Examining individual differences in how interaction behaviors change over time: A dyadic multinomial logistic growth...

Area of Science:

• Social Psychology
• Quantitative Psychology
• Communication Studies

Background:

• Dyadic interactions are theoretically understood through behavioral change patterns.
• Existing methods primarily analyze continuous behavioral dimensions, with less focus on categorical behaviors.
• Bayesian frameworks now enable more accessible modeling of complex behavioral changes.

Purpose of the Study:

• To provide a primer on using multinomial logistic growth models for analyzing within-dyad behavioral change.
• To examine how listener and discloser behaviors change during support conversations.
• To investigate the moderating role of pre-conversation distress on these behavioral changes.

Main Methods:

• Application of multinomial logistic growth models within a Bayesian framework.
• Analysis of behavioral change trajectories in categorical interaction data.
• Utilizing data from 118 dyads engaged in support conversations between strangers.

Main Results:

• Identified distinct change patterns for six types of listener and discloser behaviors during support conversations.
• Demonstrated that pre-conversation distress significantly moderates the evolution of conversational behaviors.
• Illustrated the practical implementation of Bayesian multinomial logistic growth models for dyadic interaction research.

Conclusions:

• Multinomial logistic growth models offer a powerful tool for studying categorical behavioral dynamics in dyadic interactions.
• These models enhance our understanding of how individual differences, like distress, shape interactional processes.
• The findings support the refinement of theories on dyadic interaction by incorporating dynamic, categorical behavioral analysis.