Collective Behaviour in Evacuation Processes

By Arturo Cuesta, Orlando Abreu, and Daniel Alvear

We are affected in our decision-making by what others around us are doing. This can also happen during emergency evacuations. Previous research helps us interpret fundamentals of collective behaviour during evacuation. Current methodologies are often based on anecdotal recollections of actual evacuees (interviews, surveys, etc.) and/or theoretical frameworks[1]. The response of others has been proved to influence our own response to ambiguous threat cues[2]. This can also happen when the fire alarm is unclear as to where the fire might be located[3].

In such conditions, we interact with others to decide what to do[4,5]. Proximity also seems to be an important factor: We are more influenced by people who are close than by those who are farther away[3]. We also cooperate in emergencies[4–8]. Past incidents such as the Beverly Hills Supper Club fire have shown that social groups tend to escape or succumb together[9]. Evacuation groups can emerge spontaneously[4] or be formed based on social ties[10] and the organizational/situational context[5].

To sum up, the evacuation process can be considered as a social process in which people are likely to make consensus decisions, decide on a plan of action, and act together[5, 11–13].

We wanted to investigate how people make decisions in emergencies— that is, if and to which extend they make decisions and act together[14]. The philosophy behind our study was that collective behaviour denotes a reduction in the behavioural variability of people: the smaller the dispersion of a behavioural variable, the greater the cohesion among the members of an evacuation group.

study involved a set of evacuation experiments at the bus terminal of a multimodal station in Madrid (Spain) involving 150 individuals. The bus terminal consisted of a waiting area with 10 boarding gates (to access the bus departure bays) and five emergency exits (see Figure 1). We randomly assigned six to nine individuals to each boarding gate (See Figure 2). Three different trials were conducted for a set of 30 potential evacuation groups where individuals, due to proximity, were likely to interact with each other (through verbal and/or non-verbal communication). The trials were recorded by using video cameras and data was statistically analysed afterwards.

Figure 1. Layout of the bus terminal: Boarding gates (BGs) and emergency exits

Figure 2. Participants accessing the concourse during one trial

Proximity was found to be an important factor in the formation of groups. Most individuals at the same boarding gate reached a consensus decision regarding the direction of evacuation and used the same exit. However, choosing the same exit was not enough to determine the collective behaviour.

The study also allowed us to measure whether participants responded together (as a group) and maintained cohesion during the evacuation movement. The behavioural cohesion was observed in eight potential evacuation groups during both the response and the movement phases of evacuation (26%). Nine other potential evacuation groups showed behavioural cohesion during the movement phase of evacuation (30%). Although participants of these groups did not respond collectively, they adapted their speed to maintain proximity to each other during their movement and left the concourse together.

Emergency evacuations often depend on social interactions among individuals. Previous research has provided useful information to interpret fundamentals of collective behaviour during evacuation qualitatively. This study illustrates an example of a quantitative approach that can be employed for analysing collective evacuation response. Detailed information on this topic can be found in the paper associated with this study[14].

Arturo Cuesta, Orlando Abreu, and Daniel Alvear with the University of Cantabria, Spain 


1. Gwynne, S., Galea, E.R., and Lawrence, P.J. 2006. The introduction of social adaptation within evacuation modelling, Fire and Materials, 30:285–309.

2. Latane, B., and Darley, J.M. 1968. Group inhibition of bystander intervention emergencies, Journal of Personality and Social Psychology, 10(3):215–221.

3. Nilsson, D., and Johansson, A. 2009. Social influence during the initial phase of fire evacuation—Analysis of evacuation experiments in a cinema theatre, Fire Safety Journal, 44: 71–79.

4. Turner, R.H., and Killian, L.M. 1987. Collective Behaviour. Englewood Cliffs, NJ: Prentice Hall, Inc.

5. Aguirre, B.E., Wenger, D., and Vigo, G. 1998. A test of the emergent norm theory of collective behaviour, Sociological Forum, 13(2).

6. Jones, B.K., and Hewitt, J.A. 1986. Leadership and group formation in high-rise building evacuations, Fire Safety Science- Proceedings of the First International Symposium, 513–522.

7. Fahy, R.F., Proulx, G., Aiman, L. 2012. Panic or not in fire: Clarifying the misconception. Fire Matter, 36, 328–338. doi:10.1002/fam.1083.

8. Quarantelli, E.L. 1954. The Nature and Conditions of Panic. American Journal of Sociology 60:267– 275.

9. Feinberg, W.E., and Johnson, N.R. 2001. The Ties that Bind: A Macro-Level Approach to Panic, International Journal of Mass Emergencies and Disasters 19:3.

10. Sime, J.D. 1980. movement towards the familiar: person and place affiliations in a fire entrapment setting, Environment and Behaviour.

11. Connel, R. 2001. Collective behaviour in the September 11, 2001 evacuation of the World Trade Center, Disaster Research Center, University of Delaware.

12. Kuligowski, E.D., 2016. Human behaviour in fire. SFPE Handbook of Fire Safety Engineering, fifth ed., 2070–2114.

13. Gwynne, S.M., & Kuligowski, E. D., 2016. Representation of Human Behaviour in Fire. In Evacuation Modeling Trends, 55–80. Springer International Publishing.

14. Cuesta, A., Abreu, O., and Alvear, D. 2016. Methods for measuring the collective behaviour in evacuees, Safety Science, 88:54–63.

© SFPE® | All Rights Reserved
Privacy Policy