Skip to main content

Verified by Psychology Today


The Opposite of Empathy

How projection became empathic accuracy.

Wikimedia Commons
Toni Frissell
Source: Wikimedia Commons

When the young psychologist Rosalind Dymond arrived at Cornell University in 1946, she set out to design some of the first experiments to measure empathy. Dymond began by assessing empathy with the Thematic Apperception Test (TAT)—a set of cards depicting images of archetypal personalities and dramatic scenes created by psychologist Henry Murray and artist Christiana Morgan. Subjects scrutinized the images and told stories about the figures in the pictures. These stories were frequently drawn from the subject’s own experiences.

In these early experiments, Dymond judged a subject’s capacity for empathy by evaluating the kinds of stories they told. Stories were rated “good” if they described the thoughts and feelings of the person depicted in the image; “fair” if the accounts only touched on the person’s external characteristics; and “poor” if they simply named the figures [1]. Subjects who offered good descriptions were deemed to have empathy, and interviews revealed that these subjects also had greater insight into their own relationships. This characterization of empathy as the ability to tell in-depth, imaginative stories of another’s feelings and circumstances was closely tied to empathy’s early aesthetic meaning .

In later experiments, however, Dymond sought to measure empathy between individuals interacting with one another. She divided fifty-three social psychology students into small groups that met three times so students could get to know each other. Each person ascribed six personality traits to another member of their group, and then judged which personality traits the other person would attribute to themselves. Dymond now defined empathy as the ability to accurately predict how another person saw themselves. Students were empathic if their predictions closely corresponded to the ratings other students gave themselves [2].

Empathy was no longer a matter of making up intricate stories but of correctly predicting another’s response. Dymond's revised definition of empathy appeared in her 1952 paper as: “the imaginative and accurate transposing of oneself into the thinking, feeling and acting of another” [3]. As empathy transformed into an accurate appraisal of how someone else felt and thought, its links to story-telling and aesthetic projection faded.

The psychologists Irving Bender and Alfred Hastorf at Dartmouth College extended Dymond’s pioneering experiments to discover that undergraduates were quite poor at predicting their friends’ responses on personality scales. Students tended to project their own feelings into their forecasts of others’ responses. Projection comprised “the attribution to others of one’s own needs, interests, and attitudes” [4].

In one experiment, a majority of undergraduates engaged in projection by making forecasts of others' preferences that were highly correlated with their own. Only 20 out of 50 undergraduates empathized with other students by making predictions that were more closely aligned with the other students' scores than their own scores [5].

Hastorf and Bender discovered that projection was not only more common but also more intense than empathy. Projection was personal and merely referred to the self, whereas empathy was objective, cognitive and truly perceptive.

Empathy was now the opposite of projection.

The finding that many undergraduates had little empathic accuracy prompted psychologists to propose that empathy should be trained. In 1952, Dartmouth offered a new course, “Introduction in Human Relations,” aimed at increasing students’ sensitivity to the attitudes and feelings of others. The Harvard psychologist Gordon Allport expressed concern that the social sciences had fallen far behind the rapid developments made in the natural sciences. He viewed the failure to understand social relationships as an existential threat: the chances for the survival of humanity were poor, he mused, “unless we can improve mankind’s understanding and control of social and personal factors” [6].

In 1949, Dymond declared empathy to be a neglected field of study in psychology. Thirty years later the African-American psychologist Kenneth B. Clark again lamented the lack of in-depth studies of empathy [7].

Empathy is a popular topic of research today, but its cultivation is still neglected. Despite a number of recent initiatives, most schools and universities lack dedicated programs to foster empathy. We have been keen to train our intelligence and individualism in western societies but have not similarly educated our empathy or altruism, as biologist and Buddhist monk Mathieu Ricard has noted [8]. Now, at the dawn of a new decade, it is time for our social education to promote sensitivity to others' experiences.


[1] Rosalind Dymond. (1948). A preliminary investigation of the relation of insight and empathy. Journal of Consulting Psychology 2, 228-233. She published under the name R. Dymond Cartwright later in her career.

[2] R. F. Dymond. (1949). A scale for the measurement of empathic ability. Journal of Consulting Psychology 13 (2), 127-133. 128

[3] R. F. Dymond, Anne Hughes, & Virginia Raabe. (1952). Measurable changes in empathy with age. Journal of Consulting Psychology 16, 202-206, 202, my italics.

[4] A. H. Hastorf, & I. E. Bender. (1952). A caution respecting the measurement of empathic ability. The Journal of Abnormal and Social Psychology 47 (2S), 574-576, 574

[5] Bender & Hastorf. (1953). On measuring generalized empathic ability (social sensitivity). The Journal of Abnormal and Social Psychology 48 (4), 503-506.

[6] Gordon Allport, “Radcliffe Alumnae Behavioral Science, 1960 –Development of Social Relations of Harvard,” HUG 4119.50, Box 3; Folder 95, Gordon Allport Papers, Harvard University.

[7] Leonard S. Cottrell, Rosalind F. Dymond. (1949). The empathic responses: A neglected field of research. Psychiatry 12 (4), 355-359; K. B. Clark. (1980). Empathy, a neglected topic in psychological research. American Psychologist 35(2), 187-190

[8] Matthieu Ricard. (2013). Altruism. NY: Little Brown.