Affect dynamics or response bias? The relationship between extreme response style and affect dynamics in a controlled experiment

Intensive longitudinal data (ILD) have become a popular data format to capture people's momentary affect in everyday life. Besides describing persons' average affect over time, ILD are also often used to describe affect dynamics - that is how affect changes over time -, such as intraindividual variability or moment-to-moment temporal dependencies. Given that ILD studies mostly use self-report rating data, there is an increasing concern that response biases, such as extreme responding, might impact the estimates of affect dynamics. In this study, we assessed the relationship between extreme responding and affect dynamics in a controlled experiment. In a highly powered sample with N=1,398 persons, we measured extreme responding using background questionnaires, and repeatedly induced affect using a probabilistic reward task with T=140 trials per person. Our results suggest that people with high extreme response style trait levels show substantially higher measures of affect variability. However, extreme responding is neither associated with moment-to-moment temporal dependencies nor with participants' reactivity to affective stimuli. We conclude with a discussion on the importance of evaluating measurement in ILD for psychological assessments, and outline potential areas for future research to improve assessments of affect dynamics.