The following is an excerpt of the intervention methodology. For more information, please see the full text of the article on the publisher's website or through your institution's library.

Study 1 – murder trial
 

Method

The methods for these studies are identical except for the type of trial (murder vs. battery) and resulting semantic differences in the measures. The data, stimuli, and measures from the pilot and current studies are available in the OSF Project: https://osf.io/k5fpa/?view_only=9880bbc7e5fc47b6be7748995d2656b8.

Design
The study employed a 2 (Defendant Race: Black vs. White) × 2 (Mock-Juror Race: Black vs. White) × 2 (UBJ Video: present vs. absent) × 2 (Jury Instructions: UBJ vs. general) between-subjects design.

Trial stimuli
The murder trial transcript was edited from an actual trial used in past research  […] The defendant was charged with murdering his wife. The defendant claimed his wife’s death was an accident – she threatened to kill herself and the gun went off when he tried to take it from her. During the trial, three prosecution witnesses (police officer who responded to the 911 call, a detective, and the coroner) and two defense witnesses (a mental health counselor and the defendant) testified. Witnesses were subjected to direct and cross-examination. The transcripts included photos of the defendant, police officer, detective, medical examiner, defense and prosecuting attorneys, and the judge. Finally, to improve attention, the trial transcripts were broken into three sections and each section was followed by two true/false questions.

Independent variables
Defendant Race. Participants were randomly assigned to trials involving a Black or White defendant.  […]

Unconscious Juror Bias (UBJ) Video. The Western District of Washington (United States District Court, Western District of Washington, 2017) developed the UBJ video (duration = 10:53 min.) used in courts today based on the assumption that it reduces jurors’ implicit biases. Prior to reading the trial summary participants were randomly assigned to view this video or not. The UBJ video can be found here: https://www.wawd.uscourts.gov/jury/unconscious-bias

Unconscious Juror Bias (UBJ) Instructions. The Western District of Washington (United States District Court, Western District of Washington, Citation2017), also developed the UBJ instructions, which are administered pretrial and posttrial. Participants were randomly assigned to read these instructions or the general jury instructions. The general jury instructions do not mention unconscious bias, whereas the UBJ instructions define unconscious bias and explain how it can affect evidence interpretation, credibility perceptions, and decisions.

Measures
Exploratory factor analysis (EFA) with Varimax rotation (when more than one factor was indicated) was used for all of the multi-item measures below to determine the number of factors and their structure. […]

Verdict. Mock jurors provided a dichotomous murder verdict (guilty/not guilty).

Culpability. There were five items for which participants rated the defendant (two items) and victim (two items) on responsibility for the victim’s death or injury (1 =  not at all responsible to 7 = completely responsible), and defendant motive to kill/injure (one item; 1 = strongly agree to 7 = strongly disagree).  […]

Defendant Credibility. The defendant credibility scale […], consisted of 10 items.  […]

Counterfactual Endorsement. Participants read nine statements about the case, six with counterfactual (‘if-then’) implications and three without (e.g. ‘The defendant loved his wife.’). Participants indicated how strongly they agreed with each statement on a seven-point scale (1 = strongly disagree to 7 = strongly agree). EFA with Varimax rotation revealed that four of the counterfactual items loaded on two factors (murder and suicide) with acceptable reliability  […]

 […]

Procedure
Participants were recruited through Qualtrics Panels via an email with the survey link.  […]. All participants were provided with pretrial instructions (general or UBJ), read the trial summary that was divided into three sections and answered two attention checks after each section. Participants were then given trial-specific instructions followed by either posttrial general or UBJ instructions. They then rendered verdicts, completed the culpability, credibility and counterfactual scales, and manipulation checks.

Study 2 - battery trial

The procedure for Study 2 was identical to that of Study 1.

Trial stimuli
 [...]The defendant (starting quarterback) was charged with battery resulting in serious bodily injury of a teammate and known rival. In an altercation between the defendant and victim, the defendant threw the victim off himself, causing the victim to collide with an eyewitness and fall over a bench. As a result, the victim tore his anterior cruciate ligament (ACL), suffering permanent loss of mobility in his leg. The defendant claimed he was acting in self-defense and the injury was accidental. Jurors were instructed on the relevant law for the charge. The trial summary included photos of the defendant (race manipulation), the judge, and prosecuting attorney, as well as an image of the ACL injury. Finally, to improve attention, the trial summary was broken into two sections. Each section was followed by two true/false questions.

Independent variables

Study 2 manipulated the same variables in the same manner as Study 1 (i.e. defendant race, UBJ video, and UBJ instructions).

Measures

EFA was again performed for all of the multi-item measures below.

Verdict. Mock jurors provided dichotomous verdicts (guilty/not guilty) for the Battery with Serious Bodily Injury charge.

Culpability. Culpability items were the same as those used in Study 1 except the crime (murder vs. battery) and outcome (death vs. injury) differed. These items were expected to load on two factors: defendant and victim culpability. Only the defendant culpability scale had good internal consistency (Cronbach’s alpha = .72).

Defendant Credibility. The same 10 items as Study 1 loaded on a single factor  [...].

Counterfactual Endorsement. Participants responded to eight statements about the case, five having counterfactual implications and three that did not (e.g. The victim’s injuries were an unfortunate accident). Participants indicated how strongly they agreed with each statement on a seven-point scale (1 = strongly disagree to 7 = strongly agree).  [...]
 

Manipulation and Attention Questions. Study 2 employed the same manipulation check questions and similar attention check questions as Study 1.