Who am I?: The influence of affect on the working self-concept

Abstract

Two experiments investigated the impact of affect on the working self-concept. Following an affect induction, participants completed the Twenty Statements Test (TST) to assess their working self-concepts. Participants in predominantly happy and angry states used more abstract statements to describe themselves than did participants in predominantly sad and fearful states. Evaluations of the statements that participants generated (Experiment 2) demonstrate that these effects are not the result of (1) participants describing positively and negatively valenced information at different levels of abstraction, or (2) valence-based affective priming. Further, half of the participants in Experiment 2 were led to attribute their affect to the manipulation prior to completing the TST. This manipulation eliminated the influence of affect on the working self-concept. Taken together, these results are consistent with theory and research on the informative functions of affect.

Keywords:

• Affect and cognition
• Affect-as-information
• self

Acknowledgments

This work was partially supported by a grant from the National Science Foundation (BCS-0956309) to LMI.

We would like to thank Thomas Galdi for programming the second experiment and Danielle Dean for serving as the second coder for both experiments.

Notes

¹Data on racial identity were available for 88 participants.

²These questions were included at the end of the study rather than following the affect induction for two reasons. First, we did not want to draw participants' attention to the fact that we were attempting to alter their affect. Second, if participants became aware that the videos may have affected their feelings before the TST data were collected, participants may have correctly attributed their affect to the video clips. Such attributions can result in the elimination of affective influences on subsequently measured variables (i.e., the TST), as we will demonstrate in Experiment 2. The measures that we used are limited in that they rely on participants' recollections of their earlier feelings rather than capturing their actual feelings at the time they were manipulated. We addressed this concern in a follow-up study designed to test our manipulations directly.