Hands quicker than eyes when it comes to screening

Oct 24 2007

That fleeting moment of regret between clicking the wrong icon and seeing an unwanted web page pop onto the screen could make a huge difference in improving the accuracy of visual searches in medicine and homeland security.

Visual screening is critical to such things as early cancer diagnosis and airport security, but paradoxically the more rare the object being searched for becomes, the lower the screeners' accuracy in finding it when it is there. Screeners also tend to respond more quickly when the targets become more rare.

“Even though they're not under time pressure, people tend to hurry,” said Stephen Mitroff, an assistant professor of psychology and neuroscience at Duke University. “They get used to not seeing anything and they're just hitting that key over and over.”

When people are asked to search for an item that will appear only once in 100 images, they might miss as many as a third of the objects they're supposed to be finding. Studies of radiologists looking at images to find cancer have shown similar error rates.

But those error rates fall by more than half if test subjects are given the opportunity to immediately reconsider and correct their choices, according to a new study by Mitroff and Duke graduate student Mathias Fleck that appears in the November issue of the journal Psychological Science.

“The high error rate in studies of rare targets appears to come from what we call execution errors – the observers did notice the target, but they responded too quickly,” Fleck said. The sensation should be familiar to most web- and channel-surfers.

But that's not all there is to it, Fleck added. Studies that tracked the eye movements of trained radiologists found their gaze tended to linger longer on a trouble spot in an image, even when the radiologist went on to say the picture was clear. Their visual system partially picked up on the problem, but the viewer ultimately didn't recognize the target, Fleck said. So, rare-target errors are both in perception and action.

Each issue should be dealt with separately to improve airport security and cancer screening, Mitroff and Fleck recommend. Perhaps moving the airport screener to a spot where somewhere where the screeners can't feel the unconscious time pressure of seeing a long line of travelers would help.

The next phase of the Duke research will compare experienced video game players against non-gamers in visual searching tasks. Fleck's hunch is that the gamers, who spend a lot of time coordinating their hand movements with visual decisions, have a higher accuracy rate at the search-and-click tests. Whether their improved accuracy comes from being able to see faster or to click slower remains to be seen.