Medical imaging experts better at solving optical illusions—study shows training can improve visual perception

Medical imaging experts are adept at solving common optical illusions, according to research from four UK universities, including the University of East Anglia.

The correct analysis of medical images from scans, such as MRI, is critical for diagnosing cancer and many other conditions.

A new study, "Specific Visual Expertise Reduces Susceptibility to Visual Illusions," published in the journal Scientific Reports, shows that people who do this professionally are also more accurate at judging the size of objects in common optical illusions.

In other words, medical imaging experts also literally see better in everyday life.

The research is also the first to show that people can be trained to do better at solving visual illusions, which was previously thought to be near-impossible.

Senior researcher Dr. Martin Doherty, from UEA's School of Psychology, said, "Optical illusions are designed to fool the brain. They can be a bit of fun, but they also help researchers shed light on how our brains work.

"We wanted to better understand whether people who are very experienced and skilled in visual recognition do better at solving optical illusions."

Participants were shown a series of visual illusions that made it hard to correctly judge the size of two similar objects—and asked to identify the larger one.

The object size differences varied, giving the research team an estimate of how much participants' judgments were affected by the illusions.

They tested 44 radiographers and radiologists and compared their scores with a control group of 107 non-experts.

First author Dr. Radoslaw Wincza, from the School of Medicine and Dentistry at the University of Central Lancashire, said, "Many illusions are effective even if you know it is an illusion, and until now it was generally believed you could not train yourself to avoid the illusory effects.

"But this research suggests that training aimed at accurately perceiving objects in medical images has the effect of making experts less susceptible to visual illusions.

"This is the first time that it's been shown that people can be trained to do better at solving visual illusions. And they could perhaps even be used for training medical image analysts in the future.

"This is particularly important, given that 60 to 80% of diagnostic errors are perceptual in nature," he added.