Evacuation Behavior in a Subway Train Emergency: A Video-based Analysis

Abstract

How do people behave in the seconds after they become aware they have been caught up in a real-life transport emergency? This paper presents the first micro-behavioral, video-based analysis of the behavior of passengers during a small explosion and subsequent fire on a subway train. We analyzed the behavior of 40 passengers present in the same carriage as the explosion. We documented the first action of the passengers following the onset of the emergency and described evidence of pro- and anti-social behavior. Passengers’ first actions varied widely. Moreover, anti-social behavior was rare and displays of pro-sociality were more common. In a quantitative analysis, we examined spatial clustering of running behavior and patterns in passenger exit choices. We found both homogeneity and heterogeneity in the running behavior and exiting choices of passengers. We discuss the implications of these findings for the mass emergency literature and for evacuation modeling.

Reconciliation in human adults: a video-assisted naturalistic observational study of post conflict conciliatory behaviour in interpersonal aggression

“…Repeated viewings of video captured behaviour allow us to describe how human reconciliation, and its prospective elements, unfold sequentially. This methodological novelty has the potential to generate new theoretical insights, as evidenced in previous CCTV observational work around human violence (Philpot et al, 2019), mass emergency responses (Philpot & Levine, 2022) and bystander intervention (Levine, Philpot & Kovalenko, 2020). This proposed video-observational approach further satisfies recent calls for more direct naturalistic observations of human conflict behaviour (Wrangham, 2008;Richer, 2017), particularly of adults as opposed to children (de Waal, 2000).…”

Beyond ‘stampedes’: Towards a new psychology of crowd crush disasters

“…Further research could also broaden methodological approaches to include analysis of real‐life behaviour from CCTV footage (see Philpot & Levine, 2022) or indeed fieldwork and virtual reality experiments. Crushing disasters and other emergency events are inherently difficult to study, and ideally should be approached using a combination of data sources and methods.…”

3D Social Research: Analysis of Social Interaction Using Computer Vision

“…OpenPose can track multiple individuals within the same frame. Other software packages (e.g., RealSense's Skeleton Tracking SDK) also track human movement, but OpenPose is more efficient, consistent, and accurate when following multiple people (Cao et al 2021; Raaj et al 2019). Raaj et al's (2019) version of OpenPose improves the algorithm's tracking by (largely) resolving the substantial challenge of following individuals across frames.…”

“…Other software packages (e.g., RealSense's Skeleton Tracking SDK) also track human movement, but OpenPose is more efficient, consistent, and accurate when following multiple people (Cao et al 2021; Raaj et al 2019). Raaj et al's (2019) version of OpenPose improves the algorithm's tracking by (largely) resolving the substantial challenge of following individuals across frames. The expansion assigns a consistent ID to each individual throughout the video (though some limitations remain, which we discuss in the reliability section below).…”

“…Such local social influence typically results in wave-like patterns of crossing pedestrians. Philpot and Levine (2021) use a similar approach to analyze local social influence in evacuation behavior during subway train emergencies. Movement in space is thus a crucial component in many social processes at the core of foundational sociological theories.…”