Strategies for Dealing with Mental Contamination

Strategies for Dealing with Mental Contamination in Everyday Life

Exposure control is sometimes possible and can be considered to be a preventive strategy. As professors, we might choose to grade student papers without knowing the identity of each student so as to avoid the halo effect. Similar strategies are occasionally feasible in the realm of employment. When many professional orchestras audition new musicians, they ask the musician to play behind a screen, so that the decision maker is unaware of the person's gender, age, race, and any other potentially biasing information.

What if exposure control is not possible or is not desirable? Although we remain pessimistic about people's abilities to detect and correct for contamination once it enters their minds, we offer three suggestions for strategies that can help.

Examining the Test–Retest Reliability of One’s Judgments.

When it comes to debiasing a judgment of a specific stimulus, it can be useful to observe one's reactions over time under different circumstances – essentially, to observe the test–retest reliability of the evaluation. Doing so may allow one to tease apart whether the evaluation is a reliable reaction to the stimulus or a spurious, contaminated reaction to something about the circumstances under which it is evaluated.

Consider an example: One of us adopts such an approach when grading students' final papers. Because this activity takes place over the course of several days, she knows that her judgments might be affected by order and contrast effects as well as various situational factors (e.g., whether she reads a paper at the office between appointments or at home in front of the fireplace with a glass of wine in hand and classical music in the background). Accordingly, after reading each paper, she writes a tentative grade and some comments on a separate sheet. Then, after reading all of the papers, she looks at each paper again and gives it another grade. If the two grades she assigns do not match, she suspects contamination and gives the paper further scrutiny (e.g., she compares it to other papers in the same grade ranges).

Such a strategy is not perfect, of course. Although it can help us detect contamination, it does not tell us how to correct for it. The professor who determines that her evaluations were biased by the setting must still decide whether the paper grades were artificially inflated by the fireplace, music, and wine, or whether they were artificially deflated by the hectic, noisy office. Moreover, the test–retest strategy does not allow people to tease apart contaminants that are present at both times. For example, we have a friend who has a closet full of clothes she never wears that she ordered from a catalog. Each time she looks at the catalog she loves the way the clothes look, but she underestimates how much this is due to the fact that beautiful models are wearing them. When the clothes arrive, she is invariably disappointed at how they look on her.

Examining the Covariation of One’s Responses with Potential Contaminants.

It can be useful to examine our general patterns of response to different subgroups of stimuli. A faculty colleague of ours, for instance, once noted a suspicious trend in his evaluations of student papers: Heavy smokers tended to receive lower grades than nonsmokers. As it turned out, the papers that reeked of stale cigarette smoke gave him a headache as he read them, which then caused him to evaluate the work more negatively. The test–retest strategy would not have helped him here (because smokers' papers reliably reeked and reliably elicited headaches), yet by comparing his evaluations of smokers' and nonsmokers' work, he was able to identify a contaminant.

This strategy can be useful in the workplace. Recall Ms. Green from the prestigious law firm. If she notices that, in general, her reactions to African-American candidates are less favorable than her reactions to European-American candidates, this should give her reason to pause. Indeed, this is precisely the logic behind some affirmative action guidelines. Employers are required to keep records of group-related employment patterns (e.g., the percentages of women and minorities who apply, are interviewed, are hired) so that group-linked disparities can be more readily detected (Crosby & Cordova, 1996). The existence of group disparities does not prove that discrimination has occurred, but it should prompt further investigation by the employer. In other words, even though individual contaminated judgments are unscented, certain patterns of judgment do smell "fishy," thereby alerting us to the possibility of unwanted bias.

Once again, we must point out that this strategy is limited. Perceivers are not terribly adept at covariation assessment (Nisbett & Ross, 1980), and therefore may fail to detect suspicious patterns. When they do succeed, the patterns may be open to multiple interpretations and the implications of global trends for individual cases may be unclear.

Considering the Opposite.

Some kinds of mental biases can be undone by asking people to "consider the opposite," whereby they imagine different outcomes than the ones they had been thinking about. Simply thinking about whether a specific hypothesis is true, for example, increases people's beliefs in its validity because the person focuses more on explanations as to why it could be true than why it could be false. Several studies have found that asking people to consider the possibility that the opposite hypothesis is true is sufficient to undo the bias of one-sided thinking (Lord, Lepper, & Preston, 1984).

For this strategy to work, however, people must be able to mentally simulate alternative outcomes or hypotheses (Hirt & Markman, 1995). In many everyday cases of contamination, people may be so locked into one way of looking at something that it is difficult for them to imagine alternatives. This is especially likely to be true, we suspect, when people have contaminated affective reactions. The affect may anchor their judgments to the extent that it is difficult for them to imagine they could feel otherwise. Suppose Ms. Green tried to imagine how she would have evaluated Mr. Jones if he were White. Her negative feelings might anchor her thinking to the point where it is difficult to perform this mental simulation.

Furthermore, people do not always know what aspect of a situation to mentally undo. Consider the men in the Dutton and Aron (1974) experiment who were approached by the woman on the scary bridge. Presumably, they had little idea that they were misattributing fear to attraction to the woman. Even if it occurred to them to consider the opposite, they might not know which aspect of the situation to transform. Should they imagine that they had met the woman at a different time of day or season of the year? Without some idea about the source of contamination, people do not know which mental simulation to run.