Scientists use the running of the bulls to gather data on the running of the humans

This is an Inside Science story.

People walking alone walk relatively quickly. A crowd walks slowly. But how does a crowd move when there is, say, a massive bull charging at them? To answer this, scientists analyzed the movement of a crowd of runners during the running of the bulls in Pamplona, Spain, in 2019.

The San Fermín festival in Pamplona, Spain, hosts the world's best-known running-with-bulls event. Every morning for a week each year, festival officials send six bulls charging down a set of narrow, blocked-off streets toward waiting crowds of people.

In 2019, pedestrian dynamics researcher Daniel Parisi, from the Technological Institute of Buenos Aires, attended the festival -- though he was not running himself. Instead, Parisi was gathering data from a set of cameras roughly four or five stories above the route.

Pedestrian dynamics is the science of studying how crowds move and is useful for anyone designing or planning spaces like buildings or streets. Stadium exits that allow people to file out quickly without forming a human traffic jam may do so thanks to insights from pedestrian dynamics.

But there's a gap in scientists' knowledge. While scientists have a lot of data from crowds moving at walking speeds, when it comes to crowds running, there's much less out there. It's hard to predict when something will create a stampede of people in the real world. A running crowd can be dangerous if people start to trip each other.

"In a laboratory, this kind of experiment with real danger is not ethical. But this festival is almost like laboratory conditions," said Parisi. The street is easy to monitor, the runs happen several times and in the same place, and the stimulus -- the sudden appearance of a charging bull -- is a consistent motivational force. "So this is not experimental, but it's quasi-experimental because it's repeated in the same conditions over and over."

By taking the videos and using software to tag and track the position of each runner, Parisi and his colleagues were able to explore whether scientists can use real-world events like this to look for insights. And in doing so, they made two discoveries.

Firstly, Parisi and his team found that while the crowd could be at times fast or dense, there was a combination of speed and density that never occurred due to runners tripping each other and falling. "It's intuitive," said Parisi. "With restricted personal space, you can't run very fast."

Their second result, however, was unexpected. While it's traditionally assumed that loose crowds move fast and dense crowds slow, Parisi found that as the bulls first approached the runners, for a short amount of time, the crowd became both denser and faster. "This is surprising and never was observed before," said Parisi. He notes that this may be somewhat unique to the event's particular circumstances -- people starting relatively spaced out and stationary, only to be sent scrambling in extreme ways to avoid the bulls. The runners also know to expect the bulls, even if they don't know precisely when they'll show up.

The researchers' paper, published last week in the journal PNAS, shows that by using the right technology, pedestrian dynamics researchers can indeed look to this kind of festival or extreme event for insight about crowds and their behavior, though more work needs to be done.

"I think that the methodology holds some promise," said Rachel Berney, an urban design scholar at the University of Washington. She said she'd view the research as exploratory -- it's good to review any finding with other methods before relying on it -- but that this type of research could provide insight for urban designers.

Carlo Ratti, director of the Massachusetts Institute of Technology's Sensible City Laboratory, said in an email to Inside Science that he thought conducting the study at the running of the bulls was interesting and that "the paper shows how the Big Data revolution is helping researchers document and describe previously unexplored environments, throwing light on problems which were not quantifiable just a few years ago."

As for Parisi himself, he was content to observe the runners without experiencing the event first-hand. "Maybe if I were 20 years younger, but not now," he said. The festival has many other less risky activities he enjoyed instead.

Inside Science is an editorially independent nonprofit print, electronic and video journalism news service owned and operated by the American Institute of Physics.