Why do we miss what is in front of us?

Participants watched a video of two teams passing basketballs and counted the passes made by one team. Partway through, a person in a gorilla suit walked through the game from one side to the other while the players carried on around them, on screen for five seconds.

Half of the 192 people tested saw a woman with an umbrella in place of the gorilla. Across both, 46 per cent did not report the unexpected event; for the gorilla alone, 56 per cent did not. Shown the video a second time they were consistently surprised, and some said “I missed that?!”

The version in which the gorilla stops in the middle of the players, turns to the camera and thumps its chest is an extra condition, reported in the paper’s discussion. It lasted nine seconds and was shown to twelve people. Six of them noticed.

The effect is stronger when the counting task is harder, and when the unexpected object is less like what is being attended to. Watching the team in black, 58 per cent noticed the black gorilla. Watching the team in white, 27 per cent did.

Twenty-four radiologists searching lung scans for nodules were given a last case with a gorilla in it, more than forty-eight times the size of the average nodule they were hunting. Twenty did not report it, and eye tracking showed that most of those who missed it had looked directly at where it was.

For most of them the eyes did their part. What did not happen is whatever turns that input into something a person can report. So the missing part is not absent from the input. It is absent from the report. Whether it is also absent from the experience is a harder question, and a more unsettling one.

An alternative account is that the gorilla was experienced briefly and then not encoded, which would make this a memory failure rather than a perceptual one.

The two are hard to separate, because the only access anyone has to what a participant experienced is a report made afterwards. The authors think blindness the more parsimonious reading: someone who had seen the gorilla and forgotten would not be shocked to be shown it. This is the same measurement problem that runs through the study of consciousness.

Later work has pushed from both sides. Ward and Scholl had people report anything unexpected the moment they saw it, and some still missed a salient event that broke the pattern they had come to expect, which the authors take as a failure of perception rather than of memory. In 2025 five experiments with more than 25,000 online participants found that, taken as a group, people who said they had noticed nothing could still report the colour, shape or location of what they missed better than chance, and leaned towards saying no. Their authors read this as seeing that inattention degrades rather than abolishes.

The laboratory version uses a demanding task and a brief event. Driving, reading and conversation are also demanding tasks, and the effect has been demonstrated with real objects in real settings.

One study staged it outdoors. Participants ran after a confederate along a 400 metre route, counting how many times he touched his head, and passed a driveway eight metres off the path where three men were staging a fight. At night seven of twenty noticed it. In daylight nine of sixteen did. The study was built around a court case in which a Boston police officer was convicted of perjury and obstruction of justice for saying he had not seen an assault he ran past. He later won an appeal and was not tried again.

What is not known is the rate. Nobody can count the things they did not notice, which is a limit on the question rather than on the answer.