The topic of this article is a sort of personal odyssey that takes me from my earliest research interests more than two decades ago to today.
It began with an interest in shyness and social relations. Shy people act carefully and cautiously. They are conservative investors in the interpersonal marketplace, and they don't embrace the social risks that others might. The prospect of disapproval and embarrassment is too much for the shy individual, so disapproval avoidance can overwhelm the desire for approval and all the joys that new acquaintances and relationships can bring.
Test anxiety was another, later interest for me. The chief component of test anxiety is the "worry" that identifies it, worry about doing poorly and suffering the consequences (e.g., poor outcomes, shame, helplessness). Ironically, it is the cognitive interference that stems from worry that interferes with performance, so test anxiety is one of the most self-defeating and useless reactions to performance pressure that you can imagine.
Another set of individual differences that caught my imagination, later, was low self-esteem and depression. One interest was the cognitive bases of dysphoria. For instance, we found that people tend to expend effort on tasks where they believe they have some ability (e.g., hobbies, school, etc.). And, when success occurs as a result of their efforts, people generally will infer that they have the ability to continue to perform well. However, dysphoric individuals may think too much about the causes of their behavior. They appear to apply a principle (called the discounting principle) in which one potential cause of an outcome is discounted when other plausible causes are considered. Dysphoric individuals were less likely than their happy friends to infer that they had ability when they had expended a great deal of effort. Instead, they cited effort as the cause of their performance, and ended up with more doubt, as a consequence, about their ability.
My research interests were based on things that go wrong in people's thinking, feelings, and actions. Today, in light of the very healthy positive psychology movement (e.g., Seligman & Csikszentmihalyi, 2000), what I was studying could easily be characterized as negative psychology. The aim of positive psychology is to move away from a preoccupation only with repairing the worst things in daily life, and refocus on building in the best possible qualities. However, it is also clear that knowing more about the sorts of processes associated with shyness, anxiety, and dysphoria could make a huge contribution toward healthy, positive living as well.
I didn't see the connections among these topics at the time. But the research questions seemed related. The connection became apparent when I first read about and started some research on self-handicapping (Berglas & Jones, 1978). The phenomenon of self-handicapping seemed to "have it all," and my interests came together to focus a research program on it.
A defining feature of contemporary Western society is the extraordinary emphasis placed on achievement and success. Striving for success pervades human interaction in the classroom, on the playing field, and in the corporate boardroom. This emphasis on success suggests that people would be highly motivated to harness all their resources to maximize their potential. People should reach for any and every advantage to facilitate their accomplishments.
However, the phenomenon of self-handicapping suggests otherwise. Self-handicapping is a strategy in which people who are uncertain about their likelihood of success, paradoxically, will sabotage their own performance. In the original demonstration of it, research participants were given a first task, then some feedback about their performance, followed by a second exposure to a similar task.
For half the participants, that first task was challenging, but solvable, and it probably felt as though they had done well. For the remaining half, many of the items were not solvable (there were no correct answers). In both conditions, the experimenter returned and declared the participants' performance to be very good. The success almost surely felt unearned, even illegitimate, for the individuals taking the test comprised of many unsolvable items (termed the noncontingent condition, because their outcomes would appear to be noncontingent on their efforts).
Before repeating the task a second time, the participants were given the opportunity to take one of two experimental drugs, one that was said to be performance enhancing (called "Actavil") and one that was said to interfere with performance ("Pandocrin"). (Actually, all the pills were inert, and not really these drugs at all.) Those noncontingent success participants (those who took the test with no actual correct answers) were more likely to select Pandocrin, the debilitating drug, than Actavil, the performance-enhancing drug. They handicapped their future performance.
The notion of self-handicapping added a proactive twist to the common-sense wisdom that people make excuses for their poor performance. People not only make excuses after the fact, but also plan social occasions so that self-protective excuses are already in place. By handicapping, they were able to protect themselves from the prospect of failing on the second trial of the task. Since this first study, many illustrations of self-handicapping have been reported (e.g., selecting impossible tasks, withholding effort, using alcohol, procrastinating, etc.). The evidence that people will protect themselves from threats to their self-regard is strong and clear, and this research shows that this happens proactively, as a stage-setting device.
Self-handicapping was an exciting phenomenon to study because it brought together many parts of personality and social psychology. There are individual differences. Some people self-handicap more than others (e.g., people with shaky self-esteem). The strategy also has interpersonal elements: people are more likely to self-handicap in public than in private settings, suggesting a strategy designed in part to protect one's standing in other's eyes.
But I found it most compelling because of the type of doubt that a noncontingent success inspires. The individual who succeeds, but doesn't see his or her ability as responsible, ends up feeling self-doubt that ability can be called on again, when trying for future success. This is self-doubt - self-doubt about ability, or competence. Self-doubt occurs when people have uncertainty about their competence (a statistical analogy is that they have a "wide confidence interval" around their estimates of their competence).
Berglas and Jones (1978) captured it well in their observation that "it is better to fail because one is lazy than because one is stupid" (p. 205). In this illustration, it is clear that self-handicapping involves a trade-off between two classes of stigma that someone may reveal. The preference is to protect estimates of one's competence, even at the expense of appearing to be lazy. The threat to one's sense of self-worth is powerful when one's ability is drawn into question; compromising other personal qualities, while perhaps a high price to pay, is better than confirming one's incompetence. The self-handicapper opts for uncertainty about competence because that is preferable to certain information that competence is not as present as one would like to believe.
Before leaving the topic of self-handicapping, let me mention a few intriguing research findings. First, it is people with moderate levels of self-esteem who are most likely to self-handicap (Tice, 1991). These individuals may have the most uncertainty about their competence, and thus self-doubt. Both men and women cite self-reports of handicaps in their lives (e.g., bad mood, too little time to prepare, etc.), but mostly men engage in behavioral forms of self-handicapping (withdrawing effort, procrastination, drug use). Finally, people handicap more when the handicap and their performance are public events than private. All these findings point toward the self-protective basis of self-handicapping (Arkin & Oleson, 1998).
One variation on the self-handicapping strategy is called other-enhancement (Shepperd & Arkin, 1991). In comparative or competitive contexts, performance is judged by relative rather than absolute standards. In a series of studies, we found that participants would provide a competitor with a performance-enhancing advantage (e.g., some music that would facilitate performance, rather than neutral music or music that would interfere with performance). Then, if the participant is outperformed by the competitor, ambiguity is preserved about the basis of the failure. One advantage of the other-enhancement strategy is that an individual does nothing to interfere with his or her own performance (which then can be optimal), but still can preserve some uncertainty about the basis for a poor relative performance, if it occurs. Too, providing enhancements or advantages to a competitor can appear altruistic, another side benefit.
Overachievement is still another strategic behavior designed to protect one's sense of self-worth. To use a biological metaphor, the self-handicapper and the overachiever could not look more different, phenotypically. The self-handicapper is likely to withdraw effort; the overachiever is likely to expend heroic effort. Unlike the self-handicapper, the overachiever avoids failure and strives for success, while the self-handicapper flirts with disaster. Genotypically, however, the two behaviors may be inspired by the same motivational force: self-doubt.
We found that the topic of overachievement had received almost no attention in the psychological literature. This may be because overachievement is conventionally seen as a pretty good thing, with success and achievement so highly prized in our culture. When mentioned, overachievement was defined in objective terms, such as by comparing an individual's college grade point average with his or her predicted performance (using some standardized test).
We decided instead to develop an inventory that would tap the experience of being an overachiever. Our idea was that overachievers are doubtful about their ability, and compensate for this by putting in extra effort. So, the two subscales we created were (a) Concern for Performance and (b) Self-Doubt (the items on each are displayed in the Appendix). We then explored the impact of being high on this two-dimensional scale.
Overachievers do show distinct patterns of cognition, affect, and behavior. We have found recently that overachievers do not enjoy their hard work, they tend to experience relief (not joy) when they achieve success, they tend to avoid tasks if they can—but work hard if the task is unavoidable, they experience powerful negative emotions if they fail, and they often describe their actions as extrinsically rather than intrinsically motivated. In terms of behavior, overachievers do study more than their peers, and they do score a little bit higher on college exams than their peers.
However, the scale is called the Subjective Overachievement Scale (SOS), because it is not about actually overachieving but is instead about feeling like an overachiever.
One neat illustration of overachievement was uncovered in a recent doctoral dissertation (Lynch, 1999). Participants were led either to feel self-doubtful or to feel self-confident. Half of each group then read a newspaper article we created which emphasized either ability as an essential ingredient (for determining feelings of self-worth) or performance as the essential ingredient. In one article, focus was placed on "toil over talent." The other had a focus on natural ability, or "talent over toil." Many quotes from coaches, teachers, admissions officers, and the like told about them focusing on people's desire to succeed (a performance focus) or the talent (an ability focus).
Self-handicapping was found when self-doubt was combined with the message that self-worth should be judged on the basis of natural ability; overachievement emerged when self-doubt was combined with the message that self-worth should be judged on the basis of performance. Neither strategy emerged among those who felt self-confident.
The central role of self-doubt in two such different self-protective, strategic behaviors as overachievement and self-handicapping led us to focus more intensively on self-doubt per se.
In a recent set of studies, we guessed that people with enduring feelings of self-doubt would be more threatened when self-doubt arises, and their feelings of self-worth would decline (Hermann, Leonardelli, & Arkin, in press). Participants were asked to recall either a large number or a small number of instances in their lives in which they felt confident about their ability to perform well in some important arena. The ironic finding was that, among self-doubters, the more instances of self-confidence they were asked to generate, the more self-doubtful they felt, and their feelings of self-worth took a dive. They focused on their feelings of difficulty in generating the large number of instances of confidence, took their retrieval difficulty to heart, and their self-esteem dropped.
This is ironic because participants were engaged in an activity that should hold promise for boosting self-confidence and, consequently, self-regard: generating a long list of positive experiences. Low self-doubt individuals did not show the drop in self-esteem; the listing task actually tended to consolidate their already positive self-regard. They appeared to be affected by the content of the items they listed, rather than any difficulty in the process of generating them from memory.
What might be the origins of doubts about oneself? Berglas and Jones (1978) speculated that some people may grow up with a chaotic reinforcement history, producing uncertainty about overall self-worth. Some parents might also send strong messages about how to judge self-worth. Consider the implications of two self-worth messages ("what counts is ability" versus "what counts is what you do") that parents and other significant figures might send. When self-worth is highly contingent on meeting such a standard, uncertain self-esteem and feelings of doubt may arise if the standards are unclear or the prospect of meeting them is uncertain (Crocker & Wolfe, 2001).
Feelings of self-doubt can send people looking for meaning in their lives, with an eye toward shoring up self-worth. One illustration is materialism, where people place a strong emphasis on some asset, acquisition, or recognition. Self-doubt is a predictor of a materialistic orientation, and we also found that provoking additional feelings of self-doubt heightened materialistic tendencies among self-doubtful people (Chang & Arkin, in press). This is a precarious basis for judging self-worth; materialism has been linked with poor psychological functioning and lower self-esteem and life satisfaction.
Self-doubt is a powerful motivational force. We have seen that it produces a range of protective, fear- and anxiety-inspired, conservative, and safety-oriented strategies for dealing with such uncertainty. Unfortunately, we have also seen that these strategies for dealing with feelings of self-doubt tend to sustain it, rather than reduce it. It seems easy for people to enter a spiral in which self-doubt could become a chronic driving force, producing lifelong tactics of self-handicapping, overachievement, materialism, and the like. We hope the more that we know about the role of self-doubt, its causes, and the strategies that accompany it, the more we can know about how to avoid such a self-defeating spiral.
Arkin, R. M., & Oleson, K. (1998). Self-handicapping. In J. M. Darley & J. Cooper (Eds.), Attribution and social interaction: The legacy of Edward E. Jones (pp. 313-347). Washington, DC: American Psychological Association.
Berglas, S., & Jones, E. E. (1978). Drug choice as a self-handicapping strategy in response to noncontingent success. Journal of Personality and Social Psychology, 36, 405-417.
Chang, L. C., & Arkin, R. M. (in press). Materialism as an attempt to cope with uncertainty. Psychology and Marketing.
Crocker, J., & Wolfe, C. T. (2001). Contingencies of self-worth. Psychological Review, 108, 593-623.
Hermann, A. D., Leonardelli, G. J., & Arkin, R. M. (in press). Self-doubt and self-esteem: A threat from within. Personality and Social Psychology Bulletin.
Lynch, M. E. (1999). Self-handicapping and overachievement: Two strategies to cope with self-doubt (coping strategies, competence). Dissertation Abstracts International, 59(10), 5621B.
Oleson, K. C., Poehlmann, K. M., Yost, J. H., Lynch, M. E., & Arkin, R. M. (2000). Subjective overachievement: Individual differences in self-doubt and concern with performance. Journal of Personality, 68, 491-524.
Seligman, M. E. P., & Csikszentmihalyi, M. (2000). Positive psychology: An introduction. American Psychologist, 55, 5-14.
Shepperd, J. A., & Arkin, R. M. (1991). Behavioral other-enhancement: Strategically obscuring the link between performance and evaluation. Journal of Personality and Social Psychology, 60, 79-88.
Tice, D. M. (1991). Esteem protection or enhancement? Self-handicapping motives and attributions differ by trait self-esteem. Journal of Personality and Social Psychology, 60, 711-725.
This article is based on Dr. Arkin's Psi Chi Distinguished Lecture presented on May 3, 2001, during the annual meeting of the Midwestern Psychological Association. Additional articles in this series may be found on the Psi Chi national website (www.psichi.org/pubs/search.asp?category1=10)