Riley Crane wins computer science challenge

Physicist Riley Crane works at MIT’s human dynamics lab. Dr. Crane measures how information moves – on sites like Facebook, for example – and asks:

Riley Crane: How does society respond to some stimulus, whether it’s anything ranging from an earthquake or a tsunami to a call to find ten red balloons?

‘Ten red balloons’ refers to a contest sponsored by the Pentagon, which challenged teams to be the first to find ten red balloons, randomly hidden all across the U.S. Crane’s MIT team was the first to find all ten – in under 9 hours – with public help in the search.

Riley Crane: We made a website that allowed anybody to sign up by putting in their e-mail… so we could track how the information was getting from person to person…

Crane’s now analyzing this data to discover what rules might govern the spread of information online. His aim, he said, is to explore the opportunities that come with living in such a connected world. He says this field is so new, scientists aren’t exactly sure what they’re going to find, or how to use it. But, he added, by mastering these rules, we’ll get more effective at using digital tools to communicate information about a crisis, like an earthquake.

Riley Crane: You could also ask the question of: how do diseases spread, and to what extent are spreading phenomena similar in the digital world versus the real world.

Dr. Crane talked about some of the challenges of figuring out how to use digital tools to try to understand patterns of disease spread. He said he doesn’t just use computer data.

Riley Crane: We’re starting to use phones, and sensors, to be able to track how things are spreading in the real world. If we can figure out how to better collect data using phones as a platform, we can think about starting to solve health and other challenges in the real world.

One of the projects he’s working on at the MIT media lab involves a small community that will receive android phones. He’s going to study their communication patterns, their mobility patterns, and learn more about how ideas (and even applications) spread. His team is going to complement this by asking this community to complete surveys regarding their health, and health behavior.

Riley Crane: We can see, for example, how their mobility patterns or who they communicate with impact how likely they are to get the flu.

He said that you have to use found data. In other words, pick a community where an illness might spread during the course of your experiment. (Though Crane emphasized that he’s not wishing illness on anyone!)

Riley Crane: You have to look for clever ways of finding data where something else like that might also happen to be going on. There was a previous study in our research group which was studying an entire dorm, and there just happened to be a flu outbreak. So there are a lot of cases like that, where there’s and occurrence during the study that then allows you to ask new questions.