The Cognitive Roots of Stereotyping

Adam J. Oliner

Massachusetts Institute of Technology

19 October, 2000

Revision 2

Table of Contents



Social Perception and Memory

Social Consequences




Perhaps one of the human mind’s most complex functions is the governing of a person’s social behavior. One aspect of this function is the task of storing and retrieving information about other people in order to interact with them or predict their behavior. The brain has developed organizational mechanisms that group data based on similarities. While this storage system assists with organization and use of the stored information, it can have the unwanted side effect of associating characteristics with subjects for whom the association may be inaccurate. This generalization, specifically as it applies to humans, is known as stereotyping. Recently, society’s awareness of, and sensitivity to, social discrimination based on stereotypes has become heightened. Yet, if stereotyping is, indeed, a result of subconscious cognitive functions, what can the individual or the society do to counter discrimination?

The Cognitive Roots of Stereotyping

Certainly, human social interaction has been a major subject of psychological research. The brain has developed mechanisms that help humans interact effectively and efficiently. Some of these mechanisms serve to permit one person to predict a target person’s behaviors or characteristics. These predictions are often made based on physical characteristics and the social group in which the target implicitly belongs. This is commonly called stereotyping. “A stereotype is a socially shared set of beliefs about traits that are characteristic of members of a social category” (Greenwald & Banaji, 1995, p. 14). Unfortunately, these stereotypes do not always accurately reflect the individual. This can result in discrimination, sexism, and racism. What therefore, can a society interested in eliminating this behavior do to counter ingrained cognitive functions?

Social Perception and Memory


Hirschfeld (1996) writes, “the prevalent point of view in psychology is that [stereotypical] thought is a by-product of the way information is organized and processed” (p. 8). He goes on to assert that information is processed into categories in order to both reduce the sheer quantity of information and “extend our knowledge of the world by capturing nonobvious similarities [among] their members.” Information about objects (including people) is placed into these categories based on the similarities between the object and the potential categories. One appealing theory is that, if the object and category have a high enough value of shared characteristics, it is likely the object will be placed in that category (Vosniadou & Ortony, 1989). These are not quantitative theories on cognitive categorization; they are qualitative statements about human information organization to which many studies make reference.

“People are perhaps the most complex stimuli we encounter, in part because they simultaneously belong to multiple social categories (e.g., age group, ethnicity, sex)” (Macrae, Bodenhausen, Milne, 1995, p. 397). It turns out that, even beyond the unscientific concept of similarity, selecting the most appropriate category can be complicated. Macrae et al. discussed studies in which the underlying mechanisms of category dominance were sought. The studies they described involved subconsciously priming subjects for either the category “woman” or “Chinese” through what was described to the subjects as a vigilance task. They were then asked to perform an ostensibly unrelated task of categorizing a string of characters as words. The subjects were shown words that fit into one of three categories. The first was a set of control words (the names of countries, etc.). The second and third sets were words which either fit the stereotypical categories of woman or Chinese, respectively, but were stereotypically neutral for the other. For example, the words “thoughtful, friendly, and emotional” were considered typical for the category woman but neutral for the category Chinese. The studies anticipated that the subjects would be faster in recognizing those words that were stereotypical of the primed category. In fact, they discovered that the subjects also actively inhibited the words which were stereotypical of the unprimed category. Their results provide evidence that “both excitatory and inhibitory mechanisms” (p. 400) are at work in the categorization process.

Once information is stored in categories it can be used to make predictions and inferences about new category members. For example, if a person has seen an orange basketball before, they have already placed this object into a basketball category consisting of several characteristics; it bounces, has a certain texture and groove design, is spherical, and is orange. If that person then encounters a black basketball, they might be able to still associate it with the category of basketballs based on its texture and shape. Once this association has been made, the person can presume that the black basketball will also bounce. This metaphor can be extended to, for instance, carnivorous jungle cats and their propensity for devouring humans; the evolutionary benefits of categorization become clear. The ability to make these inferences is behind stereotyping. “Stereotypes guide judgment and action to the extent that a person acts toward another as if the other possesses the traits included in the stereotype” (Greenwald & Banaji, 1995, p. 14).

Argument exists regarding the degree to which categorization is carried out automatically, as a cognitive process, versus being within the subject’s control. “A long tradition has conceived of stereotyping and prejudice as an automatic and inevitable consequence of categorization…Specifically, people’s memberships in fundamental categories such as age, gender, and race seem to be attended to automatically” (Lepore & Brown, 1997, p. 275). In contrast, Macrae et al. (1995) contends that “target categorization [is] more flexible…we acknowledge that it can, under certain circumstances, be under the perceiver’s executive control” (p. 405). The latter performed studies that involved the subconscious influencing of category formation. Subjects were primed either for the category “woman” or “Chinese.” They were then shown a video that included a Chinese woman. Finally, they measured “the accessibility of applicable stereotypes (e.g., considerate, gracious, calm for Chinese) about the target” (Macrae et al., 1995, p. 399). They discovered that the subconsciously primed category was more accessible relative to a control (again, countries of the world). This argues for the implicit formation of categories. That is, categories can be created without the attention or intent of the subject. However, this only applies to when the priming was subconscious. The question remains whether attention paid to the categorization can affect the outcome. In other words, can a person explicitly select a category, overriding natural tendencies?

Factors affecting retrieval

Once the category has been formed and the information thus stored, the person can access that information from the context of the category. Retrieval of the categorized data, the act of stereotyping, can be investigated in a way similar to what was done for categorization. Is the activation of a stereotype inevitable? What factors affect prejudice? Milne, Macrae, Bodenhausen, Thorn, & Castelli (1997) define automatic versus controlled processes as follows:

Automatic processes are mental activities that occur outside of awareness in a largely involuntary, unintentional, and effortless manner. Controlled processes, in contrast, are activities that possess precisely the opposite qualities: they are intentional, controllable, effortful, and are implemented in the presence of conscious awareness (p. 472).

These two opposite definitions seem to suggest that a process must be one or the other. The evidence seems to suggest that this is not the case. Indeed, it seems that, like categorization, stereotyping occurs automatically, but can be controlled if given proper attention.

Automatic stereotype activation

Lepore and Brown (1997) describe automatic stereotype activation as the activation of categorically associated “nodes.” “When encountering a category member the group node is activated, and the excitation spreads from it to other connected nodes…” (p. 275). These excited nodes are the stereotypic characteristics. What is significant about the definition is the totally involuntary nature of the process of stereotype activation. According to Greenwald and Banaji (1995), “recent reviewers have very effectively documented the unconscious or automatic operation of stereotypes” (p. 4). Stereotypes can be activated by single words, and trait inferences were spontaneously formed from behavior-describing sentences. A related study found “that imposing time pressure on a judgment task…increased the level of ethnic stereotyping in subjects’ judgments” (Greenwald & Banaji, 1995, 18). With more complicated tasks, where subjects have an opportunity to evaluate stereotypes, automatic stereotype activation is not usually observed. “Whereas the mere presentation of words appears sufficient to activate stereotypic associates, the same cannot be said of persons” (Milne et al., 1997, 486). But, is this exception the result of natural “node wiring,” or a product of conscious effort?

“Automatic reactions can be modulated by attention and intention; they can be inhibited and suppressed; and they can be coherent and [planned]” (Milne et al., 1997, 473). Milne also refers to an experiment by Gilbert and Hixon (1991) that demonstrated stereotype activation primarily when subjects were depleted of their cognitive resources (such as through distraction or fatigue). Subjects were exposed to a video of an Asian card dealer, and subsequently the activation of stereotypes was measured. Results show convincingly that a person whose cognitive resources are decreased are equally able to identify the ethnicity of the card dealer, but are unable to access stereotypes associated with the category. This argues that stereotype activation is not absolutely automatic. “Rather, it is conditional upon the availability of sufficient attentional resources for perceivers to locate and retrieve stereotypical material from its residence in the mind” (Milne et al., 1997). That is, stereotypes are activated by the mind as a secondary process to garner more information about a target from categories; they do not happen as part of the primary recognition process.

Social Consequences


There are many consequences associated with categories and stereotyping. The benefits of this cognitive organizational technique have been discussed. The negative consequences, however, beg further investigation. One such problem is discrimination. Discrimination is the judgment of an individual based on the categorical characteristics associated with the target’s social or ethnic group. When these category attributes do not fit the target, Hirschfeld (1996) calls this “illusory correlation.” Many studies have shown the effects of discrimination on judgment.

Some of the most dire consequences of discrimination can be seen in the setting of the legal system. Bodenhausen (1988) presents results from a study in which subjects were given evidence about a supposed crime in order to decide guilt. “These subjects judged the defendant to be more likely to be guilty when he belonged to an ethnic group for which the crime was stereotypic than when he was ethnically nondescript” (p. 727). With a stereotyped defendant, subjects were even more likely to recall incriminating evidence. They also judged neutral evidence more harshly. Similar behavior has been found in the context of actual criminal cases.

The halo effect is a tendency for positive characteristics to be associated with other positive characteristics. “In much halo effect research, physical attractiveness plays the role of the objectively irrelevant attribute that influences evaluative judgment on various other dimensions” (Greenwald & Banaji, 1995, 9). There have been many studies showing this halo effect. They are summarized nicely by Greenwald and Banaji (1995) below:

…attractive males and females are judged to be kinder, more interesting, more sociable, happier, stronger, of better character, and more likely to hold prestigious jobs. Similarly,…essays attributed to a female student were judged by male students to be of higher quality when the stimulus materials included a photo that showed the author to be physically attractive…defendant attractiveness was associated with judges levying smaller fines and lower bail levels in actual misdemeanor cases (p. 9)

Appearance discrimination, just like racial or gender discrimination, is a result of illusory correlation.

Countering discrimination

Evidence presented in this paper has shown stereotyping to be a controllable cognitive process. Consequently, discrimination should also be under the subject‘s control. “The finding that implicit cognitive effects are often reduced by focusing judges’ attention on their judgment task provides a basis for evaluating applications…aimed at reducing such unintended discrimination” (Greenwald & Banaji, 1995, 4). Greenwald and Banaji (1995) go on to suggest three potential ways of dealing with discrimination, namely: blinding, consciousness raising, and affirmative action.

The idea of blinding is to provide no cues whatsoever as to the social category or stereotype under which the target might fit. If no revealing attributes are present, it seems impossible for discrimination to occur. Of the three solutions, this one is infallible. Where this technique can be applied, it should be. Unfortunately, it is practically difficult to shield all stigmatizing characteristics from a judge.

Affirmative action is the process by which the negative effects of discrimination are proactively countered. Those traits that were considered negative become desirable under an affirmative action plan. This method attempts to compensate not only for past damages, but also for potential future discrimination. It assumes an inability to prevent the stereotyping in the first place.

There are connections that cause nodes corresponding with stereotypic traits to be activated when a group node is triggered (Lepore and Brown, 1997). These links can be weakened by conscious attention. As the cognitive connection between a group and a stereotypic trait is decreased, the chances of unintentional discrimination are also lessened. Awareness of illusory correlations and potential discrimination can allow conscious prevention.


Each method for dealing with discrimination is appropriate in certain situations. It is important, though, for society to be aware of the possibility and causes of implicit stereotyping. Only by understanding the source of the problem can society hope to solve it. Stereotyping occurs as a consequence of categorization. The categories themselves are a cognitive organizational structure for information. Categories allow humans to better participate in complex tasks like social interaction. This social interaction is perhaps one of the most intricate and high-order tasks undertaken by the brain.


Bargh, J. A., & Uleman, J. S. (1989). ). Unintended thought. New York, NY: The Guilford Press.

Bodenhausen, G. V. (1988). Stereotypic biases in social decision making and memory: Testing process models of stereotype use. Journal of Personality and Social Psychology, 55 (5), 726-737.

Gilbert, D. T., & Hixon, J. G. (1991). The trouble of thinking: Activation and application of stereotypic beliefs. Journal of Personality and Social Psychology, 60, 509-517.

Greenwald, A. G., & Banaji, M. R. (1995). Implicit social cognition: Attitudes, self-esteem, and stereotypes. Psychological Review, 102 (1), 4-27.

Hirschfeld, L. A. (1996). Race in the making. Cambridge, Massachusetts: The MIT Press.

Lepore, L., & Brown R. (1997). Category and stereotype activation: Is prejudice inevitable? Journal of Personality and Social Psychology, 72 (2), 275-287.

Macrae, C. N., & Bodenhausen, G. V., & Milne, A. B. (1995). The dissection of selection in person perception: Inhibitory processes in social stereotyping. Journal of Personality and Social Psychology, 69 (3), 397-407.

Milne, A. B., & Macrae, C. N., & Bodenhausen, G. V., & Thorn, T. M. J., & Castelli, L. (1997). On the activation of social stereotypes: The moderating role of processing objectives. Journal of Experimental Social Psychology, 33, 471-489.

Vosniadou, S., & Ortony, A. (1989). Similarity and analogical reasoning. New York, NY: Cambridge University Press.

© Adam Oliner 2001