Journal of Personality and Social Psychology2001, Vol. 81, No. 5, 828-841Copyright 2001 by the American Psychological Association, Inc.0022-3514/01/S5.00 DOI: I0.1037//0022-3514.81.5.828Imagining Stereotypes Away: The Moderation of Implicit StereotypesThrough Mental ImageryIrene V. Blair, Jennifer E. Ma, and Alison P. LentonUniversity of Colorado at BoulderResearch on implicit stereotypes has raised important questions about an individual's ability to moderateand control stereotypic responses. With few strategies shown to be effective in moderating impliciteffects, the present research investigates a new strategy based on focused mental imagery. Across 5experiments, participants who engaged in counterstereotypic mental imagery produced substantiallyweaker implicit stereotypes compared with participants who engaged in neutral, stereotypic, or no mentalimagery. This reduction was demonstrated with a variety of measures, eliminating explanations based onresponse suppression or shifts in response criterion. Instead, the results suggest that implicit stereotypesare malleable, and that controlled processes, such as mental imagery, may influence the stereotypingprocess at its early as well as later stages.Implicit stereotypes are social category associations that becomeactivated without the perceiver's intention or awareness when heor she is presented with a category cue. Over the past decade,substantial evidence has accumulated for the influence of implicitstereotypes on judgment and behavior (for reviews, see Bargh,1999; Blair, 2001; Greenwald & Banaji, 1995). With that work hascome concern over the controllability of such effects and theimplications thereof (e.g., Banaji & Greenwald, 1994; Bargh,1999; Devine, 1989; Fiske, 1989). If an individual is unaware ofthe stereotype's influence or does not have the means to control it,is stereotyping inevitable? Should individuals and organizations beheld legally responsible for discrimination that results from im-plicit stereotypes? At the core of these questions is the issue ofcontrol. Can people control the influence and expression of im-plicit stereotypes?One answer to this question is provided by theories of stereo-typing that specify very distinct roles for implicit and explicitprocesses (see Bargh, 1999; Bodenhausen & Macrae, 1998;Brewer, 1988; Devine, 1989; Fiske & Neuberg, 1990). Accordingto those theories, stereotyping begins with the activation of im-plicit stereotypes and ends with their application to judgment orbehavior. If a perceiver is motivated to avoid stereotyping, he orIrene V. Blair, Jennifer E. Ma, and Alison P. Lenton, Department ofPsychology, University of Colorado at Boulder.Jennifer E. Ma is now at the Department of Psychology, University ofKansas.Implementation of Multicultural Perspectives and Approaches in Re-search and Teaching (IMPART) grants from the University of Coloradogreatly facilitated the research and writing of this article. We thank Chris-tine Johnson and Olympia Yearwood for their assistance with the research,and we are grateful to researchers at the University of Colorado Stereo-typing and Prejudice (CUSP) laboratory for their comments on early draftsof this article.Correspondence concerning this article should be addressed to Irene V.Blair, University of Colorado, Department of Psychology, MuenzingerPsychology Building, Boulder, Colorado 80309-0345. Electronic mail maybe sent to [email protected] may control the application of stereotypes by suppressingthem, compensating for their influence, or concentrating on indi-viduating information (Devine & Monteith, 1999; Dunton & Fa-zio, 1997; Fazio, Jackson, Dunton, & Williams, 1995; Fiske, 1989;Plant & Devine, 1998). What the perceiver is not able to do,according to current theory, is alter the activation of implicitstereotypes.A different answer to the question of control is provided byrecent research that suggests people can moderate the activation ofimplicit stereotypes as well as control their later application, giventhe right strategies and conditions (e.g., Blair & Banaji, 1996;Gilbert & ffixon, 1991; Gollwitzer & Schaal, 1998; Macrae,Bodenhausen, Milne, Thorn, & Castelli, 1997). For example, Blairand Banaji (1996) demonstrated that a counterstereotype (CS)expectancy produced significantly weaker implicit stereotypesthan a stereotype expectancy, especially under reduced cognitiveload. However, the work supporting the moderation of implicitstereotypes has been criticized as either internally or externallyinvalid (Bargh, 1999).The goal of the present research is to investigate mental imageryas a new strategy to moderate implicit stereotypes. Mental imageryis the conscious and intentional act of creating a mental represen-tation of a person, object, or event by seeing it with the "mind'seye." We focus on mental imagery for several reasons. First,research over the past several decades has shown that mentalimagery has many of the same characteristics as a real experience,including concrete details, causal sequences, logical constraints,concomitant emotional arousal, and similar neurological charac-teristics (Dadds, Bovbjerg, Redd, & Cutmore, 1997; Kosslyn,1994, 1995; Taylor & Schneider, 1989). As a consequence, mentalimagery has a more powerful impact on learning, decision making,and behavior compared with other methods of processing infor-mation (Bower, 1972; Gregory, Cialdini, & Carpenter, 1982;Paivio, 1971; Pham & Taylor, 1999; Taylor, Pham, Rivkin, &Armor, 1998).Second, mental imagery affects the cognitive system in a man-ner that makes it a theoretically plausible strategy to moderateimplicit processes. In particular, mental imagery increases the828SPECIAL SECTION: IMAGINING STEREOTYPES AWAY829accessibility of related cognitive, emotional, and behavioral rep-resentations (Carroll, 1978; Johnson & Sherman, 1990; Strack,Schwarz, & Gschneidinger, 1985). For example, when one imag-ines an airplane crash, knowledge of recent accidents, the potentialcauses and consequences of such an accident, and one's emotionaland behavioral readiness for the accident all become accessible.When asked about the future likelihood of such an event, theincreased accessibility of these constructs produces an exaggeratedestimate of its probability (Carroll, 1978; Gregory et al., 1982;Sherman, Cialdini, Schwartzman, & Reynolds, 1985; Sherman,Skov, Hervitz, & Stock, 1981). In this manner, mental imagery canfunction much like other sources of priming (see Bargh,
View Full Document
Unlocking...