Professional judgments of personality are different in some ways from the everyday judgments we make of one another (see here, for background). Last week, I began looking at "expert dimensions" -- dimensions of personality that experts, in particular, notice and employ to understand others.

One such dimension is that of mental defense. Defense mechanisms represent ways of thinking that block out or otherwise manage thoughts, feelings, and motives a person may have that are so unpleasant or threatening that they are difficult to keep directly in mind. People vary in how they use such mental defenses, with some using crude approaches that so distort reality that the individual's capacity to comprehend the world is compromised. Other people are relatively adept at using defenses so that their distortions of the world around them are more subtle.

Researchers use one of several methods to classify and rank mental defenses according to how well they work. This week, I will examine the work of Professor Phebe Cramer of Williams College in Massachusetts, who has developed a measurement approach to assess defenses.

Cramer's work is somewhat different from the work I described last week by George Vaillant, but it is complementary to it and provides additional evidence for how defenses work in a person's life. In scientific lingo, the two researchers have produced "convergent" evidence in that they used similar but slightly different theoretical models, along with different measurement approaches, with different participants, and despite such variations ended up with similar findings. 

In Cramer's approach, there are three levels of defense she labels denial, projection, and identification.  Cramer thinks of these in developmental terms from least-to-most sophisticated, with the least sophisticated denial defenses emerging earlier in life and then declining to some degree, and both the mid-range projective and the high level identification defenses emerging for the first time around adolescence and continuing through adulthood.

Cramer has developed a method of measuring psychological defenses that depends on a standardized psychological assessment, as opposed to an interview or ratings from psychologists. Participants respond to six pictures from the Thematic Apperception Test. The six pictures are of men and women, by themselves, or with a second person, or with several others, and either indoors or outside. After each picture is shown, the person being assessed tells a story about the character(s) in the picture, which is recorded verbatim.

In Cramer's method, trained coders then evaluate the participants' stories for the presence of defense mechanisms. Stories are coded for the presence of each of the three defenses. Denial is coded when there are major omissions or positive distortions of the picture. Projection is coded when anger, anxiety or other emotions and intentions are added to a picture that are not usually present. These might include magical powers, things falling apart, concerns over protection, and similar unusual perceptions. The third defense-level, identification, is scored if one character is hoping to imitate, take over or learn the skills or qualities of another person.

Some examples can make this clearer. Imagine that a study-participant, Jonathan, is shown a picture of two men with a gun in the woods. As I understand the system, if Jonathan said, "Two men are in the woods enjoying the scenery..." that would be scored for denial, because Jonathan distorted the picture in a positive way and failed to incorporate the gun. If Jonathan said, "Two men are in the woods hunting and an angry wind is blowing and there is going to be a tornado in which they will be killed," that would reflect a projective defense because it involves concerns over death and the addition of a violent storm. If Jonathan's story was, "The first man is learning to hunt and the second one is teaching him how to track a deer and then to shoot it and bring it home so that they have meat to eat over the winter," that would illustrate defenses at the level of identification, because one person is learning from another, and constructive, plausible motives are assigned to the people.

Cramer has studied defenses developmentally using data from several longitudinal studies. I will try to summarize some of her key results from studies reported in 2002, 2008, and 2009 (references below). All three studies compared defense style to an indpendent criterion constructed of expert observers' impressions.  In the 2002 study, the observers were four graduate and doctoral level students in psychology who interacted with the participants for several days, engaging in informal social interactions, and interviewing the participants. The observers then described the participants' personality features using the California Q-sort, a list of 100 attributes. In the 2008 study, clinical judges (all Ph.D.'s) described the participants based on their overall interview and test records.  The 2009 study included mothers' ratings of their childrens' behaviors, including temper tantrums, lying, jealousy and irritability.

Some of the more consistent and important findings across Cramer's studies, as I read them, are these: The use of denial in five to ten-year-olds is common. During their early teen years and young adulthood, however, individuals normally reduce their use of denial and increase their use of projection and identification.  Those who continue to use denial between the ages of 11 to 13 exhibit more undercontrolled behavior than others - temper tantrums, lying, and jealousy, as rated by their mothers (2009). By about 23 years of age, men and women who use more denial than others are relatively behaviorally unstable, lacking in clear thinking, flamboyant, egotistical, and unconventional (2002). Those who employ more identification-based defenses during late adolescence exhibit better social competence than others (2008). Competence is defined as the capacity to define and maintain an identity, to achieve life goals such as doing well in school and to choose a career and exhibit good self-control.  After adolescence, a person's competence will vary somewhat, but individuals who use identification defenses appear to be better protected from steep declines in such capabilities relative to others.

Cramer's findings use somewhat different methods and a different theory than others (see here), yet arrive at findings consistent with those of other experts in the field -- the use of more mature defenses predicts better behaviors and life outcomes.

These findings suggest that defensiveness is a dimension that experts understand and that predicts important life outcomes. The few English-language terms related to defensiveness (see here) suggest that ordinary people may overlook or have trouble explicitly describing the defense mechanisms of those they encounter, and (more speculatively) this may limit the understanding ordinary people may have of one another relative to experts.


Cramer, P. (2002). Defense mechanisms, behavior, and affect in young adulthood. Journal of Personality, 70, 103-126.

Cramer, P. (2008). Identification and the development of competence: A 44-year longitudinal study from late adolescence to late middle age. Psychology and Aging, 23, 410-421.

Cramer, P. (2008). The seven pillars of defense mechanism theory. Social and Personality Psychology Compass, 2, 1963-1981. (Downloaded 12/2/2010 from

Cramer, P.(2009). An increase in early adolescent undercontrol is associated with the use of denial. Journal of Personality Assessment, 91, 331-339.

Cramer, P. (n.d.). The defense mechanism manual. Unpublished document. Williamstown, MA: Williams College.

Copyright © 2010 by John D. Mayer

Most Recent Posts from The Personality Analyst

Six Lessons Preschoolers Learn about Personality

What children learn can teach scientists about personal intelligence

Can You Tell the Cat Lovers From the Dog Lovers?

Looking for clues to personality in a person's choice of pets

What You Can Learn From a Handshake

Was your parents' advice right?