How simple rules determine pedestrian behavior and crowd disasters[1]
Summary
Proposes a cognitive-science model of pedestrian behaviour based on behavioural heuristics rather than Newtonian force. Guided by visual information — the distance of obstructions in candidate lines of sight — pedestrians are modelled as applying two simple cognitive procedures to adapt walking speed and direction. Though simpler than earlier approaches, the model predicts individual trajectories and collective patterns of motion in good quantitative agreement with a wide variety of empirical and experimental data, and reproduces self-organisation phenomena including spontaneous unidirectional lane formation and stop-and-go waves. Combining the pedestrian heuristics with body collisions generates crowd turbulence at extreme densities, a phenomenon observed during recent crowd disasters.
So what
The practical value is the emergence of crowd turbulence from ordinary walking rules plus body contact. Turbulence at extreme density is not a separate catastrophic mode requiring panic or malice to explain; it falls out of normal people navigating normally once bodies are touching. That supports the systemic-failure framing over the panic framing, and it means the intervention point is density itself rather than crowd behaviour. Note this is a simulation study validated against empirical data, not a measurement of a real crowd.
Read the paper
- Publisher (DOI) ↗may be paywalled
- PubMed ↗abstract · free