Challenges and opportunities in emotion dynamics research: an interdisciplinary perspective

Recent advances in the acquisition of intensive longitudinal data (ILD) have enabled affective science to investigate emotional experience in increasingly ecologically valid contexts. These developments highlight the complexity of psychological systems and the challenges involved in understanding emotions. In this perspective paper, we argue for a conceptual adjustment in how measurement is practiced in emotion-dynamics research. Drawing on affective science, computational modeling, and the philosophy of science, we propose that measurement should be treated as an interaction that may partially structure and modify the phenomenon being measured. This perspective has implications for the design, acquisition, and analysis of ILD. We identify three complementary directions for future research. First, measurement frameworks should be theory-informed and transparent, with methodological choices and relevant hyperparameters explicitly documented to support cumulative science. Second, subjective self-report should be complemented, where appropriate, by behavioral, physiological, and environmental measures to provide partially independent perspectives on complex psychological processes. Third, greater attention should be given to temporal scale in ILD measurement and analysis, including multiscale dynamical network approaches that connect proximal, observable, and potentially perturbable processes with higher-level psychological phenomena. Together, these directions support a shift from describing statistical associations toward a more process-oriented understanding of psychological phenomena.