Usually, I don’t ruminate, only from time to time: the predictive value of trait and state measures of rumination for the intensity of affective states

Emotions and their flexible regulation are essential in adapting to our environment. Individuals differ significantly in their ability to regulate their emotions, and research has traditionally focused on these individual differences. However, evidence emerges that traditional laboratory measures and questionnaires capturing trait emotion regulation may not correspond to emotion regulation ability in real life as assessed by repeated measures designs. These processes appear to show different variability between and within individuals, implying that characteristics regarded as traits in psychology – e.g., emotion regulation effectiveness and typical use of strategies – may not reflect everyday functioning.One of the most extensively researched emotion regulation strategies is rumination, a cognitive process that involves repeatedly and passively dwelling on negative feelings and their consequences. Interestingly, recent results suggest only a weak correlation between state and trait measures of rumination. Furthermore, evidence supports that increased state (but not trait) rumination is associated with higher emotional reactivity, the emotional reactions experienced in response to a negative occurrence. However, based on the evidence described above, the temporal, within-person examination of rumination may predict emotional reactivity even more accurately.The present study aims to confirm that within-person, state measures of rumination hold a more substantial predictive value for emotional reactivity compared to between-person trait rumination. We also investigated the association between state and trait rumination in this longitudinal study.The sample comprised 247 individuals who provided data for 28 days in an experience sampling method (ESM) study. Participants received 8 short, discretional questionnaires daily, assessing their momentary affective states, emotion regulation, and perceived stress. 14265 observations were obtained, then analyzed using mixed-effects models.Our results suggested that all state, aggregated mean state, and trait measures have independent predictive validity for emotional reactivity. However, the magnitude of these effects overlapped, and therefore, the importance of the measures could not be ranked. On the other hand, when negative affect alone was considered, the mean of state measures of rumination emerged as the strongest predictor. This composite indicator can essentially be considered a trait measure to some degree, however, it incorporates the everyday variability of the process within an individual, thus differing from traditional, self-report trait measures that can be subject to recall bias and lower ecological validity. Thus, it could be regarded as a measurement ‘halfway between’ trait and state indicators that could possibly combine the advantages of both, thereby having the best incremental validity in terms of predicting emotional reactivity.Furthermore, we found only a moderate positive correlation between trait and state rumination indicators. Certain psychometric and contextual differences could account for this result, as well as the different self-report techniques tapping different sources of self-related information.Our findings provide valuable insight for understanding the within-individual dynamics of emotion regulation and their relationship to stress and affectivity, nuancing the common trait approach and overcoming some of its limitations. Thereby, our results corroborate the need to re-evaluate the traditional measurement of psychological constructs. Additionally, our findings have clinically relevant, practical implications for the personalized prediction of emotional responses to stress through simple self-report tools applicable in everyday contexts.