3-D TV Disparities Said to Cause Physical, Mental Strain

Visual disparities in 3-D TV images can cause physical strain in viewers, according to recent research at the University of California at Berkeley. While proper viewing conditions can help avoid most problems (dark room, central seat far from the screen), 3-D cinematographers also need to follow careful rules or risk making people sick, researchers said. The debate over 3-D TV heated up earlier this week when Samsung issued a warning about possible health effects.

Even if the physical strains of 3-D are avoided, other disparities can cause mental strain akin to vertigo, according to other investigators at the University of Washington. "If [cinematographers] confined 3-D to animations, then there wouldn't be a problem [since] the brain doesn't have the same expectations about cartoons," said Aris Silzars, founder of Northern Lights, a display technology consultancy (Sammamish, Wash.)

For the emerging crop of 3-D movies that incorporate real-world scenery, movie makers have one extra job beyond those of previous animations: Minimizing the so-called vergence-accommodation conflict. The conflict arises from the fact that 3-D displays often cause distortions in perceived 3-D structure compared with the percepts of the real scenes the displays depict.

"The only thing we have any data on is what we call the vergence-accommodation conflict, which our lab has shown really does cause fatigue, discomfort, eye strain and headache in some cases," said professor Martin Banks, who led the research into 3-D eye strain at the UC-Berkeley, which was funded by the National Institutes of Health.

According to Banks, when viewers direct their eyes at nearby objects or scenes, their gaze converges. When they gaze into the distance they, diverge, or what optometrists call "vergence." Conversely, focusing the eye muscles to bring something into sharp focus is called "accommodation."

In the real world, vergence and accommodation are synched to the same distance, but in the world of 3-D stereoscopic glasses decouple the two, forcing the brain to cope with a disparity between the vergence and accommodation distances.

"Normally the distance to which you converge your eye and accommodate your focus are the same, so understandably your brain has coupled these together," Banks said. "The problem is a stereo or 3-D display breaks this coupling, and the reason why is that the disparity between the images that are being presented to the two eyes might specify something behind the screen or in front of the screen. In those two cases you have to converge your eyes to a different distance than the screen, but you still have to accommodate the screen because that is where the light comes from."

The Berkeley researchers have performed two studies that reached the same conclusion: 3-D TV induces a vergence-accommodation disparity in the brain that manifests itself in a statistically significant number of people as fatigue, discomfort, eye strain and headaches. So far, test subjects have been between 18 and 30, but Banks plans to test 50-year-olds to determine whether they are immune to the vergence-accommodation conflict, as he suspects, since older eyes do not focus as well.


A stereo viewing chair that rotates was designed by UC-Berkeley researchers to allow manipulation of visual and vestibular cues to self-motion and body orientation


Banks recommends two solutions to minimize eye fatigue and other symptoms: viewers should sit very far from a 3-D TV screen as they would in a theater. The farther from the screen, the smaller the difference between the convergence and the accommodation distances, Banks said. The second is to shoot 3-D videos so that action is located at the front of the screen, not off one side or projecting out of the screen into a crowd. By minimizing the vergence-accommodation disparity, the possibility of fatigue and other symptoms will recede in most people, according to Banks.

"The better 3-D movies, like Avatar and Pixar's Up are clearly trying to minimize this conflict with the way they do the cinematography," said Banks.

Even if the theater experience is duplicated in the home, not all of the potential problems associated with 3-D TV will be resolved, according to University of Washington professor Robert Patterson. Patterson and Silzars presented their results at the 2009 Society of Information Display. According to the researchers, three other cognitive disparities can cause fatigue in addition to vergence-accommodation: binocular disparity, linear-perspective disparity and texture perspective disparity.

For instance, if a 3-D scene depicts a football player running the length of the field, depth has to be compressed into a few yards by the cinematographer to prevent too much parallax, causing viewers to see double-image ghosting. Linear perspective (geometric angles) and the texture perspective (distant objects that are less detailed) signal the brain that the distance traveled was 100 yards, thereby creating cognitive disparities.

"The reason why high-level cue conflict leads to discomfort in immersive stereo displays is that the intuitive reasoning system is attempting to make reasoned sense out of this incoming, conflicted perceptual information," Patterson and Silzars wrote.

These higher-level cues create conflicts in the brain that causes mental strain in addition to the physical eye strain of the vergence-accommodation disparity, they added, which becomes intolerable over time, resulting in a confused state of mind akin to vertigo.

"When you take away cues, you create in the brain what is called the 'doll house' effect. The scene can be very precisely constructed, but it does not look real (it looks like a doll house) because your brain is saying that its missing cues that are always present when viewing real scenes," said Silzars. "It creates a vision-to-brain conflict that can make you feel nauseous and dizzy or worse."

Worse still, according to Patterson and Silzars, is that the closer a set of cues comes to matching reality, the more discomfort viewers experience. That means the problem could get worse as cinematographers attempt to make 3-D experiences even more realistic. One solution, according to Silzars, would be to confine 3-D content to animations since the brain does not expect cartoons to contain the same realism as normal video or film.

"In Avatar for instance, the scenes that were shot with real backgrounds were less convincing than the animated scenes because the brain has lower expectations for animations," said Silzars.

By R. Colin Johnson, EE Times