz, Aaron Kesselheim, and Tom Mauet for serving as actors and consultants; to James

Greiner, George Kimball Smith, and anonymous reviewers for the Conference on Empirical Legal Studies and Journal

of Empirical Legal Studies for commenting on drafts, and to Germar Townsend, Tess Gemberling, Carol Ward, Judy

Parker, Barbara Lopez, and Bert Sky for excellent research and administrative support.

bs_bs_banner

Journal of Empirical Legal Studies

Volume 9, Issue 4, 765–794, December 2012

765

witnesses play an increasingly prominent and important role to inform those layperson

decisionmakers. In civil litigation, experts appear in the vast majority of trials (Gross 1991),

and courts have held in the criminal context that the right to effective counsel includes the

right to funding for expert testimony (Ake v. Oklahoma).

In the U.S. legal system, expert witnesses are hand-picked by the litigants, coached

toward favorable opinions, and compensated for continuing work only as long as those

opinions are favorable (Robertson 2010). These experts are thus subject to various cog-

nitive biases, which have been extensively documented in the social sciences (Risinger

et al. 2002). Reflecting on his own experience, John Langbein (1985) has explained that

“those of us who serve as expert witnesses are known as ‘saxophones[,]’ . . . a musical

instrument on which the lawyer sounds the desired notes.” Courts have criticized experts

who altogether “cast aside [their] scholar’s mantle and [became] a shill” for the party

that retained them (Mid-State Fertilizer Co. v. Exchange National Bank of Chicago). The

problem is not limited to outliers, however. In one study, both federal judges and attor-

neys cited “experts abandon[ing] objectivity and becom[ing] advocates for the side

that hired them” as the most frequent problem with expert testimony (Johnson et al.

2000).

Experimental studies have shown that witnesses interviewed by partisan attorneys

prior to testifying tend to deliver more biased testimony than those interviewed by nonpar-

tisan attorneys (Sheppard & Vidmar 1980). In criminal cases, scholars have documented a

significant bias of expert opinions in the direction of the side that requested the opinion

(Otto 1989). One litigation-oriented study (Gitlin et al. 2004) compared the opinions

rendered by radiologists retained by plaintiffs’ attorneys in asbestos cases to opinions

rendered by a panel of “independent” radiologists retained by defense attorneys.1The

plaintiffs’ experts found physiological abnormalities 95.9 percent of the time, while “inde-

pendent” reviewers found abnormalities only 4.5 percent of the time. These sorts of biases

may operate subconsciously (e.g., an anchoring heuristic driven by the litigant’s suggestion

as to the amount of damages) or consciously (e.g., a desire to maintain a stream of income

from the litigant).

Expert biases may be the subject of cross-examination in trials, but that is typically a

zero-sum game, since both litigants tend to exploit these biases. Although court appoint-

ment of experts is a possible solution, it is almost never actually employed, for cultural and

economic reasons, in the adversarial U.S. legal system (Gross 1991; Cecil & Willging 1993).

Without a real alternative to biased expert witnesses, it is not practicable for judges to simply

exclude all such testimony; instead, the exclusion doctrine—based on Daubert v. Merrell Dow

Pharmaceuticals—operates only in marginal cases. It is not a solution for the everyday

instances of litigant-induced biases.

When layperson factfinders (whether judge or jury) are left to rely on traditional

expert witnesses who are subject to these biases, the factfinders receive relatively weak

1We put “independent” in scare quotes because the study authors admit that the research was done on behalf of

defense attorneys. See Oliver et al. (2004), arguing that the Gitlin “study was done at the behest of attorneys for

defendants in asbestos litigation.”

766 Robertson and Yokum