•  ()
  •  ()
  • Print this Story
  • Email this Story

The Lesson From Boston

What video can and cannot do.

text size AAA
May 10, 2013

This article originally appeared on Al Jazeera.com on April 25, 2013.

By Miriam Gamoran Sherin

The site of the Boston Marathon was described by one Boston official as the "most photographed crime scene in the history of crime scenes". Runners and fans began posting photographs on social media soon after the explosions took place. On the photo-sharing service Instagram, more than one million photographs were posted to the #prayforboston hashtag, many containing images from the marathon. 

Knowing the potential information these images might contain, authorities requested that videos, particularly of the finish line, be uploaded to an FBI website. 

More than one week after the bombings, as memorials continue, we all know now, these images did in fact play a key role in identifying, capturing and ultimately charging the 19-year-old bombing suspect, Dzhokhar Tsarnev. Yet video itself is simply a record, an image. It is what investigators and the public did with those images that made the difference. They took notice. 

Noticing is a daily part of our lives. We do it all the time. We notice when the milk is getting low, when the traffic light changes to yellow, when a light bulb is out. Much of our noticing is unconscious, but it is still an active process. In other words, two people in the same room at the same time might notice different things because they have different goals and expectations for a situation. The first thing I notice when I walk in a bedroom is if the bed is made; but to my teenage daughter, whether a bed is made just is not noteworthy. 

In a well-known experiment, psychologists Dan Simons and Christopher Chabris examined inattentional blindness. They created a video of a small group of college students passing a basketball. Participants were asked to count the number of passes completed during the 30-second video. About half-way through the video, a person dressed in a gorilla costume enters the scene, stands in the front and pounds his hands on his chest. Surprisingly, most viewers failed to notice the gorilla - our noticing is selective. Who would expect a gorilla in the middle of a basketball game? It does not even register.

Those agents who combed through images from last Monday made sure not to miss the gorilla. They prepared themselves to notice. They likely broadened their noticing, to be open, in particular, to anomalies and "absent events". 

It is easy to take in the distress visible in a sea of people reacting to an explosion moments ago. It can be harder to notice the one calm person in the scene. Similarly, when FBI special agent Richard DesLauriers released images of the suspects to the public, he was very non-specific in his description of the suspects; there was no mention of their age or race, for example. This was certainly purposeful, so as to not skew our noticing, to not influence our expectations of what we would see when we looked at the images. And we rose to the challenge. 

Traffic at the FBI.gov website where the images could be viewed reached an all-time high. Within minutes, the online community Reddit had reportedly identified the hats the Tsarnaev brothers wore. Other images of suspects were also identified by the public, such as the one taken by David Green that showed Dzhokhar Tsarneav shortly after the bombing took place. 

Users around the world began to discuss the videos, commenting on the suspects' backpacks, gaits and ethnicity. In doing so, many took advantage of the permanence and editability of video and photographs - they can be viewed over and over again, from different perspectives and with different goals in mind. They can also be edited, such as was done with David Green's photograph, to highlight the suspect who appeared only in the background. 

To be sure, using images in this way is not without risks. Last week, two young men were incorrectly identified as suspects in the bombing, with their names and images plastered over some internet sites. And yet, the payoffs in this case were great. Suspects identified, one dead, one captured. 

Will video always provide such clues? It depends on what we notice.

- Miriam Gamoran Sherin is a professor of education and social policy at Northwestern University.