If You Can’t Notice a Gorilla in Plain Sight, How Can You Testify as a Witness?

By Daniel Simons | December 14, 2011 8:48 am

by Daniel Simons, as told to Discover’s Valerie Ross. Simons is a professor of psychology at the University of Illinois, where he studies attention, perception, and memory—and how much worse people are with those skills than they think. He is the co-author, with fellow psychologist Chris Chabris, of The Invisible Gorilla.

Late one January night in 1995, Boston police officer Kenny Conley ran right past the site of a brutal beating without doing a thing about it. The case received extensive media coverage because the victim was an undercover police officer and the aggressors were other cops. Conley steadfastly refused to admit having seen anything, and he was tried and convicted of perjury and obstruction of justice. Prosecutors, jurors, and judges took Conley’s denial to reflect an unwillingness to testify against other cops, a lie by omission. How could you run right past something as dramatic as a violent attack without seeing it? Chris Chabris and I used this example to open our book because it illustrates two fundamental aspects of how our minds work. First, we experience inattentional blindness, a failure to notice unexpected events that fall outside the focus of our attention. Second, we are largely oblivious to the limits of perception, attention, and awareness; we think that we are far more likely to notice unexpected events than we actually are.

Chabris and I have studied this phenomenon of inattentional blindness for many years. Our best-known study was based on earlier work by Ulric Neisser: We  asked subjects to count how many times three players wearing white shirts passed a basketball while ignoring players wearing black who passed their own ball. We found that about 50 percent of subjects failed to notice when a person in a gorilla suit unexpectedly walked through the scene.

The mismatch between what we see and what we think we see has profound implications for our court system. As our research has shown, we can fail to notice something obvious if we are focused on something else. Yet, most jurors likely hold the mistaken belief that we should see anything that happens right before our eyes. Kenny Conley was convicted on the strength of that intuitive belief. Many others likely languish in jail due to similarly mistaken beliefs about the accuracy of memory. By studying these limits of attention and memory and our beliefs about them, we identify cases in which our beliefs diverge from reality. Ideally, we can then reveal these “invisible gorillas” in the court system.

In Conley’s case, the beating took place in the middle of the night, following a high-speed chase through the streets of Boston that culminated in Conley chasing a suspect on foot and apprehending him nearly a mile later. Conley presumably was focused on chasing the suspect, much like our subjects were intently focused on counting passes. And, that intent focus meant that he, like our subjects, was less likely to notice unexpected events falling outside that focus of attention.

Although the analogy from inattentional blindness in the lab to the situation faced by Conley seems reasonable, Chabris and I had a nagging concern: No studies had tested the effects of inattentional blindness under real-world conditions like those faced by Conley. There is plenty of anecdotal evidence (as well as observational studies) showing that people do miss unexpected events in the real world, but we wanted to verify that our “inattentional blindness” explanation for Conley’s experience was plausible.

In the study we published in i-Perception (conducted by Chabris and his students, Adam Weinberger and Matthew Fontaine, at Union College), we did our best to simulate some of the key aspects of Conley’s experience. Our subjects were asked to jog behind an experimenter, monitoring how many times the experimenter touched his hat. As they were jogging along a predefined route, they ran past a simulated fight scene in which two other experimenters were “beating” a third experimenter. We found that even in broad daylight, only 56 percent of subjects noticed the fight. At night, only 35 percent did. Presumably, even fewer people would have noticed had we been able to ramp up the intensity to something consistent with a high-speed police chase of a murder suspect.

Coupled with the Conley case, this finding illustrates how flawed our thinking about inattentional blindness can be. The jury convicted Conley at least partly because they intuitively believed that he must have seen the beating. That belief, what we call an illusion of attention, is one of many ways that we misunderstand our own minds. It follows naturally from how we experience our world. We are only aware of those unexpected events we happen to see, and we are unaware of all the ones we never saw. The pattern we experience, then, is of always noticing unexpected events, and it’s rare that we experience our failures of awareness. If you missed the gorilla and nobody ever asked you about it or showed you what you missed, you would continue believing that you would notice it. Inattentional blindness is not the only courtroom problem caused by our imperfect brains. For example, when we recall a vivid experience, it often passes before our mind’s eye in great detail. But we rarely check the accuracy of that imagery, and when we do, the details often are distorted. Our assumptions lead us to trust confident testimony about memories (“That’s the guy who robbed me. I’m sure of it.”) when decades of research on memory, such as the great work by  people like Elizabeth Loftus, reveals its fallibility. These assumptions we make about our memory lead to flawed decisions by juries: we assume someone must be lying when they simply can’t remember; we trust confident witnesses more than we should; we believe that people must have noticed anything falling within their field of view.

This summer, Chabris and I published a paper in PLoS One based on a national survey we conducted to explore how pervasive these mistaken beliefs about memory actually are. The survey was a representative sample of the American population, and we found that 63 percent thought memory works like a video camera and 78 percent believed that unexpected objects will automatically grab notice, even when attention is focused elsewhere. These flawed intuitions help explain how so many people can be convicted of murder on the basis of eyewitness testimony, only to be exonerated later via DNA testing. Unless we can inform lawyers, judges, and juries about these assumptions and the limits of perception, we will continue to see innocent people convicted solely on the basis of eyewitness testimony.

One of the big remaining questions in the study of the limits of attention and perception is whether some people are better able to notice and remember than others. Undoubtedly there are individual differences, but my guess is that the variability across people is far smaller than the similarities among us.While some people may notice a bit more than others, we all have severe limits on how much of the world around us actually reaches our awareness at any moment.

Correction: The study in i-Perception was conducted by Chris Chabris and his students Adam Weinberger and Matthew Fontaine, not Daniel Blakely.

CATEGORIZED UNDER: Mind & Brain, Top Posts
  • Dunc

    One of the big remaining questions in the study of the limits of attention and perception is whether some people are better able to notice and remember than others.

    I have trouble directing my attention much of the time (I’m supposed to be working… ;)) In the classic gorilla experiment, I’d completely lost track of the number of passes (and thus given up trying to count them) long before the gorilla showed up, so yeah, I had no trouble spotting it.

    I often notice that I pick up on stimuli that others seem unaware off.

  • Christopher Chabris

    Hi. Thanks for publishing this piece about our work! Just a small correction: the i-Perception study you describe was done by me with my students Adam Weinberger and Matt Fontaine. I don’t know a Daniel Blakely. Thanks,


  • Jim

    Mr. Simons, I don’t know if you’ll be checking the commentary on this, or if Ms. Ross will be able to relay the message, but I much appreciate your book. It has helped me a great deal in my professional life (I often work as a trainer) and was entertaining and enlightening.

  • bob

    50% of people didn’t notice the gorilla? Were exactly half of your test subjects blind?

  • Valerie Ross

    @Chris: Sorry for the mistake, and thanks for letting us know — we’ve updated the post with the correct names!

  • Jim

    bob, I’ve run this experiment myself with groups of up to 20 people, all with at least a bachelor’s degree. Several had held jobs that required extensive observational skill, including one surgeon. What was fascinating was that in the versions of this that I’ve run, it’s 50% of the people, but education does not appear to play a role at all. The surgeon, incidentally, did not notice the gorilla.

    (read the book – it’s fascinating)

  • Dan Weiss

    National Geographic did a Three part Series called “Brain Games” a few months ago also demonstrating the same thing amongst others that showed the limitations to our perceptions.

    One must always remember we are only spiritual beings in a human form and limited by the sensory inputs allowed by that limitation..

  • norsenomad

    Could this relate to the percentage of us that will persistently perceive the negative, when no bias or even a positive bias is expressed? Those that will always look for the hidden message which must of course be negative? Then again, some percentage of the population will always perceive the positive, when negative aspects are presented? Would the term “Predetermined Perception (of reality)” apply to either? Are both simply alternate ways of filtering the real world to fit our expectations?

  • Jay Warner

    I knew to expect the gorilla, yet almost missed him as I concentrated to count the passes. The second time, switching my brain from passes to gorilla I could feel the former dropping away. A ‘high functioning autism spectrum’ person (i. e., the ones I’ve known) has great difficulty switching from topic a to b. We value these folks’ concentration ability; we don’t also expect them to scan for divergent inputs. And you and me, we are different? How?

  • http://www.the-gardener.org David Kline

    Take this a step further. I learned, as a technical writer in a noisy environment, to intentionally block out the noise. The noise was still there; it just didn’t register. (This was so successful that I did not pursue my idea for noise cancelling earphones which were finally developed several years later.) Your experiments deal with visual perception. Somehow my internal noise blocking ability transferred to visual events such that when visualizing a physical project, the design of a workbench for instance, the external visual field before my open eyes, though still there, just doesn’t register. In other words, inattentional blindness or deafness are matters closely aligned with how we focus our attention. We can turn a deaf ear or a blind eye, but the perceptual field continues to be generated in accordance with our sensors; what we do with this field is a matter of focus, subject to our intent, or to involuntary distractions.

  • Tiktaalik

    Maybe this explains how I can drive my car for 20 minutes while thinking about something else and have no memory of having driven the car or the environment I passed through, yet somehow I arrive at my destination…

  • Pingback: Farnam Street If You Can’t Notice a Gorilla in Plain Sight, How Can You Testify as a Witness?()

  • floodmouse

    The more you micro-focus on a specific task, the less aware you are of other things going on around you. When I was in grade school, I could concentrate on reading a book to the extent that I wouldn’t hear people speak to me, and they assumed I was snubbing them. I learned this was a bad idea in public. Ironically, people who have less ability to micro-focus (or who aren’t using it) might be more likely to notice gorilla.

    On a tangent, there was a documentary featuring the stage illusionists Penn and Teller, who demonstrated how easy it is to draw someone’s eyes away from an object coming out of the magician’s sleeve. Once your attention is directed to the relevant portion of the screen, it’s easy to see, but unless you know it’s coming, your eyes will be drawn to a different part of the frame. I think this was a “Nova” documentary about how the brain works, but maybe that is my memory playing tricks on me. :)

    “Our assumptions lead us to trust confident testimony about memories (‘That’s the guy who robbed me. I’m sure of it.’) when decades of research on memory, such as the great work by people like Elizabeth Loftus, reveals its fallibility. ” – Another example from my personal experience: I was the third car in a minor three-car collision. The three cars were traveling in the same direction at a slow speed, and my car was low to the ground behind the other two cars. I told the police officer that I thought the girl in front of me failed to brake because I didn’t see her brake lights, and I believed she struck the front car first–but I correctly pointed out that I couldn’t be SURE if she struck the front car first, because from my vantage point, I could not see the front bumper of her car. The girl in question simply told the police officer, definitively and confidently, that I hit her first, causing her to slide into the other car. The police office accepted her version of the story without investigating further. In the absence of a video camera or corroborating evidence, I will never be any surer of what really happened than I was that day when I admitted my ignorance. I just wonder what would have happened if I had very confidently assured the police officer that the girl in front of me was responsible for the accident, without bringing up the little matter of what I was ACTUALLY able to see in the field of vision from my rather low-slung front seat.

  • Pingback: If You Can’t Notice a Gorilla in Plain Sight, How Can You Testify as a Witness? - Association for Psychological Science()

  • Yes man

    Did you repeat this with counting passes from black players? The gorilla is black too. Test subjects just don’t register black guys?

  • Brian Too

    Magicians routinely manipulate our attention. It’s wrong to suggest that eyewitness testimony is inaccurate as a blanket assertion. However eyewitnesses are vulnerable to seeing what they expect to see, especially if they have been prepared in any way.

    In the Penn and Teller documentary there was a fascinating section. One of the magicians (Penn?) wanted to know how one particular part of a trick worked. They couldn’t understand how an action taken in plain sight was never seen by the audience. The show noted that the action could be seen however, so long as the face of the magician was covered. Even though this was not the primary focus of the audience!

    Somehow the experience of being directly looked at by the magician steals some of our attention.

  • Pingback: Our Fathers, Ourselves & Much More | Red Weather Review()

  • Patrick F

    I’ve seen the video twice in psychology classes. Even having seen it, if I’m counting passes as directed, I miss the gorilla despite KNOWING it’s there. I can easily believe the cop missed the beating.

  • Jim G.

    Did anyone notice how many times the players in black passed the ball? Of course not! The issue is not the lack of awareness to what is going on but of focus and concentration on what has been determined to be important. Why was the police officer running past the beating? Was he out jogging for exercise or chasing after someone who just committed murder? Ignore the instructions about the video and just watch it. You will notice the gorilla. Anyone will.

  • TRJc

    It is significant that both the gorilla and the unimportant players wore black. Repeat the experiment with the instruction to count the passes by the players wearing black and people will notice the gorilla. Or substitute for the gorilla a person dressed as a polar bear, and again count the passes by the players wearing white. Whenever the animal suit is the same color as the significant players, it will be noticed. Whenever the animal suit is the same color as the unimportant players, it will mostly not be noticed. Perhaps the most interesting would be to repeat the experiment with players wearing white and yellow. In this case the gorilla would neither belong to the important color, nor the unimportant color, but to a third color.

  • Wang-Lo

    First time through, I lost track of the number of passes and didn’t notice the gorilla either. I spent most of the time watching the sexy girl in the yellow bikini.


  • Peter

    Fyi people, this experiment is impossible to be done correctly when you know the gorilla is coming. I read and knew it was coming. But when the video was played in my psych class without background information, the majority of us missed the gorilla.


Discover's Newsletter

Sign up to get the latest science news delivered weekly right to your inbox!


See More

Collapse bottom bar