Beyond Cause to Consequence: The Road from Possible to Core Self-Revision

Abstract

Two studies addressed the ultimate consequences and pathways running from repeated possible self-revisions to gradual revisions in core selves over time. As hypothesized, greater prior experiences of downward possible self-revision ultimately predicted greater subsequent declines in core self-integrity (e.g., greater self-doubt, lower self-esteem). However, also as hypothesized, this effect was mediated by the relative use of defensive versus remedial attributions for past downward self-revision experiences. In closing, we unpack how the present work extends prior work by situating possible selves and motivated self-attributions as complementary systems that can slowly undermine as well as expand the integrity of core selves over time.

Keywords::

• Possible selves
• Motivated self-attributions
• Core selves
• Self-revision

Notes

^1. Of course, Epstein distinguished his approach from Kelly's Personal Construct Theory that viewed people as naïve scientists who—like scientists—developed naïve theories, or constructs, to better understand, predict, and control their life experiences. In particular, Epstein's model did not view people as rationale scientists determined to systematically constructed and refined self-theories to better explain and control the data of life experiences. Rather, they were motivated scientists whose self-theory was driven by “hot” motivations to fulfill basic needs (e.g., hedonism, self-esteem, relatedness) other than accurate understanding over the course of life. Relatedly, Epstein argued that his model (vs. PCT) placed a greater significance on the role of emotion in driving the self-theory.

^2. By “self-assessment error”, we generally mean any error in the cognitive appraisal or behavioral expression of one's true core competencies. Thus, for example, a student who excels in computer science rather than art but believes they are better at art than computer science has made a self-assessment error. In the context of the present paper, self-assessment errors are problematic because new possible selves are translated from one's perceived (vs. actual) core self-competencies. Thus, the self-assessment error can lead to translational errors such that the student translates possible selves (e.g., become an artist) that he believes are realistic generalizations from his perceived core competencies but, in fact, are completely unrealistic generalizations because he does not actually have the true core competencies (e.g., artistic competencies) required to support that possible self-pursuit. The longer the self-assessment error remains in place, the more errors occur in the translation of his perceived core self-competencies into unrealistic new possible selves (artist) that are more likely to fail than realistic possible selves (computer scientist, programmer, etc.) translated from his true core self-competencies (computer science not art). Consistent with this point regarding the problems with self-assessment errors, Dunning and colleagues noted that “to the extent that people misjudge themselves, they may suffer costly consequences by pursuing wrong paths and missing opportunities to take advantage of special skills and resources they truly own (Dunning et al., 2004, p. 70).”

^3. Of course, remedial attributions may identify other types of self-assessment errors such as errors in the initial strategy or effort one takes to successfully realize one's true potential. Thus, for example, he could realize that his performance and competency within the sciences depends upon the quality of the strategies and effort he invests with his better performances (vs. worse) restricted to prior conditions in which he used effective study habits and high effort (e.g., the 99% perspiration/1% inspiration formula for scientific success). Whatever the source of the error, early remedial attribution permit the young man to identify and resolve it to improve the translation of his core self-competencies into better new possible self-competencies.

^4. Importantly, in contrast to the 26 excluded in Study 2 for never having experienced a revision, only four participants were excluded from Study 1. Although there may be other methodological differences that could explain this difference (retrospective vs. prospective survey design), it is very likely that the differences in actual college experience between the pre-college (Study 2) and college (Study 1) sample explain the differences in possible self-revision experiences. The greater number of Study 1 participants who reported prior revisions experiences may merely be because their greater college experience also provided greater opportunities to experience possible self-revision relative to their Study 2 counterparts. That is, unlike the Study 1 sample in which students had an average of 1–2 years of college experience when we assessed them, we assessed the Study 2 sample before they ever arrived on campus. Thus, there may have been fewer Study 2 participants than Study 1 participants with prior possible self-revision experiences because we assessed them before (vs. 1–2 years after) they had actually experienced the transition into the intensely evaluative university environment, where direct challenges to unrealistic possible selves frequently arise from educator attempts to steer “wayward” students toward more realistic career possibilities.

^5. To lend further credence to our hypothesis that attributions for failure mediate the relationship between self-revisions and self-integrity outcomes, we estimated an additional model with direct paths from the T1 measures to the T2 outcome measures of self-esteem and well-being. The model fit was excellent, RMSEA = .000, but the utility of RMSEA is limited in this model because nearly all possible paths are estimated, and so the model-implied correlation matrix is to be expected to be nearly identical to the sample correlation matrix. As (or more) importantly, the additional paths from T1 control measures to T2 outcome measures were not significant, and so in keeping with our hypotheses the model discussed here does not include any direct paths from T1 to T2 outcomes.

No potential conflict of interest was reported by the authors.

This publication was partially supported [grant number T32-MH19728s] from the National Institute of Mental Health. The contents are solely the responsibility of the authors and do not necessarily represent the official views of the National Institute of Mental Health.