The Dual Nature of Stereopsis – Part 2

Several readers shared such good observations through comments to Part 1 that I decided to do a Part 2.  You don’t normally think of Borish’s Clinical Refraction as a source for deep discussion about stereopsis, but I had the occasion to consult old reliable (the 2nd edition  was published in 2006) about something else, only to be reminded of how good the treatment of stereopsis is in the book!  First, a comment by J.J. Saladin (Chapter 21) who quotes Griffin and Grisham in describing stereopsis as the barometer of binocularity.  Saladin was so taken with the term that he titled his section on stereopsis:  The Barometer of Binocularity and Visual Function.  He goes a step further regarding stereopsis and says: “It is the single best indicator of the overall function of both the sensory and motor portions of the visual system.”

Second, in their chapter on Fusion and Binocularity, Daum & McCormack write:  “The purpose of binocularity is to enhance the vision over what it would be monocularly and particularly by way of stereopsis.  Stereopsis, which means ‘seeing solidly’, contributes to the judgment of depth and distance and participates in the recognition of solid objects … As compared with monocular viewing, binocular vision and stereopsis also help provide better motor control (e.g. when reaching for a target or completing fine motor tasks); they also provide quicker and more accurate cognitive information.”

Daum and McCormack differentiate depth and distance perception, both processes operative in the perception of three-dimensional space.  Distance perception is the judgment of how far something is from you or from some other reference point or object in absolute terms of measurement.  The wall is 10 ft. from me; or the picture is 6 ft. from the ground.  Depth perception is a relative term indicating the space between objects or the relative depth intervals between a number of different points in space that can be judged by monocular cues or binocular cues, with stereopsis being a limited to binocular depth perception based on horizontal disparity cues.

Here’s a classic picture of railroad tracks extending into the horizon.  You might make reasonable judgments about absolute distances along the way, and you’d be doing this based largely on monocular cues.

Here’s a picture of an outdoor scene in which you could make some reasonable estimates of the distance between you and the table, you and the edge of the deck, between the table and the edge of the deck, about the depth of the pile of leaves and so forth, based on monocular cues.   But it would be easier not to mention more enjoyable sizing up these relationships based on binocular cues.

And here’s a conventional vision therapy setup, with binocular stereoscopic cues arranged by having two fusable vectograms, one seen by the left eye and one by the right eye, and vary the degree of lateral separation.  We don’t concern ourselves with judging absolute distance between points.  Rather we’re focused on the relative depth intervals and judgments.  Is the Quoit (rope) closer to me than where the the silver ball (on Woolf Wand) appears to be or further?

Let’s talk for a moment about monocular cues to depth.  We can divide them into static cues and dynamic cues.  This picture represents a static cue – texture gradients.  As the textural elements of the bales of hay get further from us, they appear smaller and closer together, providing a cue to relative distance.

Here’s another monocular, static cue, superposition or occlusion – which leads you to interpret that the square is relatively closest to you, the circles is the furthest, and the triangle is in between.

And in this one, we lump together shadow, luminance and linear or aerial perspective that give you a strong sense of being at the top of the stairs looking downward with a sense of space between each step.

So much for static monocular cues.  Here’s an example of kinetic or dynamic monocular cues, clouds looming over Clearwater Beach, FL.  Is the storm approaching or receding?  The motion will give you a clue.  Here’s some nice looming video for you.

Two other kinetic cues are motion parallax and the kinetic depth effect.  Motion parallax is induced by differential retinal motion that is stimulated by head motion.  Let’s do motion parallax.

Close one eye and hold the two index fingers of each hand up one slightly higher and behind the other.  Fixate your far finger and as you move your head side-to-side you’ll see that your far finger appears stationary while your near finger appears to move opposite to the direction of your head.  Switch to fixating the near finger and the far finger appears to move in same direction as your head is  moving.  

Here is a monocular stereokinetic depth effect that is a particularly strong example of the kinetic depth effect.  In contrast with motion parallax, this monocular cue is caused by movement of the stimulus rather movement of your head.

Think of all these and other monocular cues to depth that serve the patient with strabismus or suppression, or those patients who are labeled as “stereoblind” because they cannot appreciate random dot stereograms.  This serves as the back drop for envision how their binocular stereoscopic cues or stereoscopic three dimensional cues (S3D) now have to be weighted together with monocular cues to depth in various contexts and scenes.

– Leonard J. Press, O.D., FCOVD, FAAO

7 thoughts on “The Dual Nature of Stereopsis – Part 2

  1. Interesting comments from those old fossils….. One of our “Stereo Sue” patients described the circle behind Topper suddenly one day as a “globe.” I have always had good steropsis and call Topper’s background the “moon.” In the night’s sky, I attend to the moon’s waxing and waning and think of it as 2D rather than a 3D solid.

  2. That’s really interesting, Beth. I never thought of Topper’s circular backdrop as a globe, as it is very clearly limited to a single plane when viewed stereoscopically. That gives insight into your patient tending to think of things as having a 3D contour, even when only 2D can be appreciated. Prognostically, I’d say that’s a good sign.

    I really appreciate your insights on your perception of the moon as being 2D based on what you *attend* to. I do the same.

    And conversely, it is a reminder of why, when I ask if a vectogram appears like it’s moving, many patients with strabismus will observe its lateral displacement “left or right,” even once they have started to appreciate its z-axis displacement forward and back. I try to control for this with symmetric movement of the vectogram slides, but with a strong ocular dominance off-center, they may continue to perceive lateral displacement.

    Samantha Slotnick, OD, FAAO, FCOVD

    • Agreed, Samantha. Something we’ve onli scratched the surfacce of is the stereoscopic volume of space that constitutes the Vieth-Mueller Circle. My hunch is that as strabismus patients convert from the strong oculocentric shift you mention, it takes awhile before getting all the “stereopscopic vectors” to aggregate into a symmetrical volume of space that constitutes a true binocular V-M circle. When I used to listen to the old masters talk about space volume, that’s what I envisioned.

      • Interesting, good point, Len. Of course, oculo-motor asymmetries, paretic muscles, etc., would also contribute to asymmetric weighting of the horopter.

        But that horopter… now that’s a scary word… not such a juicy blog-topic for most….

        Samantha Slotnick

  3. Len,

    This truly is a wonderful addition, detailing the ways in which monocular cues — static and kinematic — contribute to depth perception. And yet “depth perception,” per se, still does not approach the profound experience of steropsis, the perception of solid space.

    I just wanted to take a moment to point out that if you take a moment to view each of these 2-D images, first with 2 eyes, then closing an eye, you ought to immediately appreciate a sense of increased depth. (If not, consider a binocular vision evaluation!)

    Monocularly, the “line” of the bus picture draws your eye immediately down the length of the bus. The blur effect at near adds to a profound sense of depth.

    Monocularly, the railroad tracks drop down and back, away from the viewer.

    Monocularly, the porch provides a pronounced sense of local space, practically inviting a walk to the corner of the balcony. And the tree in the background drops farther back, separating from the bannister.

    Monocularly, the boy at the table separates in space from the instrument which is to his left, also along the table.

    Monocularly, the farm seems less oblique, less angled to the right, and the far end of the image recedes back and centers itself somewhat.

    — The superposition diagram is somewhat flawed, as the linear perspective is contrary. Relative to the horizon, we judge depth with items below the horizon appearing closer to us, the lower they are in our visual field. In the case of the square and the triangle, note that their bases are at the same height. Therefore, they appear to be in the same plane, particularly with one eye closed. The circle, however, does not offer a base location (as it is occluded). Therefore, in this image, the circle appears slightly receded while the square and triangle are immediately on top of one another, and lack any depth. They are 2-D shapes, not geometric solids.

    Monocularly, the aerial view of the stairs took a moment (for me) before the steps separated. This is more subtle, perhaps because it is at an unusual angle. The texture at the ledge of each step contributes to texture gradient changes. In addition, the assumption that each step is the same width but appears to get smaller moving up from the bottom of the picture is another indication of distance (like train tracks). This coincides with increasing shadow in the distance.

    Monocularly, the beach scene offers a greater fullness to the clouds (due to shadowing), but this particular picture lacks the resolution to offer much depth from texture changes with ocean distance.

    Monocularly, even the “motion parallax” demo becomes more prominent, with the use of relative size of the two hands, and the line of the left arm.

    Monocularly, the 2D images in the video are, by-and-large, similarly monocularly depth-rich. (The ocean liner was not at a telling angle, and the texture of the waves is not always even and predictable, so it was the weakest of the lot).

    For those interested in obtaining an in-depth read on stereopsis, I strongly recommend Saladin’s review article, published in 2005:

    Saladin JJ. Stereopsis from a performance perspective. Optom Vis Sci. 2005 Mar;82(3):186-205.

    Thanks again for the exploration!

    Samantha Slotnick, OD, FAAO, FCOVD

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s