Vision scientists at University of California, Berkeley and Massachusetts Institute of Technology (MIT) have discovered an upside to the brain mechanism that can blind us to subtle visual changes in the movies and in the real world.
For example, viewers fail to notice that Harry Potter's T-shirt changes from a crewneck to a henley shirt in the 'Order of the Phoenix' or in 'Pretty Woman' Julia Roberts' croissant inexplicably morphs into a pancake.
People fail to notice the changes because of a 'continuity field' in which they visually merge together similar objects seen within a 15-second time frame, researchers found.
"Unlike in the movies, objects in the real world do not spontaneously change from, say, a croissant to a pancake in a matter of seconds, so the continuity field is stabilising what we see over time," researchers said.
"The continuity field smoothes what would otherwise be a jittery perception of object features over time," said David Whitney, associate professor of psychology at UC Berkeley and senior author of the study.
"Essentially, it pulls together physically but not radically different objects to appear more similar to each other," Whitney added.
"This is surprising because it means the visual system sacrifices accuracy for the sake of the continuous, stable perception of objects," he said.
Conversely, without a continuity field, people may be hypersensitive to every visual fluctuation triggered by shadows, movement and myriad other factors.
For example, faces and objects would appear to morph from moment to moment in an effect similar to being on hallucinogenic drugs, researchers said.
"The brain has learned that the real world usually doesn't change suddenly, and it applies that knowledge to make our visual experience more consistent from one moment to the next," said Jason Fischer, a postdoctoral fellow at MIT and lead author of the study in the journal Nature Neuroscience.
To establish the existence of a continuity field, the researchers had study participants view a series of bars, or gratings, on a computer screen. The gratings appeared at random angles once every five seconds.
Participants were instructed to adjust the angle of a white bar so that it matched the angle of each grating they just viewed. They repeated this task with hundreds of gratings positioned at different angles.
The researchers found that instead of precisely matching the orientation of the grating, participants averaged out the angle of the three most recently viewed gratings.
"Even though the sequence of images was random, participants' perception of any given image was biased strongly toward the past several images that came before it," said Fischer.