How Do You Miss a Gorilla? By Counting Something Else

Participants watch a video of people passing basketballs and are asked to count the passes made by one team, which requires real concentration. Partway through, somebody in a gorilla suit walks into the middle of the scene, faces the camera, beats their chest and walks off, taking around nine seconds. Roughly half of participants report seeing nothing unusual, and many refuse to believe it happened until the video is replayed. The study was published by Christopher Chabris and Daniel Simons in 1999 and became one of the most widely known results in psychology.

Several findings show the failure is in attention rather than in the eye:

• •Eye tracking shows participants look directly at the unexpected object
• •They fixate on it for a measurable period and still do not report it
• •Making the counting task harder increases the failure rate
• •Making it easier decreases it
• •Expecting something unusual almost eliminates the effect
• •So the determining variable is what attention is committed to

The most robust finding around the effect is that people are certain it would not happen to them, which is a separate phenomenon in its own right. Surveys find that a large majority believe they would notice the unexpected object, and that belief is held just as firmly by people who have just failed to notice it. The explanation is that experience gives no evidence of what was missed, since a person who fails to see something has no memory of not seeing it and nothing to correct the impression of a complete and continuous view. The subjective sense of seeing everything is therefore untroubled by the repeated demonstration that it is wrong.

Several neighbouring effects are distinct and are easily run together. Change blindness is a failure to notice that something has altered between two views, which requires a comparison across time rather than a failure to see something present. The attentional blink is a brief window after noticing one thing during which a second is missed, measured in fractions of a second. Banner blindness describes learned avoidance of a familiar location on a page. And selective hearing at a noisy party is the same phenomenon in sound, where one conversation is followed and the rest is present and unprocessed until a name cuts through it.

The effect has consequences well beyond the laboratory and is cited in several fields. Drivers looking for cars at a junction fail to see motorcyclists and cyclists, which the accident literature records as a consistent pattern and which is why riders are advised to assume they have not been seen. Radiologists searching for one abnormality miss another, demonstrated by inserting an image of a gorilla into a scan that most specialists then failed to report. Aviation training addresses it directly, since a crew working a problem can miss an instrument warning. And it is a recognised limit on eyewitness reliability, since a witness concentrating on a weapon frequently cannot describe the face above it.

Committing attention to a demanding task makes people fail to report something obvious even while their eyes are fixed on it, which shows the failure is in attention rather than in vision. Almost everybody believes it would not happen to them, including people who have just demonstrated otherwise, because missing something leaves no trace to notice. Drivers miss motorcyclists and radiologists miss second findings for the same reason.