Beliefs about emotion: implications for avoidance-based emotion regulation and psychological health

ABSTRACT

People’s beliefs about their ability to control their emotions predict a range of important psychological outcomes. It is not clear, however, whether these beliefs are playing a causal role, and if so, why this might be. In the current research, we tested whether avoidance-based emotion regulation explains the link between beliefs and psychological outcomes. In Study 1 (N = 112), a perceived lack of control over emotions predicted poorer psychological health outcomes (increased self-reported avoidance, lower well-being, and higher levels of clinical symptoms), and avoidance strategies indirectly explained these links between emotion beliefs and psychological health. In Study 2 (N = 101), we experimentally manipulated participants’ emotion beliefs by leading participants to believe that they struggled (low regulatory self-efficacy) or did not struggle (high regulatory self-efficacy) with controlling their emotions. Participants in the low regulatory self-efficacy condition reported increased intentions to engage in avoidance strategies over the next month and were more likely to avoid seeking psychological help. When asked if they would participate in follow-up studies, these participants were also more likely to display avoidance-based emotion regulation. These findings provide initial evidence for the causal role of emotion beliefs in avoidance-based emotion regulation, and document their impact on psychological health-related outcomes.

KEYWORDS:

• Beliefs
• implicit theories
• emotion regulation
• avoidance
• psychological health

Disclosure statement

No potential conflict of interest was reported by the authors.

Notes

1. Self-efficacy beliefs also predicted all outcomes independently (without including the implicit theories control measure).

2. In each of the three analyses (loneliness; satisfaction with life; and clinical symptoms) we used the summed total avoidance scale as the mediating variable and the summed total score for the DASS as the global indicator of clinical symptoms. We also conducted supplementary analyses with multiple mediators to examine the independent contribution of the behavioural and cognitive avoidance subscales. Cognitive and behavioural avoidance were both significant intervening variables in all models.

3. Baron and Kenny (1986) causal steps approach to mediation requires that the effects of X on Y (path c), X on M (path a), and M on Y (path b) are significant, as well as that the effect of X on Y controlling for M (path c′) is smaller than c by a non-trivial amount. However, mediation effects may still be observed in the absence of a significant total effect (path c) and/or a direct effect (path c′).

4. Once again we included measures of general implicit theories of emotion (Tamir et al., 2007), and personal beliefs (De Castella et al., 2013). An analysis of the specificity of measures indicated that consistent with Study 1 and with previous research comparing general and personal measures (De Castella & Byrne, 2015; De Castella et al., 2013, 2014), people’s theories about their own emotions uniquely predicted all dependent variables over and above general implicit theories.

5. For all variables analyses were repeated using only the emotion regulation self-efficacy measure to examine whether efficacy beliefs predicted outcomes independently (without including the implicit theories control measure). Regulatory self-efficacy beliefs explained significant variance on all variables: Cognitive Avoidance R²=.14, p<.001; Behavioural Avoidance R²= .16, p < .001; Loneliness R²= .16, p < .001; Life Satisfaction R²=.06, p<.01; Clinical Symptoms (DASS) R²= .14, p < .001.

Additional information

Funding

This work was supported by Australian National University.