Toward a generative model for emotion dynamics

Area of Science:

• Psychology
• Computational Neuroscience
• Affective Science

Background:

• Emotion theories often describe emotions as reactions to life events.
• Process theories highlight a feedback loop between environment, attention, emotions, and action.
• Existing theories are primarily verbal, lacking clear predictions for intensive longitudinal data.

Purpose of the Study:

• To develop a formal, generative model of emotion dynamics.
• To bridge the gap between verbal emotion theories and quantitative data from experience sampling.
• To provide a foundation for a complete generative model of emotion dynamics.

Main Methods:

• Formalizing the relationship between situations and emotions.
• Developing a basic generative model of emotion dynamics.
• Testing the model against empirical phenomena in emotion time series data.

Main Results:

• The incomplete generative model reproduced nine empirical phenomena in emotion time series.
• These phenomena relate to temporal statistical associations and distributional forms of emotions.
• Demonstrates the utility of a formal approach to modeling emotion dynamics.

Conclusions:

• Generative models offer a powerful framework for advancing emotion theory.
• This approach can enhance emotion measurement, study design, and statistical analysis.
• Further development can lead to a complete generative model of emotion dynamics.