Imagining Discrimination

Evidence that at least sometimes,
- people "are just imagining it";
- that if you encourage a victim mentality people will perceive themselves as more discriminated against;
- that people are prejudiced against people who are thought to be prejudiced;
- that people might blame others for discriminating against them to feel better about themselves;
- and that all this screws the lives of the "victims" up:

"Feldman Barrett (1996; Feldman Barrett & and Fong, 1996) proposed an elaboration of the Lazarus and Folkman model, drawing on signal detection theory (SDT) to explain variations in the primary appraisal process. Although it was originally used as a model for understanding perceptual errors (misses and false alarms) in judging psychophysical signals, SDT has been applied to judgments in many psychological domains. Feldman Barrett and Fong (1996) argued that there are different psychological and interpersonal costs associated with misses and false alarms when applied to appraisals of threat. They suggest that people weigh the psychological costs of each type of error when making threat appraisals, thereby providing a motivational explanation for people’s judgment strategies...

SDT’s most significant theoretical contribution to understanding the judgment process lies in its ability to separate an observer’s actual judgment behavior into two subprocesses: sensitivity and response style or bias (Harvey, 1992). Sensitivity has been defined as an observer’s ability to accurately detect sensory information when it is present and its absence when it is not present. A target’s sensitivity to prejudice would reflect her or his ability to accurately detect the presence or absence of cues indicating prejudice and discrimination. Sensitivity may vary because people differ in their perceptual abilities or because of the properties of the stimulus. A stimulus’ probability of occurrence, intensity, and imminence (i.e., proximity to danger) will affect its ambiguity, and therefore a perceiver’s sensitivity (McNicol, 1972; Miller, 1979; Paterson & Neufeld. 1987).

In contrast to sensitivity, response style or bias is defined as the observer’s tendency to favor one response over another, independent of the base rate for the stimulus. Thus, a response bias for prejudice exists when an individual judges a situation or person as prejudiced or discriminatory more or less frequently than prejudice or discrimination objectively occurs in that environment. Response bias (i.e., the placement of an observer’s decision criteria) is influenced by two factors: the observer’s beliefs about the base rates of the stimuli; and the goals that she or he has when making a judgment about a stimulus (Egan, 1975; Green & Swets. 1966; 1974; 1-lealy & Kubovy, 1978), in particular, the perceived severity and consequences of a miss or false alarm (Feldman Barreu & Fong, 1996). There is no requirement that individuals are consciously aware of their response biases, and in fact they may function outside the observer’s awareness (Harvey, 1992)...

The notion of judgment outcomes can be applied to perceptions of prejudice and discrimination. For example, an African American individual may be in a situation where he or she is barred from entering a store that is about to close, but he or she notices that the manager allows a European American to enter. In this scenario, sensitivity is indicated by whether or not the African American individual notices the incongruence at all; response bias is indicated by how the person interprets the incongruence. A person with a stringent decision criterion may not judge the event to be discriminatory. A person with a more lenient decision criterion, however, will be more likely to perceive the event as discriminatory. The extent to which a person has a stringent or lenient decision criterion (response bias) is likely to be a function of many things, including his or her previous experience with prejudice in that environment (i.e., the perceived base-rates), and his or her need to be self-protective versus accurate (i.e., the goal associated with making the judgment). The goal associated with making the judgment is strongly linked to the perceived cost of making a judgment error (i.e.. the cost of a miss versus the cost of a false alarm)...

A third party observer who is independent of the target and the store manager can be used to determine the presence or absence of the stimulus criterion. This third party observer is not necessarily “objective.” but is independent of the victim—perpetrator system. Although the third party observer may have motivations that influence where he or she sets the stimulus criterion, they are not the same motivations a.s those of the perceiver (which constitute bias). Thus, the actual absence or presence of the stimulus cue is decided by an external source; it is ambiguous and probabilistic, but the relativity is taken out of the hands of the perceiver/target, and this allows us to distinguish between the decision criterion, which is bias, and the stimulus criterion, which is not...

Failure to change a decision criteria in response to new base rates for threat can occur for three reasons. First, individuals using a zero-miss strategy may fail to calibrate to the base rates of the larger environmental context because of behavioral restriction. Avoiding certain situations of people is one way to avoid a miss. Such avoidance also prevents individuals from encountering discomfirming evidence, however, and that in turn contributes to maintaining the use of a zero-miss strategy.

Second, individuals using either a zero-miss or a positive illusion strategy may fail to detect changes in their environment because of cognitive biases. Previous experience will produce cognitive structures that direct attention to information that is consistent with the formative environment and lifter out that which is inconsistent. An individual will develop cognitive structures that facilitate or inhibit threat detection, associated with expectancy that either most, or few, experiences have the potential to be dangerous or harmful. In either case. the individual develops well-entrenched assumptions about how to interpret ambiguous stimuli (Ittlesone & Kilpatrick, 1951) and will be chronically prepared to deal with ambiguous events (Kahneman & Tversky, 1982) in a way that matches their formative environment. Previous research suggests that implicitly held expectancies mediate the large effects of context on recognition, and exert their greatest influence on the interpretation of ambiguous stimuli (Epstein & Roupenian, 1970). Expectancies that have developed over a lifetime of previous experience not only have a profound effect on judgments, but they are usually inaccessible to conscious knowledge or intention, function automatically and effortlessly, and essentially constitute a dispositional preparedness for detecting threat (Ittlesone & Kilpatrick, 1951: Kahnemafl & Tversky. 1982: Posner, 1978). As a result, the individual may not be consciously aware that he or she has been trained to detect or avoid threat and may have limited sensitivity to the increase or decrease in threat cues in a new or changed environment. In addition, decision rules are typically learned deductively (Einhorn. 1982) and are used without intention or awareness (Lewicki, Kill, & Sasaki, 1989). These decision rules structure the encoding of ambiguous infonnation such that it will be seen as confirming evidence and thereby strengthen the further use of the rule (Kahneman & Tversky. 1982). As a result, confirmatory biases will lead people to try to verify, rather than falsify, their working hypotheses about the world.

Third, individuals using either strategy may fail to calibrate to a change in environmental conditions for emotional reasons. Threat appraisals may constitute an aversive learning context that has intense emotional consequences for judgment errors. If individuals modify their learned judgment strategy in any way, they will encounter more errors of the type that they have learned to avoid. Not only will the individual suffer the full consequences of the current judgment error, but he or she will have to tolerate the emotional arousal associated with an error; the individual may even recall or even re-experience similar previous situations where he or she suffered in some way by making the error. Thus, the error will likely have a strong emotional currency because in the formative environment, it was psychologically structurally, or physically costly . Because judgment errors will be emotionally disruptive to the individual, they may retain strong reinforce ment power and may subsequently reinforce the readoption of the original appraisal strategy.

This theoretical framework suggests the critical role that previous experiences can play in affecting the judgment strategies that individuals use to determine whether or not they have been a target of prejudice or discrimination...

Ruggiero and colleagues (Ruggiero & Major, 1997; Ruggiero & Taylor, 1995, 1997) have conducted several studies testing the impact of base rates on attributing (or not attributing) negative evaluations to prejudice. In these studies, participants first completed a test of their abilities. They were then told that their tests would be graded by one of eight outgroup members (e.g., males for female participants). Participants were also told that of these eight people, either all (100%), 6 (75%), 4 (50%), 2 (25%). or none (0%), were known to discriminate against members of the participant’s group. After a delay period, participants received a failing grade on the test, making them ineligible for a lottery. Finally, participants completed dependent measures that included a rating of the extent to which they attributed the grade they received to discrimination. Consistent with the prediction that prejudice is likely to be perceived when the probability of occurrence is high, participants were most likely to judge negative feedback as prejudicial when 100% of the evaluators discriminated against their group. Similarly, a base rate of 90% led individuals to more frequent judgments of discrimination than did lower base rates (Ruggiero & Taylor, 1995). Interestingly, the relationship between probability of occurrence and judgments of prejudice was not linear, because attributions to discrimination did not differ when the base rates were 75%, 50%, 25%. and 0%. The one exception to this finding was for European American men, whose attributions to discrimination decreased in a stepwise fashion from the 100% to the 0% conditions (Ruggiero & Major, 1997). Findings from all four groups suggest that probability of occurrence does affect attributions to discrimination. The absence of complete reliance on base rates. however, suggests that other factors (i.e., response biases) were also influencing attributions...

People are more likely to identify an event that has occurred to someone else as sexual harassment when the event had negative repercussions for the target than when the same event had no repercussions (York. 1989). Additionally, the presence of positive as well as negative outcomes could reduce the perceived intensity of the negative outcomes and thereby reduce the judgments of prejudice. For instance, the positive aspects of benevolent forms of discrimination such as paternalism (Glick & Fiske, 1995; VandenBerghe. 1967) may make it difficult for people to recognize this type of differential treatment as indicative of prejudice (Swim, Cohen, Hyers, Fitzgerald. & Bylsma, 1997)...

People have beliefs about who is prejudiced against whom. These beliefs (like stereotypes) can be defined as perceived base rates or perceptions of the probability that certain people will be prejudiced (Locksley, Borgida, Brekke, & Hepburn, 1980; McCauley & Stilt, 1978). For example, participants are more likely to label a male (versus a female) instigator as sexist, even when instigators engaged in identical behavior (Baron. Burgess, & Kao, 1991, Inman & Baron. 1996). Similarly, European American instigators are labeled as racist more often than are African American instigators, even when they engaged in identical behavior (Inman & Baron, 1996).

Behaviors are likely to vary in the extent to which they are perceived to represent prejudice against one’s group (Swim. Cohen, & Hyers, this volume). Differences in judgments of what constitutes a prototypic prejudicial behavior could explain why Blacks (primarily of West Indian heritage) were more likely than East Asians to indicate that a low grade was a result of discrimination (Ruggiero and Taylor, 1997). Even though both the East Asian and Black participants underutilized base-rate information, the Black participants were more likely to do this than the East Asian participants. Ruggiero and Taylor (1997) suggest that this group difference might be the result of differences in the tendency to make internal attributions for the low grade. An alternative explanation, however, is that negative evaluations in academic contexts are less prototypic for Asian students than for Black students. Hence, the Black participants may have believed that low academic scores are a prototypical cue of prejudicial treatment more so than did the Asian participants...

People have beliefs about the extent to which they or members of their social group have experienced prejudice and discrimination. The available evidence suggests that some of these beliefs influence attributions to prejudice. In the previously described study by Ruggiero and Taylor (1996). women were asked to indicate the extent to which they themselves, and women in North America, had experienced discrimination from men. The first question represents participants’ perceived base rates for their own personal experiences with discrimination, and the latter represents their perceived base rates for women in general. Ruggiero & Taylor (1996) found that perceptions of personal experience with discrimination were associated with attributing a failing grade to discrimination. Similarly, African American teenagers who believed that they were more likely to be personally discriminated against were also more likely to indicate that scenarios describing prototypical incidents of discrimination were indicative of discrimination (Taylor, Ruggiero, & Louis, 1996). Unlike beliefs about personal encounters with prejudice. however, beliefs about the tendency for one’s group to experience discrimination were not predictive of women’s or African Americans’ judgments (see Taylor et al., 1996, for a possible explanation for the difference in predictive power for the two types of base rates)...

The social psychological literature is replete with references to the ways that goals can influence judgments about other people (Fiske & Taylor. 1991). Judgments of prejudice and discrimination are no different. When we judge a person to be prejudicial, we are using stereotype information about who is likely to be prejudiced against whom. Therefore, past research on cognitive and motivational factors influencing the use of stereotypes (e.g., Brewer. 1996; Neuberg & Fiske, 1987) is likely to inform us about how goals of accuracy and self-protection influence the types of judgment strategies that people use when making appraisals of prejudice and discrimination...

Early research on reactions to discrimination indicated that after receiving negative feedback from a male evaluator, women who did not identify the evaluators as prejudiced had lower global self-esteem than women who did make this attribution (Dion, 1975, 1986). Results consistent with these conclusions have been found for the impact of negative evaluations on African, Jewish, and Asian Americans (Dion. 1986; Dion, Earn, & Yec, 1978; Miller, Boye, & Gerard. 1968 as cited in Dion et al., 1978)...

In the late 1960s, Grier and Cobbs (1968) proposed that African Americans have a “healthy cultural paranoia.” Following this characterization, researchers have examined the tendency for African Americans to distrust and be suspicious of European Americans (Terrell & Tenell, 1981; Thompson, Nevjlle, Weathers, Poston, & Atkinson, 1990). This distrust (or “racism reaction”) is thought to stem from feelings of threat from European Americans (Thompson et al., 1990). While the distrust may be justifiable. the disruption that results from mistrust can hinder the formation of specific relationships, even when there is primary importance placed en the relationship between two individuals. For example, African American individuals who have high mistrust levels are more likely to expect their European American counselors to be less accepting, trustworthy, credible, satisfactory, and more likely to expect less help with general anxiety, shyness, inferiority feelings, and dating difficulties (Nickerson, Helms, & Terrell. 1994; Watkins & Terrell. 1988; Watkins, Terrell, Miller. & Terrell, 1989). Cultural mistrust can also affect the counseling process by affecting the amount of disclosure during a counseling session (Thompson. Worthington. & Atkinson. 1994), possibly leading to self-fulfilling prophecy combined with a confirmatory bias. These findings from the counseling literature may also apply to other forms of interactions (e.g.. Kleck & Strenta, 1980). For example, research indicates that stigmatized individuals’ expectations about how others will treat them can lead them to perceive unfavorable treatment even when none is given (Kieck & Strenta, 1980).

Second, false alarms can be associated with behavioral restriction. One way to manage the perceived presence of prejudice is to structure one’s life to decrease the likelihood of encountering it (Swim et al., this volume). While complete avoidance is unattainable for the most part (Simpson & Yinger. 1985), targets can make choices about when (or when not) to enter particular situations or interactions. For instance, women (and not men ) are likely to prefer to change groups and gender composition of groups when they anticipate being the solo member of their gender in the group and this preference is related to women’s perception that they will he treated stereotypically (Cohen and Swim. 1995). Similarly, distrust of European Americans may lead African Americans to terminate employment (Terrell & Terrell, 1981) and prematurely terminate counseling with European American counselors (Terrell & Terrell, 1984). Also, reduced numbers of African American applicants to jobs has been attributed to a desire to avoid the rejection and interpersonal stress that results from prejudice from European American employers (Pettigrew & Martin. 1987).

The behavioral restriction that results from such avoidance has costs (Stangor & Sechrist, this volume). Descriptive research from counseling psychology illustrates that avoidance has potential costs for African Americans (Pinderhughes, 1982). Past experience with misdiagnoses by clinicians and intrusiveness of social service workers has lead many African American families to avoid seeking mental health services. As Biafora, Warheit, Zimmerman, Apospori, and Taylor (1993) note. “While racial mistrust may provide an adaptive coping mechanism for some individuals, it could also be hypothesized that mistrust may be maladaptive for others in that it may motivate them to withdraw from activities that are essential if they are to access the opportunity and reward structures of the dominant society—for example, school completion and/or seeking employment” (p. 894).

Third, false alarms are associated with anxiety (Mathews & MacLeod, 1994). If targets of prejudice believe that they are going to be evaluated in terms of their social group rather than on their own merits, anxiety may result and interfere with their performance (see Steele & Aronson, 1995; Aronson, Quinn, & Spencer, this volume). For example, women can be distracted when they are solo members of their gender in a group, and this interferes with their work even when they are not treated differently from other group members (Lord & Saenz. 1985; Sacnz, 1984): this occurs particularly when women are socially worried and believe they are being scrutinized (Lord, Saenz, & Godfrey, 1987). Thus, the anxiety associated with anticipating a threat can interfere with task performance...

Behavioral restriction means that people are less likely to enter situations where their beliefs will be discomfirmed. As Pettigrew and Martin (1987) note with regard to avoiding prejudice. “Avoidance learning reduces the possibility of experiencing corrective situations, such as acceptance and positive interaction” (p. 54).

Second, schematic processing, such as interpreting ambiguous information in line with one’s beliefs or focusing on confirming rather than disconfirming evidence. may make it difficult to change decision criteria. Confirmatory biases can decrease the likelihood that people will notice changes in the occurrence of prejudice and discrimination. For instance, stigmatized individuals believe that people will treat them unfavorably even when there is no evidence of negative behavior (Kleck & Strenta, 1980). In general, the research indicating that people tend to maintain their stereotypes, despite disconfirming evidence, suggests that people who hold stereotypes about perpetrators of prejudice will do the same (Baron et al., 1991; Inman & Baron, 1996; Rettew. Billman. & Davis. 1993).

Third, the emotional currency of encountering a miss or false alarm can make it difficult for people to stop using a positive illusion or zero-miss strategy, respectively... As Pettigrew and Martin (1987) note, “...because personal and vicarious experiences as a victim of prejudice and discrimination are highly emotional, this avoidance learning is deeply emotional—and emotional condition has an extremely slow extinction curve (Solomon. 1964). For these reasons, negative black responses to recruitment efforts are often especially resistant to change” (p. 54)."

--- Appraisals of Prejudice and Discrimination / Lisa Feldman Barrett, Janet K. Swim
in: Prejudice: The Target's Perspective


Addendum: See also - Discrimination: A Self-Fulfilling Prophecy