Salient selves in uncertain futures

ABSTRACT

We examined possible selves during three distinct periods of uncertainty. Cancer survivors (Study 1a) and survivors’ romantic partners (Study 1b) rated the salience of possible selves in which the cancer did (negative possible self; NPS) and did not (positive possible self; PPS) return. Study 2 mapped PPS and NPS salience throughout the four-month wait for bar exam results. Study 3 experimentally primed possible selves among participants awaiting medical test results. PPS salience correlated positively, and NPS negatively, with indicators of health and well-being, and inducing focus on one’s NPS led to greater negative emotion and worry compared to a PPS induction, but not less positive emotion. These results illustrate the well-being implications of possible selves during periods of uncertainty.

KEYWORDS:

• Possible selves
• uncertainty
• well-being
• cancer
• bar exam

Disclosure statement

No potential conflict of interest was reported by the authors.

Notes

1. It is likely that this number has only increased in the years since their writing.

2. We used treatment status rather than remission status as our exclusion criterion. In this study, 84% of participants reported that they were in full remission, 10% were in partial remission, and 6% were not in remission. The findings remain consistent when only including participants in full remission.

3. The pattern of results was nearly identical when examining the three salience items separately, with the exception that the item “How clear was the mental picture you imagined?” was unrelated to depression and anxiety for participants’ NPS, and the item “How much do you think about this possible future?” was only weakly related to anxiety for participants’ PPS.

4. We again used treatment status rather than remission status as our exclusion criterion. In this study, 66% of participants indicated that their partner was in full remission, 20% in partial remission, and 13% not in remission. Likely due to the large drop in power, some findings (particularly those for NPS) become nonsignificant when only considering participants whose partner was in full remission. The findings remain the same as those reported below when considering participants whose partner was either in full or partial remission.

5. We also explored the possibility that possible-self salience may be more or less predictive of well-being when participants are more or less optimistic about their chances of passing the bar exam. For these models, we included grand-mean centered and person-mean centered estimates and their interactions with all other model terms. These analyses revealed an interaction between within-person expectations and within-person NPS salience on negative emotion, such that at times when participants were particularly pessimistic about their chances of passing, NPS salience was more strongly predictive of negative emotion. Similarly, although within-person NPS salience did not predict positive emotion or poor sleep overall, it did interact with within-person expectations. In both cases, NPS salience was more strongly predictive of well-being (lower positive emotion, poorer sleep) when participants were particularly pessimistic. Although speculative, these findings may suggest that negative possible selves are more powerful to the extent they accord with pessimistic beliefs about the future. No interactions with within-person PPS salience emerged.

6. Due to a survey error, an additional 97 participants completed the survey but were not randomly assigned into either experimental condition. Specifically, these participants were prompted to answer questions about their possible selves but did not write about any possible self. We suspect that this survey error created considerable confusion at the start of the survey for these participants (because they were asked about “this scenario” out of context), and thus we opted to drop them from the study rather than interpreting their emotional state as a type of control condition.

Additional information

Funding

This material is based in part upon work supported by the National Science Foundation under Grant No. BCS1251672 (PI: Sweeny).