Witnesses Can Identify Criminals With Their Eyes Closed — Here’s How

Crime stinks.

For ages, our sense of smell was considered to be weaker than that of other mammals, like dogs, for instance. That’s why when you’re walking through airport security, there will typically be a German Shepherd on a leash, sniffing for bombs—not a man with his nose up, smelling your bags. But, on the heels of some new research, it seems we humans might have a better sense of smell than we’ve historically gotten credit for.


According to Alice Rolandini Jensen for Fronteirsin.org, when it comes to spotting criminals in a lineup situation, “nose-witnesses can be just as reliable as eye-witnesses.” This belief was recently put to the test via a study designed to test the accuracy and precision of the human olfactory sense.

Jensen points out that because our sense of smell is “directly linked to the areas of the brain associated with emotion and memory,” it is conceivable that this sense would be strongest after potentially stressful events—like crimes, for instance.

To test this, researchers recreated forensic situations to put participants into the minds of true eyewitnesses. They were shown videos of criminals carrying out graphic crimes before being exposed to a unique smell designed to serve as the culprit’s “body odor.”

“They also watched neutral videos, with a similar setup,” Jensen explains.

When asked to pinpoint the perpetrator’s body odor out of a lineup of five men—with five different BOs—participants could identify the correct smell nearly 70% of the time! Even the researchers were impressed. Professor Mats Olsson, an experimental psychologist at Karolinska Institutet in Sweden, marveled, “It worked beyond my expectation.”

An important factor to consider, however, is that the success rate for olfactory incriminations was significantly better in “emotional settings,” such as violent crime scenes.

To test the results of this experiment against the results of similar ones, Olsson and his team toggled the lineup sizes. In addition to the five-person lineup, researchers gathered results from three- and eight-person lineups, as well. Additionally, they also tested various delay times between the video “test crimes” and their subsequent “test lineups,” ranging from 15 minutes to one full week.

As you would expect, the success rate of “nosewitnesses” decreased with larger lineup sizes—although, the same premise applies to eyewitness, as well—and the success rate significantly dropped when the lineup occurred more than one week after the initial exposure to the smell.

While I’m not sure forensic teams are ready to swap eyewitnesses for nosewitnesses—or crime-sniffing dogs for crime-sniffing humans—these findings build an interesting case for trusting the human nose.