Psychological crowds move differently

Introduction

Module 4 builds crowd models in which each person is a particle with a desired speed and a destination, repelled by walls and by other particles. They are powerful tools. But think about the last time you walked out of a concert with friends. You did not take the shortest path to the station at your own best speed. You stayed together, waited for the slowest, walked close enough to talk.

That is not noise around the particle model. It is a systematic difference in how people move when they move as a group, and it has now been measured. This lesson closes the psychology module by showing that the identities of lessons 2 and 3 are visible in people’s feet.

Physical and psychological crowds

A physical crowd is a set of people who happen to be in the same place: commuters crossing a station, shoppers in a high street. A psychological crowd is a set of people who share a social identity, who see themselves as “us”, whether or not they knew each other beforehand. Fans at a match, pilgrims at the Hajj and, as lesson 1 showed, strangers caught in the same emergency can all become psychological crowds.

Templeton, Drury and Philippides reviewed 140 crowd-modelling articles in 2015, under a title that sums up the problem: “From Mindless Masses to Small Groups: Conceptualizing Collective Behavior in Crowd Modeling”. Models had a picture of how crowds behave. The question was whether it matched how psychological crowds actually move.

Walking together: three signatures

In 2018 the same authors tracked 280 walking trajectories and compared people who were part of a psychological crowd with people who were not. Members of the psychological crowd:

1. walked slower,
2. walked further, and
3. maintained closer proximity to each other.

In their words, members of a psychological crowd “(i) walked slower, (ii) walked further, and (iii) maintained closer proximity”. Each signature makes sense once you think of a group as a unit that wants to stay together. Keeping together means moving at the pace of the group, not the fastest member. Staying together means taking a longer route round obstacles and other people rather than splitting. And being together means being close.

At the Hajj: identity beats density

Measured trajectories are one thing. Does identity matter in a mass gathering such as the Hajj? Alnabulsi, Drury and Templeton surveyed 1,194 pilgrims at the Hajj, in a study of “place, space and the process of cooperation”. Their headline result:

“Shared social identity with the crowd explained more of the variance than both location and density.”

In plain terms: if you wanted to predict how a pilgrim behaved towards the crowd around them, knowing whether they identified with that crowd told you more than knowing where they were or how dense the crowd was. For a field that has spent most of this course on density, that is a striking result. Density sets the physical limits; identity shapes behaviour within them.

Why models and plans need psychology

The argument is now made from inside pedestrian dynamics as well. A 2017 paper by Sieben, Schumann and Seyfried, from the experimental tradition you met in Module 1, is titled “Collective phenomena in crowds—Where pedestrian dynamics need social psychology”. When the people who run the controlled experiments make that case, it is worth listening.

For simulation, the consequences are concrete:

• represent groups, with rules that keep them together;
• give group members speeds that reflect the group, not the individual;
• expect closer spacing within groups, and so higher local densities;
• test how sensitive the results are to these assumptions, as Module 4’s lesson on reading a simulation study will ask you to do.

For event management, they are just as practical:

• groups will not split, so a route or gate that only works if people separate will not work;
• people will wait for each other at pinch points, so allow space before and after them;
• in an evacuation, people will look for the people they came with (lesson 1’s affiliative behaviour), so plan meeting points and messages that tell groups where to regroup.