Reddit So Wrong?
Researchers have conducted a study where teams were given a single mugshot for a group of "suspects" (located internationally). The teams were asked to use social media to suss out their identities. If folks used targeted tools (not random or indiscriminate contacts), they were able to find more success in the "hunt," perhaps showing that there is (limited) wisdom in large groups [source: Ball].
The "New" Manhunt
For a lot of folks, the Boston Marathon bombing suspect search gave us our first look at how technology and social media changed the process of a manhunt. Not only was the search extensive, it also was available for us to digest in several platforms. Many people around the world found themselves in a self-imposed lockdown, watching real-time updates from friends on Facebook or checking tweets updating the search on their phone at work.
Increasingly, police agencies and services are turning to social media like Twitter or Facebook to apprehend criminals. It makes sense, of course: If you have a picture or description of a suspect, it's smart to get it out to a wide audience and hope that the tips roll in. Tips in several U.S. cities can also be texted anonymously to the police.
The Boston bombings provided another huge game-changer in how manhunts work: crowdsourcing the manhunt. The most sensational story about crowdsourcing came when commenters on social news Web site Reddit began combing through images around the time of the bombings to unofficially ID "suspects." The theories, according to a law enforcement official, were not correct and "complicated" the official investigation [source: Montgomery, Horowitz, Fisher]. In part, official images of the suspects were released to mitigate some of the inaccurate accusations.
As with the Boston bombings, police will occasionally release images or salient details during a manhunt not just to disseminate facts to the public but to provoke the suspects. It worked in the Boston case, as the two suspects were forced on the run.
And although the Boston incident ended with a suspect in custody, it doesn't mean that technology was purely useful to the investigation. An FBI agent had to watch bombing footage 400 times to construct a decent narrative -- the facial recognition software that would've picked up the two suspects (who do have pictures in official databases) didn't work [source: Montgomery, Horowitz, Fisher]. Turns out, there were several good old-fashioned manhunt procedures that aided in the apprehension of the suspect, coming down to one of the very first leads: a victim who woke up from his surgical amputation, requested a pen and paper, and described the person he saw drop a black backpack and walk away shortly before the blast.