Upward Self-Revision: Constructing Possible Selves

Abstract

This research examined how and when people engage in upward self-revision to embrace new possible selves in response to social validation. First, the present study (n = 67) predicted and found that upward self-revision was more likely to occur when validations fully specify the meaning of the positive discrepancy between the desired self and the alternative self into the explicit prospect of the desired self as more likely to occur. Second, initial elevations in self-confidence mediated the effect of social validations on possible selves. The discussion focuses on implications and future directions of the present work.

Notes

¹It is important to distinguish (a) socially presented discrepancies from (b) self-represented discrepancies (e.g., Higgins, 1987). For example, Oettingen (Oettingen et al., 2001) focused on self-initiated goal revisions when people mentally contrast self-representations of present and desired goal reality. By comparison, this model focuses on revisions in response to socially presented contrasts or discrepancies between present and desired goal reality.

²The inclusion of Time 3 allows the test of two complementary predictions concerning the when confidence changes do and do not mediate the process. First, I predict that initial confidence elevations will mediate the effect of validation on behavioral commitment but, second, ultimate confidence elevations will not mediate the effect.

Note. For self-report measures, means within columns with different subscripts differ at p < .05.

³Additional evidence that self-confidence changes are instrumental in mediating the effect of validation specificity to increase behavioral commitment can also be gleaned by showing that the size of the correlation between self-confidence (T1–T2) changes and behavioral commitment increases at higher levels of validation specificity. Consistent with this prediction, the correlation between confidence changes and behavioral commitment was greater in the fully specified validation conditions (r _xy = .87, p < .001) than all other conditions (all r _xy s < .42, all ps > .05) of validation specificity.