Description
Emotional contagion models layer a scalar "emotion" or "panic" state on top of a microscopic crowd model — typically social-force-model or agent-based-model — and let it diffuse between proximate agents. High emotion modulates desired speed, repulsion strength, or route choice, producing the herding and stampede phenomena observed in real evacuations. It is the standard extension when "rational pedestrian" assumptions are not enough.
When it's used
- Evacuation modelling under panic conditions
- Coupling cognitive state with kinematic crowd models
- Generating heterogeneous crowd benchmarks for sensing systems
Limitations
- Emotion-state parameters are poorly grounded in real measurements
- Adds free parameters that are hard to validate
- Risk of overfitting to a single dataset