When Trait Reactance Moderates the Effect of Controlling Language on State Reactance

Abstract: Background: Prior research demonstrates that state reactance is a function of message characteristics like controlling language and personalities like trait reactance. However, empirical evidence on whether trait reactance moderates the effect of controlling language on state reactance is inconsistent. Aims: The primary goal of this study is to provide a culture-centered explanation on the boundary of how trait reactance moderates the effect of controlling language on state reactance. Method: Two between-subject experiments, which recruited smokers ( N = 400) and nonsmokers ( N = 431) respectively, were conducted online. They both employed the same 2 (controlling language: high vs. low) × 2 (country: China vs. US) experimental design. Results: In both studies, trait reactance only moderated the effect of controlling language on state reactance in China. Specifically, trait reactance enhanced this effect among Chinese smokers and nonsmokers. Conversely, controlling language elevated the level of state reactance among American smokers, regardless of their trait reactance. Additionally, controlling language was directly related to state reactance only among smokers. The relationship between trait reactance and state reactance was only significant among American nonsmokers. Limitations: This study used a convenience sample and a single issue. Additionally, although the experiment message was adapted from other studies on psychological reactance, future research should pay attention to the possibility of certain confounders. Conclusion: Our investigation suggests that trait reactance may only moderate the impact of controlling language on state reactance in a culture that dismisses the priority of freedom. Hence, even in a culture where the importance of freedom is underestimated, health practitioners should refrain from making people feel their freedom is limited if their target population exhibits high levels of trait reactance.