Human Predictability

I’ve written a lot about the improvisational and creative nature of everyday life, for example in my 2001 book Creating Conversations: Improvisation in Everyday Discourse. Wouldn’t it be interesting if we could get some statistical data showing exactly how spontaneous and unpredictable the average person is during the average day?

I just learned about this study*, published in February 2010 in Science, that uses cell phone usage data to measure how predictable our daily movements are. The cell phone companies have to know which cell phone tower you’re connected to when you make each call, so the researchers used data for about 45,000 people (the data were “anonomyzed” meaning no loss of privacy). They used rather complex statistical algorithms to compare the actual movements with the expected trajectories of movement if they were completely random.

For each user, they calculated that user’s predictability of movement–the percentage of movements that could be predicted. The average number across all users was a whopping .93, meaning that only 7 percent of the daily movements were not predictable! They then compared users who tended to stay in a 10 kilometer radius, with those who drove around a lot in a given day. Even the people who traveled a lot had a predictability of .93.

But we don’t always go from one location to the same next location! It turns out the predictability is largely based on the temporal information–the daily times that we move. Relying only on spatial data results in no predictability at all.

More specifically: In a given hour of the day, the data predict which cell phone tower you will be next to 70 percent of the time. That number peaks at .9 in the evening, when most people are at home, and is most variable during rush hour and lunchtime, when people are moving around more.

The researchers conclude by saying that 93% is “an exceptionally high value rooted in the inherent regularity of human behavior” (p. 1021) and that we are predictable “despite our deep-rooted desire for change and spontaneity”. And this predictability did not vary by how far you typically travel, nor did it vary by any demographic characteristics (age, gender, first language, population density of address, or urban vs. rural).

*Song, C., Qu, Z., Blumm, N., & Barabási, A.-L. (2010). Limits of predictability in human mobility. Science, 327, 1018-1021.

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s