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.

Experiment 1

Participants. Participants were non-Black (determined based on prescreening) undergraduate students who received course credit toward their Introduction to Psychology grade. Data from five participants who did the photo-description task incorrectly were deleted, as were data from eight participants who had considerable missing data. The final sample consisted of 166 participants (51.8% men; 97.0% White, 1.2% Asian, 0.6% American Indian, and 1.2% missing; Mage = 19.55, SD = 1.19). [...]

Design. A 2 (testing session: immediate vs. delayed) x 3 (condition: control vs. internal framing vs. external framing) between-participants design was used.

Procedure. Up to seven participants per session were brought to the lab and seated at individual computer stations. [...] Participants first completed an inference task for eliciting stereotyping. Across 20 trials, they saw a photograph of a person and a brief description, and were instructed to type a fitting one- or two-word inference. [...] Participants were encouraged to respond quickly, typing the first logical response that came to mind, although a specific time restriction was not imposed. Based on random assignment, participants received one of two sets of items, both of which included three critical trials. Critical trials included a photograph of a Black man and a description that could elicit a stereotypic response. [...]

Confrontation manipulation. Participants received fixed feedback that they demonstrated “slightly above average” analytic reasoning ability. [...] Control participants proceeded to the next task. Confronted participants read that their responses revealed additional information and saw a list (e.g., deductive errors, especially slow responding) with an “X” next to “Racial Bias.” The following explanation was then provided (tailored to the critical trials actually received): This means that you likely generated responses to photographs of Blacks that reflected racial stereotypes.[...]

Next, all participants completed 10 filler items to boost the cover story.

Testing session manipulation. Participants in the immediate condition completed the dependent measures next, whereas participants in the delayed condition were informed they would receive an email in 2 days to complete other tasks online. [...] Participants were informed that they would not receive feedback based on their responses. Participants then completed a second set of 20 photo-description pairs, including three trials that served as the measure of postconfrontation stereotyping. Participants who had completed Set 1 during the preconfrontation phase, completed Set 2 postconfrontation, and vice versa. After completing the second set of photodescription pairs, recall for the confrontation framing was assessed. [...]

Experiment 2

Experiment 2 investigated confrontations of both sexism and racism. [...] Additionally, Experiment 2 investigated whether participants’ chronic internal and external motivations to respond without prejudice moderated the effect of confrontations varying in motivational focus.

Participation was restricted to people who had not completed other research from our laboratory that had similar materials or involved confrontation. Thirteen participants completed the photo-description task incorrectly, and seven participants marked that their data could not be used on a postsession consent form that explained the use of deception. These data were evenly spread across conditions, and they were deleted. This left 245 participants (71% women; 81.9% White, 3.4% Black, 8.3% Asian, 4.5% Hispanic, and 1.9% “other”; Mage = 38.68, SD = 14.65). [...]

Design A 2 (type of bias: gender vs. race) x 3 (confrontation: none vs. internal framing vs. external framing) between-participants design was used, and IMS and EMS varied continuously.

Procedure The procedure was identical to Experiment 1 except where noted.

Bias manipulation. Participants randomly completed a race or gender version of the procedure. The same critical photo-description pairs from Experiment 1 were used in the race-bias condition. Critical trials were replaced by gender photodescription pairs in the gender-bias condition, which played on role restrictive, lower status, stereotypes about women. [...] As in Experiment 1, an explanation of what constitutes stereotypic responding on task, with examples, followed. Confronted participants then read the internal or external framing from Experiment 1 tailored depending on bias type condition. Participants then completed 10 filler items consistent with the cover story.

Dependent measures. Next, participants completed 20 postconfrontation photo-description pairs (Set 1 or 2, depending on preconfrontation set). Participants in the race-bias condition then completed the race-relevant IMS and EMS items (Plant & Devine, 1998). Participants in the gender-bias condition completed gender-relevant versions [...] Finally, confronted participants completed the same items from Experiment 1 concerning recall of the confrontation.

Experiment 3

Experiments 1 and 2 [...] did not address whether motivational framing is critical to confrontation effectiveness. [...] Experiment 3 investigated this issue by including a condition in which participants were informed that their responses were stereotypic (including the explanation of what constitutes stereotypic responding with examples), and comparing it to the internal and external framing confrontations, as well as to a control condition. Experiment 3 also addressed how participants perceived the internal and external framings. [...] Finally, Experiment 3 sought to replicate the gender-bias confrontation effects observed in Experiment 2, and to determine whether we would replicate the null findings suggesting that participants’ internal and external motivations did not interact with confrontation.

Participants were recruited from MTurk in return for $0.75. Participation was restricted to people who had not completed other research from our laboratory that had similar materials or involved confrontation. This left 237 participants (65% women; 78.5% White, 4.0% Black, 8.0% Asian, 6.0% Hispanic, 3.2% “other,” 0.4% missing; Mage = 39.03, SD = 12.30). [...]

Design A single-factor between-participants design was used (control condition, confrontation without motivation, internal framing, external framing), and IMS and EMS varied continuously.

Procedure The procedure was identical to the gender-bias condition of Experiment 2, with two exceptions. First, control participants were presented with the same list of possible errors/biases as confronted participants but learned they had not demonstrated any biases. [...] all confronted participants were informed of their stereotypic responses and saw nonstereotypic alternatives that would have constituted nonstereotypic responses. Participants in the internal and external framing conditions then received the same motivational framings described previously. Second, perceptions of feedback were assessed, with particular interest in whether the internal and external framings differed in line with the intended motivational focus. [...] Participants then completed the postconfrontation stereotyping task, followed by the IMS.