Imagine you unleash 100 people onto a soccer (football) pitch and encourage them all to walk around and mingle however they wish. Now imagine you send a drone directly above this crowd to film them. What will you see?

According to a recent set of studies, you’ll almost certainly notice that people are heading to the left more than to the right. In other words, seen from the drone above, the crowd is very likely to be rotating counter-clockwise.

The research behind this new observation has been summarized by many popular outlets recently, including newser.org and phys.org.

Newser wrote:

…people tended to move counterclockwise, whether in a lab, an open schoolyard, or walking alone.

According to the summary in phys.org, the discovery was accidental. The team had been studying pedestrians and found a pattern in the video data:

Following one experiment on a group of test pedestrians, the researchers couldn’t help but notice something in the video data they had recorded that, while not what they were looking for, surprised them and piqued their curiosity.

“When analyzing the experiments, my colleagues realized by chance that in 32 out of 33 experimental trials, as people moved and turned, they noticeably preferred to turn counterclockwise,” said Project Associate Professor Claudio Feliciani from the University of Tokyo’s Department of Aeronautics and Astronautics.

I suggest clicking on the phys.org story, where you can see a drone-based photo of crowd of people (where each person is anonymized as a red dot) and their leftward trajectories.

I thought of this set of studies recently as I was wandering down the hallway of an airport. I started out walking next to the right-side wall. After a few seconds I realized I had drifted toward the center. I’d veered to the left, just as the study found.

After noticing and replicating the leftward trend, the researchers conducted several follow-up studies in an attempt to explain why it happens. In research methods terms, they were looking for the mechanism, or the mediator (a term introduced in Chapter 9).

One hypothesis was that it’s a “nurture” phenomenon. People learn to drift left or right, depending on the culture they live in. That’s why the team ran some of the studies in Japan. In that country, pedestrian norms typically dictate that people walk and drive on the left, rather than the right. In Spain, where the original research was conducted, people walk and drive on the right.  

Another hypothesis was about handedness. They wondered whether left-handed and right-handed people would tend drift in different directions.

Yet another hypothesis was that people would behave differently in groups than they do alone. Specifically, they wondered whether there would be a stronger left-leaning norm when others are around.

Questions

This set of studies is a good opportunity to practice the theory-data cycle (Chapter 1). Why? Because researchers designed studies to test each of these hypotheses.

1a) We’ll start with their hypothesis that cultural norms will influence the direction people spontaneously walk. To test this hypothesis, the research team asked large groups of Spanish and Japanese volunteers to wander around an open space while their movements were tracked by drone. They later counted the proportion of volunteers drifting to the left (vs. the right) in each of the countries.

Set up the axes of a bar graph. label the y-axis, “Percentage of people drifting to the left” .  On the x-axis, label the two countries: Spain and Japan.

1b) Now, before you read about the results, sketch a graph showing what the results the study would be if the cultural-norms hypothesis was supported.

1c) Next, sketch a new graph, showing what the results of the study would be if the cultural-norms hypothesis was not supported.

2a) Now we’ll move to the hypothesis that handedness will influence the direction people spontaneously walk. To test this hypothesis, the researchers asked large groups of left-handed and right-handed volunteers to wander around an open space while their movements were tracked by drone. As before, they counted the proportion of each group drifting to the left (vs. the right).

Set up the axes of a bar graph. label the y-axis, “Percentage of people drifting to the left” .  What should you put on the x-axis for this particular study?

2b) Now, sketch a graph showing what the results the study would be if the handedness hypothesis was supported.

2c ) Next to it, sketch a graph showing what the results of the study would be if the handedness hypothesis was not supported.

3a) Finally, we’ll move to the hypothesis that the preference would be stronger when people walked around in groups than alone.

Set up the axes of a bar graph. label the y-axis, “Percentage of people drifting to the left” .  What should you put on the x-axis?

3b) Now, sketch a graph showing what the results the study would be if the group vs. individual hypothesis was supported.

3c) Next to it, sketch a graph showing what the results of the study would be if the group vs. individual hypothesis was not supported.

Are you ready for the results? Here’s a quick summary of the results of all three studies from phys.org:

Not only did they test cultural background by having parallel tests in Spain and Japan, they also investigated group size, gender, handedness and age.

“Of all these things, the only thing that stood out was that kids tend to have a stronger bias for the counterclockwise direction, so probably age plays a role in making the effect weaker or stronger,” said Feliciani.

4. In this exercise, we followed the theory-data cycle of Chapter 1. We proposed a hypothesis, specified what results would confirm (or disconfirm) that hypothesis, collected data, and reflected on the result. Look back at your graphs, and put a checkmark on the graph that best matches the studies’ actual results.

5. It turns out that none of the three hypotheses were supported by the data, so we should consider another potential mediator. Thinking creatively, consider one more reason, not tested here, that might explain why people drift to the left when they walk around. What groups would you use to test that hypothesis? Sketch a graph of the results you would expect if the hypothesis was confirmed, and a graph of the results you would find if the hypothesis was disconfirmed.