From Mindless Masses to Small Groups: Conceptualizing Collective Behavior in Crowd Modeling.

Area of Science:

• Social Psychology
• Computational Social Science
• Behavioral Modeling

Background:

• Computer simulations are vital for predicting crowd behavior during large events like festivals and evacuations.
• Existing crowd modeling literature often relies on outdated assumptions about crowd dynamics.
• There is a need to align computational models with current social psychological research.

Purpose of the Study:

• To systematically review crowd modeling literature and identify underlying assumptions.
• To evaluate the accuracy of current crowd simulation approaches.
• To propose improvements for more realistic crowd behavior modeling.

Main Methods:

• Conducted a systematic review of 140 crowd modeling articles.
• Coded articles based on how crowd structure was modeled (mass vs. small group approaches).
• Analyzed the alignment of simulation assumptions with empirical crowd behavior research.

Main Results:

• Identified two primary approaches in crowd modeling: mass and small group.
• Found that neither approach accurately simulates large-scale collective crowd behavior observed empirically.
• Highlighted a discrepancy between simulation assumptions and real-world crowd dynamics.

Conclusions:

• Current mass and small group approaches are insufficient for realistic crowd simulation.
• Future crowd behavior simulations should integrate principles from social psychology.
• Self-categorization theory offers a promising framework for modeling crowd member identification and collective behavior.