Mike Watkins admits that he used to text and drive. But Watkins, a physicist and manager of Pacific Northwest National Laboratory’s Applied Physics and Sensing Technologies division, doesn’t any more.
And not just because of Washington’s laws that prohibit texting and driving.
Watkins own research has shown him how distracting texting can be. The scientist has developed a computer algorithm that gives the cell phone the ability to recognize when a driver is sending a text message.
A survey released in December by the National Highway Traffic Safety Administration reveal that 18 percent of all drivers admitted to sending texts or emails while operating a vehicle. And nearly half of drivers from the ages of 18 to 24 admitted to texting and driving.
The agency also found that the practice increased by 50 percent in 2011, despite the fact that 35 states have laws banning texting and driving.
In late 2009, Watkins was watching the news when he heard Transportation Secretary Ray LaHood refer to texting while driving as a ‘deadly epidemic.’
“I started thinking about how you might find out if someone was texting and driving,” he said.
Watkins said he knew that he had difficulty texting while he was driving and he wondered if others had the same problem. He started reading about cognitive functions and discovered that he probably wasn’t the only person who had that problem.
Although there is already technology to identify and measure driver impairment, it often requires complicated in-vehicle equipment.
So Watkins set out to develop a way to measure driver distraction using an algorithm that enables cell phones to recognize when a driver is texting, as opposed to a passenger.
Watkins and his team at PNNL discovered there is a distinct difference in the way a driver texts, as opposed to his passenger.
To start the experiment, Watkins gave a team of six volunteers identical phones and several days to get used to texting with them. Then he had them send identical texts to him, so he could record their keystrokes and the pattern and rhythm of their texting, to create a baseline.
Then he had them step into a driving simulator and text.
“We found the time between keystrokes changed significantly,” he said. “When driving and texting, the keystrokes become erratic as our focus shifts between the tasks.”
This cognitive “time-sharing” is seen in the erratic driving patterns of the person who is texting and driving. But it can also been seen in the texting pattern, Watkins said.
Watkins said the relevant changes in texting are quantifiable and can be measured on the cellphone using an algorithm. The data being measured never needs to leave the person’s phone. And it doesn’t matter what language it is being texted in.
The technology is 99 percent accurate and has a variety of potential markets, Watkins said.
In 2009, 5,474 people were killed in the U.S. in crashes attributed to driver distraction and 448,000 were injured, according to www.distraction.gov, a government website focused on the issue. Previously, the best way to track driving while texting used a GPS device to follow how fast the phone was moving. But that method could potential tag passengers as a distracted driver.
This approach could allow software developers to create cellphone applications that provide feedback to a driver to alert them of impaired reaction times, in the same way speedometers provide information to drivers. It can also be extended to other uses.
“You could potential use it to identify fatigue in heavy-equipment operators, or to detect an anomaly in grandma’s cognitive function due to a stroke,” Watkins said.
The technology could be used by employers who want to reduce their liability risk of employees potentially driving distracted, or insurance companies, or parents who want to know if their kids are texting and driving.
“It’s all up to how the market wants to use it,” Watkins said.