Scientists studying the way cell-phone use interferes with driving performance say it's not so much dialing and driving one-handed that distracts motorists and causes accidents.
"There's lots of evidence it's really the content of the conversation, and where it puts you mentally, that is the problem," said Steven Yantis of the Johns Hopkins University department of psychological and brain sciences.
If so, legislation now under consideration in Maryland requiring drivers to use hands-free equipment when they talk and drive wouldn't solve the problem.
A bill introduced by Del. John S. Arnick, a Baltimore County Democrat, would ban non-emergency cell-phone use by drivers unless they have hands-free equipment. It's modeled on a novel New York law that took effect in December.
Support for Arnick's bill has grown since the deaths of five people in an accident Feb. 1 on the Capital Beltway. The cause is still under investigation, but the driver who triggered the accident was reportedly on her cell phone when she lost control of her newly purchased sport utility vehicle.
Similar legislation has failed in three previous sessions. And members of the Commerce and Governmental Matters Committee appeared divided on the matter at a hearing on Tuesday.
Opponents, including most cell-phone suppliers, argued that there are no accident data to support a ban on phoning while driving. Cell phones, they said, are no more distracting than passengers, and the devices are useful in emergencies.
But many scientists say talking on the telephone - whether hand-held or hands-free - draws significant mental resources away from the driving task, and in ways that passengers and car radios do not. When drivers are on the phone, they say, signals are missed, reaction times slow, and accidents are likely to increase.
"Attention is a limited-capacity process," said neuroscientist David Poeppel of the University of Maryland. "You can switch willy-nilly, back and forth, but you can't do both [tasks] at the same time."
More than 123 million Americans use cell phones, according to industry estimates, and studies show that 60 percent of their cell time is clocked in automobiles.
Data linking car wrecks to cell phone calls are scarce. But tales of near-misses and spectacular wrecks have sparked calls for new legislation in 20 states.
One of the most frequently noted scientific studies of cell-phone use and driving was published in the New England Journal of Medicine in 1997. Donald A. Redelmeier of the University of Toronto and Stanford University's Robert J. Tibshirani studied 699 cell-phone users in Toronto who were involved in wrecks in 1994 and 1995. The scientists found the drivers' risk of a collision quadrupled when they used their cell phones - comparable with driving with a blood-alcohol level at the legal limit.
The study found hands-free phones didn't cut the risk, suggesting that dialing and handling the phone were not the only source of the driving impairment.
Many researchers have looked for the other causes. David L. Strayer of the University of Utah asked 48 college students to track a target on a computer screen with their cursor. When a red light appeared, they were to press a "brake" button.
In a paper published last year in the journal Psychological Science, Strayer reported that the probability of missing a red light doubled when the students were talking politics on a cell-phone, and their reaction times slowed.
There was no impairment when the students were listening to the radio, and none when they were asked to passively repeat words they were given on the telephone. But when asked to respond with new words that began with the last letter of the words read to them, the students showed a significant increase in tracking errors.
Strayer concluded that the more demanding the phone conversation, the more likely it was to degrade driving performance.
But why should telephone conversations be any more distracting than those with passengers?
Lee Gugerty, a professor of psychology at Clemson University, studied the question with a group of students using a driving simulator. In a paper he's preparing for publication, he reports that both types of conversations degraded driver performance.
But he also found that driving degraded the phone conversation.
"When drivers were conversing over the cell phone they talked slower than when they talked to the passenger," he said. "It seems to be more difficult to have a cell phone conversation than when someone is right beside you."
But why? "We don't have a real firm idea yet," Gugerty said. Some scientists have suggested that the passengers "modulate" their conversation - slow down or stop talking - when they see the driving environment becoming more difficult, making it easier for the driver to both drive and talk.
"The person on the cell phone may not be able to modulate because he can't see the driving scene," Gugerty said.
His study seemed to support that hypothesis. "We found that the passengers had more extremely long pauses in their conversations than did the people talking to the drivers on the other end of the cell phone."
Passengers modulated the conversation so well the drivers didn't need to. But when drivers were on the telephone, the pauses in their conversations increased. It was as if the two tasks - driving and talking on the phone - interfered with each other.
"That was something a little bit surprising," Gugerty said. Driving is a very spatial task, and conversation is a linguistic task, and it was thought the two would not conflict. "But we're finding they definitely do conflict."
Carnegie Mellon psychologist Marcel Just may have captured pictures of that conflict in a brain imaging study reported last summer in the journal NeuroImage.
Eighteen people were given two simultaneous linguistic and spatial tasks - listening comprehension, and the mental rotation of shapes. Just monitored their brains with a functional magnetic-resonance-imaging machine that measures thinking effort by mapping blood flow in the brain.
Instead of doubling their activity, or handling the two tasks normally in separate locations, the subjects' brains slowed both chores. Visual-spatial processing slumped by 29 percent, and language processing fell 53 percent.
The two mental activities seemed to drain a common resource, Just said. "It could be a neurotransmitter, or some nutrient provided by the vascular system." No one's sure.
Whatever it is, when the demand exceeds the brain's resources, performance slows. "It's like what happens in a brownout," he said. "There's not enough electrical supply to go around, and every function gets a little less."
The bottleneck, said Johns Hopkins' Steven Yantis, may lie in the prefrontal cortex behind the forehead. That's where the brain decides what - amid the flood of data arriving from the senses - to attend to, process and act upon, and what to suppress. On a familiar route, Yantis said, many of us can daydream, or talk on the phone, and arrive safely without really knowing how we did it.
But when we need to make time-critical driving decisions - if a deer or a child steps into our path - "if that's delayed even half a second because we're retrieving something from memory, that short delay could result in an accident."
"I don't see a good, general-purpose solution," he said. "It's just a human limitation."
We could ban cell phone use at rush hour, or in high-traffic zones, some scientists suggested. Or, it may come down to education, and individual decisions balancing convenience and good sense.
Most cell users know their driving suffers when they're on the phone, Poeppel said. It's "introspectively obvious." At the same time, "we all like to use the cell phone, and we'd be pretty depressed if we can't."