a1 Department of Psychology, University of Texas, Austin, TX 78712-0187. email@example.com firstname.lastname@example.org http://homepage.psy.utexas.edu/homepage/faculty/swann/
Von Hippel & Trivers suggest that people enhance their own self-views as a means of persuading others to adopt similarly inflated perceptions of them. We question the existence of a pervasive desire for self-enhancement, noting that the evidence the authors cite could reflect self-verification strivings or no motive whatsoever. An identity negotiation framework provides a more tenable approach to social interaction.
(Online publication February 03 2011)
We were impressed with many of von Hippel & Trivers's (VH&T's) specific arguments. For example, they build a convincing case that self-deception may facilitate the manipulation of others and that it can be used to effectively mask clues as to one's intent. Yet when it comes to the contention that the deception of the self and others represents a core ingredient of human social interaction, we respectfully disagree.
Some of our reservations regarding VH&T's central “self-deceptive self-enhancement” thesis are purely conceptual. At one point the authors assert that “People are impressed by confidence in others” (sect. 3, para. 2) and then proceed to list the supposed benefits of confidence. Fair enough. But they then conclude that the premium humans place on confidence suggests that “self-enhancement should be ubiquitous and people should believe their own self-enhancing stories” (sect. 3, para. 2). Whereas confidence is often based on actual abilities or achievements, self-deceptive self-enhancement is thought to produce overconfidence. Surely the consequences of confidence and overconfidence are very different. In fact, we could readily imagine a complementary paragraph that lists the hazards of overconfidence and concludes that self-enhancement should be maladaptive and hence rare.
Empirical evidence of self-verification strivings provides an even stronger basis for concluding that self-enhancement plays a relatively modest role in social interaction. For example, contrary to VH&T's contention that people preferentially seek and embrace positive evaluations, there is strong evidence that people with negative self-views preferentially seek negative evaluations and interaction partners and even divorce spouses who perceive them in an overly positive manner (e.g., Swann 1983; in press).
VH&T dismiss this literature by suggesting that self-verification is merely another motive that exists alongside self-enhancement. What they fail to recognize, or at least acknowledge, is that their argument loses force insofar as self-verification overrides self-enhancement. Recent evidence suggests that this may indeed be a problem for the authors' formulation. A meta-analysis of studies that pitted self-enhancement against self-verification indicated that self-verification effects were at least as strong as self-enhancement effects (Kwang & Swann 2010). Moreover, when the response class “cognitive reactions” (which included selective attribution, attention, recall, overclaiming bias, and perceived accuracy) was examined specifically, the overall pattern favored self-verification over self-enhancement. This is noteworthy because it is precisely this response class that VH&T emphasize as the province of self-enhancement.
At the very least, evidence that self-verification strivings often trump self-enhancement strivings challenges the notion that there exists a pervasive desire for self-enhancement. More tellingly, however, such evidence also calls into question the most appropriate interpretation of much of the evidentiary basis for self-enhancement. Consider that there is convincing evidence that roughly 70% of the people in the world have positive views of themselves (Diener & Diener 1995), presumably because most people enjoy secure attachment relationships (Cassidy 1988; Sroufe 1989) and are able to systematically seek and engineer success experiences (e.g., Bandura 1982). The pervasiveness of positive self-views, in conjunction with evidence that people work to verify their negative and positive self-views, means that evidence that people seek and embrace positive evaluations could reflect either self-verification or self-enhancement strivings. Of particular relevance here, studies in which most participants embrace positive feedback may reflect a tendency for the 70% of participants with positive self-views to seek feedback that is, for them, self-verifying.
To concretize our claim, consider the Fein and Spencer (1997) article that the authors cite repeatedly. Because no measure of self-views was included in this research, it is possible that the tendency for negative feedback to amplify participants' derogation of out-group members was driven by a tendency for those with positive self-views (who presumably constituted most of the sample) to work to maintain their positive self-views. If so, then the findings may reflect self-verification rather than self-enhancement strivings. Of course, the authors also cited a study by Epley and Whitchurch (2008), which did include a measure of global self-esteem. Yet expecting the measure of self-esteem to act as a moderator in this study is problematic because there is little reason to believe that feedback regarding physical attractiveness should be moderated by global esteem (which is influenced by numerous factors besides attractiveness; for a discussion of this specificity matching problem, see Swann et al. 2007).
The authors also rely heavily on a series of studies that purport to show that people routinely claim that they are “better than average.” Whether this body of work should be viewed as evidence of self-enhancement, however, is debatable. For example, careful analyses have shown that participants are rather non-discriminating when it comes to endorsing the above-average option, even claiming that an unknown stranger performs better than average (Klar & Giladi 1997). The most straightforward interpretation of these findings appears to be that people have a very dim understanding of what an average score means and what it means to assert that they are better than average. If so, the results of such studies can hardly be regarded as evidence of self-enhancement (for a review, see Chambers & Windschitl 2004).
In short, although there are surely instances in which deception is personally beneficial, we believe that the research literature provides little evidence that such activities offer an apt characterization of human social conduct. Instead, a more tenable model of human social interaction may be offered by the identity negotiation formulation (e.g., Swann 1987; Swann & Bosson 2008). When people begin interacting, the argument goes, their first order of business is to determine “who is who.” Once each person lays claim to an identity, they are expected to honor it henceforth; failure to do so will be disruptive to the interaction and could even trigger termination of the relationship. So, people are not only motivated to seek subjectively accurate (i.e., self-verifying) feedback, but also their success in eliciting such feedback represents the interpersonal “glue” that holds their relationships together. Within this framework, social relationships are maintained through transparency and mutual understanding rather than deceit and obfuscation, and it is allegiance to truth that enables people to enjoy healthy, prosperous relationships.