Re: cross,parallel, and wall-eyed viewing

[P3D John W Roberts <roberts@xxxxxxxxxxxxxxxxx>]

>Date: Tue, 12 Nov 1996 19:16:41 -0600
>From: P3D Gabriel Jacob <jacob@xxxxxxxxxxxxxx>
>Subject: cross,parallel, and wall-eyed viewing

>In the same issue of Pop Science that I mentioned before (Dec.96) there
>is another article called FYI (for your information). It is in a 
>question and answer format. In it one of the questions is ,"Why can we
>cross our eyes inward but not outward?".
>The response they give is as below.

>>>We certainly have all the nerves and muscles needed to make our eyes
>move outward, otherwise we'd never be able to look sideways. but we can't
>simultaneously turn them outward, because that's not on the list of brain-
>approved activities. Eyes straight forward is the position that offers
>the maximum field of binocular vision. To see something up close, though,
>the eyes automatically turn inward a bit-crossing your eyes is merely an
>extreme tension of that action. but if both eyes were to turn outward, 
>binocular vision would be lost, something the eyes, "central control
>won't allow," explains Dr.William Power, an instructor in ophthamology
>at Harvard Medical School.<<

>This can be found on page 120. Now I am puzzled. Most people here on P3D
>can freeview cross-eyed and parallel-eyed, including myself. But viewing
>wall-eyed (not sure if this is the proper term) or with both eyes turned
>outward, is harder. 

I can think of a possible explanation - it gets slightly into control theory,
and uses spacecraft/telescope design and control as an analogy.

First the analogies:
 - When large telescopes started to be built in astronomical observatories,
   tracking the stars being observed turned out to be a problem. The simplest
   kind of mount is an "alt-azimuth" mount, with separate axes for left-right
   motion and up-down motion. Unfortunately, to track a star requires a complex
   relative motion between the two axes, which varies with the location of the
   telescope and the position of the star. The solution was to come up with a
   new kind of mount in which declination and right ascension were controlled
   by different components. Then for a fixed object such as a star you could
   just line up the telescope once, then track at a constant rate to compensate
   for the Earth's rotation, and the star would stay in the field of view.
   The problem with this newer kind of mount is that it's very heavy and
   expensive. In recent years, many (all??) new major telescope designs have
   gone back to alt-azimuth, with computer control to handle the tracking -
   now that computers are available, it saves a lot of money.

 - When unmanned interplanetary spacecraft are launched, they leave behind
   a human support team on the ground to handle contingencies. The designers
   of the spacecraft try to provide for problems that they think may occur,
   but it seems that almost always something comes up that requires human
   intervention. The solutions have to make use of the physical resources
   available on the spacecraft, which means that sometimes components are
   used for other than their original intended purpose. This may place a
   strain on the system, but that may be better than the risk of a mission
   failure. For instance, when the Galileo main antenna failed to open,
   the opening motor was used to "hammer" the antenna to try to force it
   open. This was risky, but if it had worked, it would have been of great
   benefit. When it was discovered that some of Galileo's thrusters tend to
   overheat and explode, the solution was to operate them in thousands of
   brief pulses rather than long continuous thrusts. This put a strain on the
   control mechanisms, but is was better than blowing up the spacecraft.
   The programmed mission profiles in these spacecraft are often rewritten
   as mission objectives and resources change - Galileo was reprogrammed to
   use its low gain antenna to transmit images, and it was directed to observe
   the impact of Comet Shoemaker-Levy with Jupiter.

Now for the tentative explanation:
  It is possible that the medium - or high-level control mechanisms of the 
  brain do not address the sideways motion of the two eyes as "left eye
  move left-right" and "right eye move left-right", but as "point of view
  move left-right" and "convergence". It's even possible that the low-level
  muscular control sections are wired up that way. Getting the eyes to
  diverge would therefore require either the ability to program in voluntary
  decoupling of part of the convergence mechanism, or application of some
  other control mechanism, or both. Decoupling the convergence mechanism in
  any way is risky, and has to be approached with caution and following a
  prescribed sequence, lest you risk losing some of your depth perception.
  Imposing an additional control mechanism on top of the control mechanisms
  can require a mental effort, and could actually be a physical strain -
  for instance if during divergence two sets of muscles are working against
  one another.
That proposed explanation would seem to address the described phenomena,
though how much of the strain is "mental" and how much "physical" is
not clear. I learned how to free-view Holmes stereoscope prints - it's
not much of a strain, but I wouldn't want to try further divergence, and
I don't do it for long sessions.

>Plus I thought that 3d would still be observable. 

It should be - convergence is not the only visual depth cue.

John Roberts


------------------------------