Follow me

Thursday, October 4, 2012

Binocular Rivalry and Head-Worn Displays


Robert Patterson, Byron-Pierce, Robert Fox
Human Factors: The Journal of the Human Factors and Ergonomics Society
Summary
                  A number of virtual system technologies are used for augmented reality through the use of artificial devices, once of the most advanced concepts in this technology is the Head Worn Display (HWD). When information is presented to only one eye then it is called monocular HWD, if the same information is displayed to two eyes than it is defined as a biocular HWD, which is different from the Binocular HWD which is displayed with binocular parallax. Different kind of devices are in the market with different applications and several advantages, but it is not a system without problems, in fact it is known that it can cause headaches, nausea, eyestrain and dizziness, studies have been performed on IHADSS (Integrated Helmet and Display Sighting System), used for a large number of helicopter pilots lamenting problems. The problems are mainly due because of interrocular differences, which is when the two eyes receive different stimulation because of unnatural viewing. When two eyes receive different stimulation for each eyes, then binocular rivalry will preclude binocular fusion, bringing a state of competition between the two eyes.
Luning is the phenomenon which refers to the case in which the counter in one eye continuously covers the background area in the other eye, this, together with binocular rivalry consists of the most troublesome problems in partial-overlap displays defining conditions of exclusive visibility and mixed visibility (respectively the case of one eye’s image exclusively visible and the case of portions of the two eyes sight visible).
Binocular rivalry is mainly caused by interocular difference in orientation, hue, luminance, contrast polarity, form, size and motion velocity, light levels etc… Rivalry is generally not provoked by stimuli lasting less than 200ms, or flickered repetitive stimuli. It generally happens to both monocular and binocular HWD’s with partial overlap and it is still questioned whether HWDs should be calibrated with the users sighting dominance. It is demonstrated (Melzer and Moffict, 1997, and Klymenko, Harding, Beasly, Martin and Rash, 1999) that luning appears less with convergent design and that overlap also results in a better performance (convergent binocular overlap is when the left eye view the right monocular flanking region and the right eye views the left monocular flanking region, while divergent binocular overlap is when the left eye views the left monocular flanking region and the right eye views the right monocular region).
Rivalry could be reduced through different proposed solutions, a significant solution propoes by Kooi (1993)  is with the use of fusible window frame in both binoculary views and monoculary viewsd scenes.
Under the point of view of the stimulus, it is demonstrated that a stimulus in one eye will dominate a rival stimulus in the other eye if the former possesses a greater contour density, higher contrast, a wider range of spatial frequencies or faster motion. It must be deeper investigated whether presenting information on a Monocular HWD will reduce possibility of binocular rivalry or if it would mask information from real word (visible thank to transparency of the information displayed).
The use of repeated brief exposures and high stimulus contrast can minimize or eliminate suppression of the displayed information, opaque monocular HWD should not be used for this reason.
Two kinds of cognitive variable have also been investigated: stimuli which are familiar or meaningful and those which engage voluntary attention.
Blake in 1998 employed a dichoptic reading paradigm in which one eyes views meaningful text and one non-meaningful text, the experiment showed that meaningful text had not special status and the user would loose its attention on it.
It is also demonstrated that practice over 10 days may help individuals control the rate of rivalry alternations, being able to manage with it. Target detection and recognition has been proved to be delayed due to binocular rivalry form some simple task, but influence on performance has to be verified.
Key Concepts
Binocular rivalry, Augmented Reality, Virtual Reality

No comments: