It feels as though you see the whole room in front of you in rich detail. The impression is strong but it is misleading. Experiments over the past decades show that people often fail to notice large changes in a scene, if the change happens at the moment their attention is interrupted. The effect is called change blindness, and it shows that your sense of a complete picture comes largely from knowing where to look when you need something.
The flicker experiments
In 1997 Ronald Rensink, Kevin O'Regan and James Clark published a study using what they called the flicker paradigm. A photograph appears for a fraction of a second, a blank grey screen appears briefly, and a version of the photograph with one thing changed appears. The two alternate. The change could be an engine of an aeroplane that disappears or a railing that shifts. Participants often needed many cycles, sometimes ten seconds or more, to find a change that was obvious once found. Without the blank, the change catches the eye because it makes a flash of motion. With the blank, that signal is lost and you have to compare the two images from memory.
The same effect has been shown in real life. In a well-known study by Daniel Simons and Daniel Levin in 1998, an experimenter asked a passer-by for directions. While they spoke, two men carrying a door walked between them and the experimenter was swapped for a different person in different clothes. About half of the pedestrians did not notice that they were speaking to someone else.
Inattentional blindness and the gorilla
A related finding is inattentional blindness: you can miss a visible object because your attention is elsewhere. In the famous 1999 video by Simons and Chabris, viewers counted how many passes a team in white shirts made with a ball. A person in a gorilla suit walked through the scene, stopped to beat their chest and walked off. About half of the viewers did not see the gorilla. Attention is selective, and what is not attended may not reach awareness, even if you look right at it.
Why the brain works like this
- Memory for scenes is sparse. The brain does not store a photograph. It stores a gist and a few details that mattered, plus knowledge of where to look to get more.
- Attention is limited. Roughly four objects can be held in working memory at the same time (see our guide on working memory), so a complicated scene cannot be compared in full.
- Expectations fill in. If a scene looks as it should, you assume it is the same. Changes that fit the context are missed more easily than those that do not.
- Importance matters. Changes to the centre of interest are found faster than changes to unimportant parts of a picture.
Why this matters in daily life
Change blindness has consequences. It is one reason for collisions in which a driver "looked but did not see" a motorcyclist, and it is a concern for air-traffic controllers and radiologists, who look at screens in which a small difference matters. It also plays a role in witness reports: confidence in an observation is a poor guide to its accuracy. A good rule is to be modest about what you think you saw when your attention was elsewhere.
How to see more
- Use a system. Scan the display in a fixed pattern, such as left to right and top to bottom, instead of letting your eyes jump around.
- Chunk the scene. Divide it into regions and check each in turn. Our Change Spot game is built so that this works.
- Look for what stands out in the difference and not the scene: ask whether anything is missing, new or in a different place, as three separate questions.
- Do not stare. Switching to a different place and back often reveals what was missed, since the first look created an expectation.
- Rest. Attention drops quickly when you are tired, and tasks of this type are one of the first things to suffer.
Memory for what you saw
Several games on this site rest on the same ability. In Vanishing Shapes you must say which item disappeared, in Pattern Shift you must tell which pattern is the copy, and in Pairs you must remember where items were. Success in each depends on how well you encode a small number of items and not on how sharp your eyesight is. If you find that you fail on a round where you "saw everything", you have met your own change blindness.