- Top of page
Reducing prejudice is a critical research agenda, and never before has counterfactual priming been evaluated as a potential prejudice-reduction strategy. In the present experiment, participants were randomly assigned to imagine a pleasant interaction with a homosexual man and then think counterfactually about how an incident of sexual discrimination against him might not have occurred (experimental condition) or to imagine a nature scene (control condition). Results demonstrated a significant reduction in sexual prejudice from baseline levels in the counterfactual simulation group. Importantly, whereas intergroup anxiety and motivation to control prejudice were not predictive factors, number of counterfactual thoughts generated independently predicted variance in prejudice reduction. Mechanisms for, and implications of, prejudice-reduction strategies including counterfactual thinking are discussed.
Sexual prejudice is defined as negative attitudes toward homosexual or bisexual persons, behaviors, and communities, and it is the motivational seed behind sexually prejudicial hate crimes. A national survey conducted in 2005 indicated that 49% of sexual minorities reported having experienced verbal abuse, 21% reported having experienced violence or property crime, and 11% reported having experienced housing discrimination (Herek, 2007). As stated by Herek (2007):
The task of confronting sexual stigma and prejudice represents not only an important practical application of our knowledge to a significant social issue. It is also a theoretically challenging area of inquiry that will yield valuable insights into human social behavior. (p. 922)
It has been well established that positive interactions with outgroup members promote favorable intergroup attitudes and reduced prejudice (Allport, 1954; Cook, 1985; Festinger & Kelley, 1951; Pettigrew, 1998; Pettigrew & Tropp, 2006, 2008). Relatively recently, attention has turned to the subtle but remarkably effective role of imagined intergroup contact, that is, mental simulations of positive encounters with outgroup members in enhancing intergroup attitudes and reducing stereotyping (Crisp & Turner, 2009; Stathi & Crisp, 2008; Turner, Crisp, & Lambert, 2007). For example, Turner et al. (2007; Experiment 3) found that heterosexual male participants who imagined talking with a homosexual man subsequently evaluated homosexual men more positively than those who had not imagined contact, and this effect was explained by lower intergroup anxiety in the group that had imagined contact with the homosexual man. Crisp and Turner (2009) noted that, although actual intergroup contact may be ideal in promoting positive intergroup attitudes and relations, lack of opportunity to engage in actual contact due to physical or social segregation or lack of motivation may be barriers to this reality.
One type of mental simulation that may promote prejudice reduction is counterfactual thinking, or alternate-to-reality reflections about how actual events might have transpired differently (Markman, Klein, & Suhr, 2009). Counterfactual thinking is a pervasive human tendency that has implications for social judgment, causal ascription, and emotional reactions including sympathy and blame (Markman, Lindberg, Kray, & Galinsky, 2007; Markman & Miller, 2006). In the case of upward counterfactual thoughts, which are naturally provoked following negative life events, the content of mental imaginings involves how a relatively favorable outcome might rather have occurred. It is believed that such thoughts aid in the construction of useful naïve theories about causality that ultimately form a blueprint for improved future outcomes (Roese, 1994). For example, in light of the current subject, someone might think, “if only people were not prejudiced, this act of discrimination might not have occurred,” and, in turn, translate this conditional supposition into future intentions (e.g., “if I challenge my own prejudicial attitudes, I will not discriminate against others this way”).
Counterfactual priming has been shown to induce consideration of alternatives with potential to debias immediate, conventional solutions to problems (Galinsky & Moskowitz, 2000; Markman et al., 2007; Roese, 1994). Yet, despite the promise of counterfactual thinking for transcending cognitive constraint, never before has it been investigated as a method that might promote prejudice reduction. The current study is premised on the notion of counterfactual simulation as a potentially debiasing problem-solving strategy; in particular, it was hypothesized that counterfactual thinking may correct the stereotypic and constricted thinking that characterizes prejudice and, in this way, promote prosocial attitude change.
In the present study, we utilized an imagined contact experimental paradigm similar to that of Turner et al. (2007), modified to include a counterfactual prime. Specifically, participants were prompted to imagine contact with a homosexual man and then think counterfactually about how an incident of sexual discrimination against him might not have occurred. It was hypothesized that sexual prejudice reduction would be greater in the counterfactual simulation condition as compared to a control condition. The current study design also allowed for assessment of actual reduction in sexual prejudice, given inclusion of pre- and post-mental simulation assessment of prejudicial attitudes, and examination of multiple potential predictors of prejudice reduction, including degree of counterfactual thought generation, intergroup anxiety, and motivation to respond without prejudice.
- Top of page
To test the primary hypothesis that participants in the counterfactual simulation condition would experience greater reduction in prejudicial attitudes toward gay men relative to participants in the control condition, a planned comparison was conducted. Results indeed demonstrated that participants in the counterfactual simulation condition (M = 1.53, SD = 4.34) reported significant reduction in prejudicial attitudes relative to participants in the control condition (M = −0.84, SD = 3.91), t(72) = 2.47, p = .02. Descriptive statistics are reproduced in Table 1 for ease of comparison. It is also shown in the table that neither intergroup anxiety nor internal or external motivation to respond without prejudice differed between the experimental and control groups, all ts < .57, all ps > .57.
Table 1. Descriptive Statistics and Group Comparisons
|Condition|| || || || || |
|Experimental (n = 37)||1.46a (0.96), 0.00, 5.00||15.76 (6.55), 6.00, 31.00||36.08 (8.98), 15.00, 45.00||16.24 (9.05), 5.00, 39.00||1.53a (4.34), −7.00, 13.00|
|Control (n = 39)||0.00b (0.00), 0.00, 0.00||16.26 (6.10), 7.00, 30.00||34.95 (8.55), 13.00, 45.00||16.59 (9.22), 5.00, 38.00||−0.84b (3.91), −13.00, 5.00|
|Total sample (N = 76)||0.71 (0.99), 0.00, 5.00||16.01 (6.29), 6.00, 31.00||35.50 (8.72), 13.00, 45.00||16.42 (9.08), 5.00, 39.00||0.31 (4.27), −13.00, 13.00|
Next, potential predictors of prejudice reduction in the experimental group, including number of counterfactual thoughts generated, intergroup anxiety, and motivation to respond without prejudice, were examined. Zero-order correlations among variables are shown in Table 2. As displayed in the table, number of counterfactual thoughts generated (r = .35, p = .04), but neither intergroup anxiety nor internal or external motivation to respond without prejudice (all rs < .17, all ps > .33), related to prejudice reduction. A linear regression analysis indicated that degree of counterfactual thought generation, the hypothesized potent factor of the current intervention, predicted 12% of the variance in sexual prejudice reduction, F(1, 34) = 4.63, p = .04. Interestingly, internal motivation to respond without prejudice was positively related to number of counterfactual thoughts generated (r = .33, p < .05), suggesting an individual difference variable that may enhance propensity toward counterfactual simulation.
Table 2. Zero-Order Correlations Among Potential Predictors and Prejudice Reduction in Experimental Group (n = 37)
|1. Prejudice reduction||.35*||−.17||.13||−.13|
|2. Counterfactual thought generation|| ||−.12||.33*||.11|
|3. Intergroup anxiety|| || ||−.48**||.48**|
|4. Internal motivation to control prejudice|| || || ||−.11|
|5. External motivation to control prejudice|| || || || |
- Top of page
The present results suggest that modification of an imagined contact intervention with a counterfactual prime concerning how an act of discrimination toward a homosexual man might not have occurred produced reduction in prejudicial attitudes—that is, pre-to-post-intervention attitude change. Importantly, number of counterfactual alternatives to discrimination generated by participants predicted prejudice reduction, whereas intergroup anxiety (a primary explanatory variable of the imagined contact effect) and motivation to control prejudice were unrelated to prejudice reduction.
Counterfactual thoughts engendered by the experimental intervention included those that altered socially prejudicial attitudes (e.g., “if only America was more accepting of homosexuals and stopped treating them as outcasts …”) and those that altered the homosexual man's behavior (e.g., “If Brian hadn't told anyone that he was gay …”). Yet, those in the latter category often continued, for example:
Not that it's not ok that he's gay, but that makes a lot of people feel uncomfortable. I would feel bad for Brian and ask if there was anything I could do, but I think it's something he'll have to get used to until society fully accepts homosexuality.
Both types of counterfactual simulations were included in the present analyses, as even counterfactual thoughts that seemingly located control in the victim of prejudice appeared to engender consideration of societal faults perceived to necessitate such accommodation.
Overall, consistent with and extending research suggesting that counterfactual priming reduces constricted problem solving (Galinsky & Moskowitz, 2000; Markman et al., 2007; Roese, 1994), findings from the current experiment are the first to suggest that counterfactual thought generation reduces sexually prejudicial attitudes. This effect was demonstrated by modifying an imagined contact experimental paradigm used by Turner et al. (2007) to include a counterfactual prime. Whereas extant research has suggested that intergroup anxiety accounts for the effect of imagined contact on prejudice reduction, priming participants in the present study to focus on their cognitive processing (i.e., counterfactual simulations) rather than on their emotional processing highlighted yet another avenue for attitude change. That is, present findings do not diminish the demonstrated influence of imagined contact and, in turn, lower intergroup anxiety, as our intervention constituted an integration of the imagined contact paradigm with a counterfactual simulation enhancement of it. Rather, the present findings suggest another mechanism—counterfactual thinking—that might function in parallel to reduce prejudice. Future research might consider the relative influences of identified prejudice-reduction interventions and, moreover, deviate from the standard “nature scene” control condition (see Stathi & Crisp, 2008) to rule out additional factors (e.g., sympathy for the victim) that might contribute to sexual prejudice reduction.