| 07 Nov 2025
Human Decision-Making in Crowds in a Virtual Flood Scenario
Abstract. Flood evacuation outcomes are critically shaped by human behaviour, yet empirical data on individual decision-making remain scarce due to the dangers and logistical challenges of collecting data during real disasters. To address this gap, this study used Virtual Reality (VR) to examine how social cues, specifically crowd behaviour, interact with factors such as crowd size, clarity of the safe destination, and floodwater level to influence evacuation choices and delays. Four within-subjects VR experiments were conducted with 84 participants, systematically testing these variables in an immersive flood scenario. Results showed that crowd behaviour strongly determined both route choice and evacuation latency, often outweighing other factors. Participants tended to follow crowds into floodwater, demonstrating herding behaviour. However, this influence weakened when water levels were very high, indicating a threshold where physical danger overrides social cues. Larger crowds and unclear destination information further increased reliance on social information and pre-movement times. These findings highlight the powerful role of social dynamics in emergency decision-making and underscore the need to integrate realistic human behaviour, particularly social influence, into flood risk models, public warnings, and evacuation planning to improve community resilience and safety.
This paper reports a series of experiments using virtual reality technology to test effects of environmental and social cues on individuals’ decisions to enter into flood water in an emergency evacuation. The topic is an important one, and the research would be of interest to both researchers and practitioners (emergency planners) who plan for floods. I am aware of the wider literature on crowd dynamics in evacuations, but there is much less on behaviour in floods, so the study is a welcome contribution. The use of virtual reality technology, while artificial does allow for experimental control of scenarios and has been used for other types of emergency evacuations. The role of social influence is crucial in these events as previous research has shown, and these studies suggest strongly that it matters in floods, in interaction with certain environmental factors. The study was situated in an appropriate and up to date review of the relevant literature. The experiments and analysis were conducted competently, and the study clearly written. Overall therefore I welcome this paper.
Â
Comments, questions and suggestions:
Â
Abstract. Consider changing ‘a threshold where physical danger overrides social cues’ to ‘a threshold where more obvious physical danger overrides social cues
Â
Is it necessary to use the term ‘herding’, which is more suitable for animals and their instincts, has been criticized in the literature (see Haghani et al., 2019)? ‘Social influence’ (also used) is a more neutral alternative term. Observing the behaviour of the majority of people is a good guide for how one should behave (Gigerenzer, 2008), particularly when those people are judged to be self-relevant in some way (Spears, 2021). The authors’ finding that pre-movement time was longer in risky conditions suggest that people are thinking about the examples they observe, rather than following others instinctively.
Â
The term ‘natural disasters’ is criticized in the disasters literature, and the term ‘hazards’ is suggested instead (with disaster being the social effects of a hazard). See for example UNDRR https://www.undrr.org/our-impact/campaigns/no-natural-disasters
Â
How was the interview data analysed?
Â
Study 1. A large effect size is expected – but why? The fact that the authors suggest post hoc that the sample size was too small indicates that this assumption was unwarranted in this case
Â
Can the authors provide a link to footage/ moving visualization?
Â
Was the questionnaire developed by the authors themselves or did they use established items? We should at least see some example items (and preferably there should be a link to the whole questionnaire, so the wording of items can be seen).
Â
Table 2 post hoc column seems to indicate that conditions were compared across experiments for VR3 and VR4, which is incorrect.
Â
Questionnaire tables should include notes reminding us what the A, B, C, D conditions are.
Â
Page 18. It is unclear what is meant by ‘chaotic crowd behaviour’ in the analysis of study 4, as there is no indication earlier that ‘chaotic behaviour’ would be varied in the visualization.
Â
The discussion makes a strong claim that social cues are more important than environmental cues, even for deep floodwater. However, the analysis of VR4 could make much more clear whether there was a significant main effect of flood water level (rather than just the interaction/ tests across the four conditions).
Â
Adrian, J., M. Amos, C. Appert-Rolland, M. Baratchi, N. Bode, M. Boltes, T. Chatagnon, M. Chraibi, A. Corbetta and A. Cuesta (2025). "Glossary for Research on Human Crowd Dynamics. This needs to be properly cited as the second edition.
Â
References
Gigerenzer, G. (2008). Why heuristics work. Perspectives on Psychological Science, 3, 20-29. doi 10.1111/j.1745-6916.2008.00058.x
Haghani, M., Cristiani, E., Bode, N. W., Boltes, M., & Corbetta, A. (2019). Panic, irrationality, and herding: three ambiguous terms in crowd dynamics research. Journal of advanced transportation, 2019(1), 9267643.
Spears, R. (2021). Social influence and group identity. Annual Review of Psychology, 72(2021), 367-390.