 |
Assimilation and Contrast
An intriguing and important consequence of categorical thinking is its tendency to distort perceptions. Typically, these distortions take the form of minimizing differences within categories ("assimilation") and exaggerating differences between categories ("contrast"). For example, when Joachim Krueger and Russell Clement (1994) asked people to estimate several daily temperatures for a nearby city, they found a smaller gap between temperature estimates for November 15th and November 23rd (dates within the category "November") than between November 30th and December 8th (dates from two different months). Both time intervals spanned eight days, and in reality the city's temperature change was not greater in the latter case than the former -- it simply seemed greater because December temperatures are, on average, different than November temperatures.
In this connection, Myron Rothbart and his colleagues (1997) tell an old Yiddish story of a peasant whose farm was located near the border of Poland and Russia, where boundary markers shifted with every international dispute:
The peasant did not know from one year to the next whether his farm was in Russia or Poland, and eventually hired a surveyor to resolve the uncertainty. After weeks of painstaking assessment, the surveyor finally announced that the farm was just inside the Polish border. "Thank God," the peasant cried with relief, "now I won't have to endure any more Russian winters!" (Rothbart, Davis-Stitt, & Hill, 1997, p. 123).
Humor aside, assimilation and contrast effects have been observed in a wide variety of domains, including estimates of line length, judgments of speech sounds, impressions of faces, and evaluations of attitudes (Brown, 1995; Tajfel & Wilkes, 1963). Robert Goldstone (1995) even found an assimilation effect in color perception. In this study, students were shown a random series of letters and numbers that ranged in color from very red to very violet. Results showed that even when a letter and number had exactly the same hue, students rated the letter as being similar in color to other letters, and the number as being similar in color to other numbers (e.g., in the diagram below, they saw the "L" as redder than the identically-colored "8").
| Red |
Violet |
Figure 2. In a study on color perception, Robert Goldstone (1995) found that the "L" above was
perceived as more red than the "8," even though the "L" and "8" were actually identical
in hue. Figure reprinted with permission of Blackwell Publishing, Ltd.
|
With respect to prejudice, the implication of this research is that differences within groups will tend to be minimized and differences between groups will tend to be exaggerated. Moreover, if these differences are consistent with well-known stereotypes, the distortion in perception may be highly resistant to change. In one study, for example, participants were unable to break free of gender stereotypes even when encouraged to do so (Nelson, Biernat, & Manis, 1990). In this experiment, people were asked to judge the height of various men and women from a series of photographs. Each photograph showed only one person, and participants were told:
In this booklet, the men and women are actually of equal height. We have taken care to match the heights of the men and women pictured. That is, for every woman of a particular height, somewhere in the booklet there is also a man of that same height. Therefore, in order to make as accurate a height judgment as possible, try to judge each photograph as an individual case; do not rely on the person's sex. (p. 669)
Despite these instructions and a $50 cash prize for the person who made the most accurate judgments, people perceived the males to be, on average, a few inches taller than the females. In other words, they were either unable or unwilling to disregard the categories "male" and "female," and the perception of men as taller than women prevailed.
 |
Page 5 of 27 |
 |
|
|
