Monday, March 8, 2010

What You See v. What You Get

"The work of science is to substitute facts for appearances, and demonstrations for impressions."
~ John Ruskin

About five years ago I was serving jury duty and was assigned to a pool being considered for a street crime case. After the normal voir dire, I was one of 12 people still sitting in the jury box. At this point the judge came in to explain a peculiarity of the case: the only evidence against the defendant was the eyewitness testimony of the mugging victim, who had seen his knife-wielding victimizer only once, on the night of the crime two years earlier. She asked, "Does anyone have a problem with that?" I raised my hand.

Standing in front of the judge like a defense attorney on Law & Order, I explained that in New York's bad old days, I had been the victim of one attempted mugging, one attempted assault and one robbery at gunpoint. And based on my experience, I did not believe the victim in this case could accurately identify an ordinary-looking attacker he had seen only once, under stress, after a two-year delay.

Here is why. The first time someone tried to mug me, the attacker was a peculiar-looking man with skin the color of lightly toasted Wonder Bread and a tight corona of dirty blond curls; he would have been easy to identify, had anybody in the mugbooks looked remotely like him.

Tangent: Have you ever sat in a police station leafing through old-fashioned mugbooks? The first time I did it, a stooped, gnarly cop right out of central casting repeatedly came to me with photos of swarthy men, or perhaps multiple photos of the same swarthy man. "Is this the guy?" I almost felt bad the first few times I reminded him that my mugger looked nothing like that.

The second man who tried to attack me was ordinary looking — so much so, several pages worth of mugbook faces looked at least a little bit like him. I was good at portraiture, so two or three days after the incident I sat down and tried to sketch him. I very quickly realized that my memory was adding false details — eyes like someone I knew, hair like someone in the mugbook. The truth was, during the incident my attention had been not on the man's face and but on the knife he was carrying.

In the third attack there were two criminals, and one put a gun to my head. They were a total blur from the get-go; my focus had been on my possessions as they disappeared down the street into the darkness. The judge reminded me that jurors are instructed to base their conclusions only on the evidence presented in the case. I reminded her that human nature was likely to overrule my best intentions. I was released back into the jury pool.

Scientists tell us our eyes show us only what evolution has decreed worth seeing. We see a narrow swath of the visible spectrum and completely miss what's outside of it — the worlds of infrared and ultraviolet, for example — while other creatures, including many insects and birds, can see a much broader spectrum. Our brains concoct three-dimensional images from the two-dimensional input provided by our stereoscopic vision. We are prey to optical illusions of many kinds.

Researchers have developed some interesting experiments to test visual acuity. One of these involves a video of students playing basketball. If you were to participate in this experiment, you would be told to find the students in the white t-shirts and watch them carefully, counting how many times they pass the basketball by throwing it and simultaneously counting how many times they pass it by bouncing it. Take a look at the video now — it's less than a minute long — and try the experiment. Be sure to keep a careful count of the passes and whether they are throws or bounces. Just watch it once — no cheating. We'll come back to the video and see how well you did in a moment.

Yesterday I attended a lecture by Marvin Chun, a cognitive scientist and professor of psychology at Yale. He spoke about fMRI studies of the brain areas that respond to faces and places. Interestingly, imagined places and faces stimulate a response almost as strong as real ones. Also interesting: gearheads respond to cars as if they were seeing members of their families!

And speaking of fMRIs, another recent study found that our eyes and brains find the images they produce are so persuasive, we are likely to believe even dodgy interpretations. Even scientists suffer from this illusion. You can download a PDF of the study here. (And thank you to the neuroscientist known on Twitter as @mocost for alerting me to it.)

Tricks of the eye can be entertaining, too. Take a look at the elephant above; notice anything funny about it? You can see many more optical illusions on this Web site.

Although our eyes lie unintentionally, they can still wreak havoc on people's lives. Between 1992 (when the Innocence Project first brought DNA testing to bear on wrongful convictions) and January 2010, some 249 defendants were exonerated of serious crimes. Some had spent decades in prison. In three quarters of the cases, the original convictions were based entirely or in part on eyewitness testimony. As this 2009 60 Minutes episode points out, our eyes lie. 

Now, about that video: Did you see this guy? Most people don't.

"Appearances to the mind are of four kinds. Things either are what they appear to be; or they neither are, nor appear to be; or they are, and do not appear to be; or they are not, and yet appear to be. Rightly to aim in all these cases is the wise man's task."~ Epictetus

No comments:

Post a Comment