Predicting Group Contribution Behaviour in a Public Goods Game from Face-to-Face Communication

Area of Science:

• Behavioral Economics
• Experimental Economics
• Social Psychology

Background:

• Public goods games in experimental economics reveal declining contributions over time due to free riding.
• Face-to-face communication (FFC) is known to reduce free riding and improve cooperation, but the underlying mechanisms are not fully understood.

Purpose of the Study:

• To investigate the mechanisms by which FFC promotes cooperation and discourages free riding in public goods games.
• To analyze both the visual (facial expressions) and content aspects of FFC to understand their impact on group behavior.

Main Methods:

• Automatic facial expression analysis of 3-minute FFC videos using a group activity descriptor and random forest classification.
• Content analysis of FFC transcripts, categorizing strategy-relevant topics and utilizing metadata.

Main Results:

• Facial expression data from 3 minutes of FFC can predict group contribution behavior towards the end of the game.
• Both facial expression analysis and content analysis indicate that FFC, especially when sustained until the end, positively influences contributions.

Conclusions:

• Facial expressions and conversation content during FFC play a significant role in promoting cooperation in public goods games.
• Further research with larger datasets is needed for a deeper understanding of the predictive power of facial expression analysis.