Understanding Prejudice
Understanding Prejudice
Return Home

Reading Room

Exercises and Demonstrations
Multimedia Center
Teacher's Corner
Directory of Experts
Links on Prejudice
About Us
Privacy Policy
Contact Us

Reading Room
The Psychology of Prejudice: An Overview

Stereotypes from Direct Experience

Stereotypes are learned not only from the mass media, but from direct experience as well. Although some stereotypes are grounded in truth (e.g., it is true that men are, on average, more physically aggressive than women), many are distortions that arise from otherwise adaptive modes of thought. To illustrate, try the following exercise: Look around you for 5 or 10 seconds and make a note of what is in your environment. Then, after you have carefully observed your surroundings, close your eyes and recall everything that you noticed. Do not read further until you have taken a few moments to give this exercise a try.

What did you recall seeing? If you are like most people, the items you noticed were the most salient things in the environment -- objects that were prominent, large, colorful, or attention-getting in some way. When we observe the environment, we do not give equal weight to every element; instead, we are highly selective. Without even being aware of it, we automatically filter what we see in a way that gives greatest weight to whatever is most salient.

Normally, this kind of automatic filtering is highly beneficial. After all, which is more important to notice -- an oncoming car or a pebble along the side of the road? Just as with categorical thinking, our focus on salient stimuli allows us to process a large amount of information efficiently. Yet also like categorical thinking, our focus on salient stimuli can lead to systematic distortions in perception, and, at times, to prejudice and stereotyping.

An experiment by Loren Chapman (1967) shows how salience can distort the judgments people make. Chapman projected a series of word pairs, such as bacon-tiger, onto a screen in front of the participants in his study. For example, in a typical series, the word on the left side of the screen was bacon, lion, blossoms, or boat, and the word on the right side was eggs, tiger, or notebook. Chapman balanced the word pairs so that each left-side word appeared an equal number of times with each right-side word, yet he found that when participants were asked to estimate the frequency of various word pairs, they reported seeing illusory correlations. For instance, people estimated that when bacon appeared on the left, eggs was paired with it an average of 47% of the time. Similarly, participants thought that when lion was on the left, tiger was the word that appeared with it most often.

Although illusory correlations can occur for a variety of reasons, one key element is that distinctive pairings are better remembered than other pairings (Hamilton, Dugan, & Trolier, 1985; Mullen & Johnson, 1990). In the case of Chapman's research, certain word pairs stood out because the two words were thematically related. Yet distinctiveness also increases when rare events or attributes are paired with one another -- a result that can sometimes lead to stereotyping.

This connection was illustrated in an experiment that presented people with brief statements describing the behavior of individuals from one of two groups: "Group A" or "Group B" (Hamilton & Gifford, 1976). Group A had twice as many members as Group B, but the proportion of desirable and undesirable behaviors represented in the statements was the same within each group. Roughly 70% of the time the statements described a desirable behavior (e.g., "visited a sick friend in the hospital"), and roughly 30% of the time the statements described an undesirable behavior (e.g., "always talks about himself and his problems"). In other words, the most infrequent -- and hence, most distinctive -- statements described undesirable behaviors on the part of the minority group (Group B).

Under these conditions, people significantly overestimated the frequency of undesirable minority behaviors. As shown by the bolded entries in Table 4, participants recalled 52% of undesirable behaviors as coming from Group B, even though the actual percentage was only 33%. Furthermore, subsequent research has shown that this kind of illusory correlation is especially pronounced when the distinctive pairings involve negative behavior and are consistent with preexisting stereotypes (Hamilton & Rose, 1980; Mullen & Johnson, 1990). In such instances, the salience of unusual pairings can strongly reinforce minority stereotypes.

Table 4. An Example of Illusory Correlation

 Statement Content Group A Group B Total
  Actual Distribution of Statements  
 Desirable Behaviors 18 (67%) 9 (33%) 27 (100%)
 Undesirable Behaviors 8 (67%) 4 (33%) 12 (100%)
  Perceived Distribution of Statements  
 Desirable Behaviors 17.5 (65%) 9.5 (35%) 27 (100%)
 Undesirable Behaviors 5.8 (48%) 6.2 (52%) 12 (100%)

Note: This table is based on data from a study by Hamilton and Gifford (1976). Even though only 4 out of 12 undesirable behavior statements involved Group B (the minority group), participants later recalled more of the undesirable behaviors coming from Group B (average = 6.2) than Group A (average = 5.8).

Previous page
Page 17 of 27
Next page