Abstract
<p>Emotion controllability beliefs are central in emotion regulation (ER), yet experimental evidence on whether such beliefs can be deliberately altered remains scarce, and no prior study has simultaneously targeted general emotion beliefs and ER self-efficacy or compared manipulations across domains (emotions vs. anxiety). In a pre-registered online experiment (data collected between April and June 2024), 290 non-clinical participants (aged 16-25 yrs.) were randomly assigned to one of four conditions in a 2 (induction direction: fixedness vs. malleability) x 2 (domain: emotions vs. anxiety) between-subjects design. General beliefs and ER self-efficacy were assessed before and after induction. The sample was restricted in age range and drawn from a single, Western, higher-educated context, limiting generalizability to more diverse populations.Fixedness induction strengthened anxiety uncontrollability beliefs (d = 0.33), with stronger belief change in the anxiety than the emotion domain (d = 0.44). Malleability induction did not produce significant belief changes in either domain. ER self-efficacy did not change significantly in either domain, possibly reflecting the domain-unspecific nature of the self-efficacy measure used. We interpret the asymmetry between fixedness and malleability inductions, and the domain-specificity of fixedness effects for anxiety, within the extended process model of ER (Gross, 2015) and the evolutionary smoke detector principle (Nesse, 2001): anxiety, as a threat-coupled primary emotion, may be particularly susceptible to fixedness reinforcement due to an evolutionarily tuned negativity bias. These findings suggest that general beliefs about emotion controllability and ER self-efficacy show differential susceptibility to experimental manipulation, with implications for the design of belief-targeting interventions.</p>