How does the eye detect the angle of incidence?

[lugita15]

I don't really know too much about the human body, but this is my rough understanding of how vision works: light comes from a source, like the sun, it hits an object, and then if it's of the right frequency it's reflected by the object and then enters one of your two eyes, where it is refracted by a series of lenses until it's projected onto the retina, where it is absorbed by a photoreceptor cell, which sends a signal via the optic nerve to the brain, which compares the two-dimensional data from both eyes in order to estimate the distances of the various objects, and then constructs a 3-dimensional construct of the world.

My question is, how does the eye get the necessary information to create a two-dimensional projection of the world? A beam of light comes into the eye and is absorbed by a photoreceptor cell, but how does the photoreceptor cell know what angle at which the light hit it? Assuming the retina is a spherical surface, it seems to me that you need two pieces of information: what is the angular position on the retinal surface at which the light hits, and what is the angle at which the light hits the surface. Only then can you deduce the angular position of the object that reflected the light, which is what you need to make a two dimensional projection. (Because light from two different places can end up hitting the place on the retina, just at different angles.) So what is the mechanism for finding out the angle of incidence?

Any help would be greatly appreciated

Thank You in Advance.

 

[SteamKing]

Well, most people have two eyes, and what we perceive visually is a result of processing the optic nerve impulses by the visual cortex of the brain. Can't see without a brain.

Folks who can see with only one eye have problems with depth perception.

 

[lugita15]

Well, most people have two eyes, and what we perceive visually is a result of processing the optic nerve impulses by the visual cortex of the brain. Can't see without a brain.

Folks who can see with only one eye have problems with depth perception.

Yes, but isn't that a separate issue? That's an issue of how the two-dimensional projections from each eye are compared with one another to estimate the distances from the eye of various objects, resulting in the formation of a 3-dimensional image in the brain. But I was asking about an earlier step in the process, the formation of the two-dimensional images.

To produce a 2D image of the world as seen by, say, the right eye, we need the angular position (not the distance) of every object relative to the right eye. And to find the angular position of an object relative to the right eye, we need to know the place on the retina that the light beam strikes, and we need the angle at which the light beam hits the retina. The first piece of information is gotten by knowing which cell the light was absorbed by. How is the second piece of information obtained?

 

[D H]

Emphasis mine:

it seems to me that you need two pieces of information: what is the angular position on the retinal surface at which the light hits, and what is the angle at which the light hits the surface. Only then can you deduce the angular position of the object that reflected the light, which is what you need to make a two dimensional projection. (Because light from two different places can end up hitting the place on the retina, just at different angles.)

That is incorrect. Draw a picture.

 

[lugita15]

Emphasis mine:
That is incorrect. Draw a picture.

Here is a picture. I apologize for my horrendous drawing skills, but I hope you get the basic idea. The sun shines two rays at two cubes, each of which reflects light to the same point.

EDIT: To clarify, are you saying that for any point on the retina, there's only one angle of incidence that light hitting that point can have?

 

[Pythagorean]

angle of incidence = angle of reflection, so you made a mistake in that drawing (your top block ha a small angle of incidence and a large angle of reflection).

 

[UltrafastPED]

The important thing about the eye is that the retina is the focal plane for the lens. The image plane is exactly the 2D projection that you are looking for ... and it all works because the eye is a "good lens", though some of us need a few corrections.

Here is a complete lecture: http://homepages.inf.ed.ac.uk/rbf/C.../LECT1/node2.html#SECTION00021000000000000000

 

[lugita15]

angle of incidence = angle of reflection, so you made a mistake in that drawing (your top block ha a small angle of incidence and a large angle of reflection).

That's just bad drawing on my part. But that can be easily remedied by rotating the block a bit, so that the rays I've drawn will obey angle of incidence = angle of reflection.

 

[Pythagorean]

if you rotate the block, you are now getting the source from a different point on the sun :)

 

[lugita15]

The important thing about the eye is that the retina is the focal plane for the lens. The image plane is exactly the 2D projection that you are looking for ... and it all works because the eye is a "good lens", though some of us need a few corrections.

Here is a complete lecture: http://homepages.inf.ed.ac.uk/rbf/C.../LECT1/node2.html#SECTION00021000000000000000

Here is my basic question: how does knowing the place on the retina that the light beam struck allow you to calculate the angular position of the object?

 

[lugita15]

if you rotate the block, you are now getting the source from a different point on the sun :)

No, you aren't. I'm keeping the location on the sun fixed, and I'm keeping the point of contact between the light beam and the block fixed, and I'm rotating the block about the point of contact.

 

[UltrafastPED]

You need to understand ray tracing - then the angular position will be clear.

Here is a simple review: http://hyperphysics.phy-astr.gsu.edu/hbase/geoopt/raydiag.html

 

[lugita15]

You need to understand ray tracing - then the angular position will be clear.

Here is a simple review: http://hyperphysics.phy-astr.gsu.edu/hbase/geoopt/raydiag.html

I do understand ray tracing. But my confusion is this: I cannot imagine any lens, or sequence of lenses, which takes two different rays, each with its own point of contact with the lens, and each with its own direction, and is guaranteed to always take them to two different points on the retina. It seems like for any lens, you can find a pair of rays that violate this condition, for the simple reason that you can control the angle of refraction by controlling the angle of incidence, so you can adjust the angles of incidence until the angles of refraction are such that the refracted rays reach the same point.

 

[UltrafastPED]

The secret is to find the image plane! It is the one set of locations where all of the tangling is untangled.

Here is a good summary of geometric ray tracing; there are several sections on image formation: from mirrors, spherical mirrors, and lenses. Your specific question is answered in "Image location by ray tracing" (page 103):
http://spie.org/Documents/Publications/00%20STEP%20Module%2003.pdf

 

[Pythagorean]

Here is my basic question: how does knowing the place on the retina that the light beam struck allow you to calculate the angular position of the object?

A single point doesn't tell you anything. It's higher level processing of the whole picture that allows organisms to infer position.

 

[D H]

I do understand ray tracing. But my confusion is this: I cannot imagine any lens, or sequence of lenses, which takes two different rays, each with its own point of contact with the lens, and each with its own direction, and is guaranteed to always take them to two different points on the retina. It seems like for any lens, you can find a pair of rays that violate this condition, for the simple reason that you can control the angle of refraction by controlling the angle of incidence, so you can adjust the angles of incidence until the angles of refraction are such that the refracted rays reach the same point.

That's exactly how lenses work, lugita. There are two distinct areas of confusion on your part. One is that you don't understand how lenses work. That other is that you apparently don't know how reflection, particularly diffuse reflection, works.

I'll start with reflection. There are two basic kinds of reflection, specular and diffuse. Specular reflection: That's how mirrors reflect light. "Angle of incidence = angle of refraction". Diffuse reflection: That's how many objects that occur in nature reflect light. In pure diffuse reflection, the light reflected from the surface is independent of the angle of incidence. The intensity depends on the angle with the outward normal to the surface. The following diagram depicts reflection from a surface that exhibits both specular and diffuse reflection:

When you look at some point on an object, some of the light reflected by that point will enter your eye at the top of the lens, some will enter at the bottom, and so on. If the object is in focus, the light that comes from that point will converge on the same point on your retina, regardless of where it entered your eye. Another diagram:

Diagram A shows how light rays emitted by a distance object such as a star enter your eye and are focused on the retina. The rays from that distance source are essentially parallel when they enter your lens. Your eye muscles have to shape the lens so that it focuses all of those parallel rays on a single point. Light from a different star will come from a different angle. That light from that other angle will be focused on a different point on your retina.

Diagram B shows what happens to light from a nearby object when the lens is shaped to focus far-away objects. The light from a nearby point does not converge on a single point on the retina. That nearby point is out of focus when you are looking at an object that is far away.

The eye muscles reshape the lens to make that nearby object come into focus. That's diagram C in the above image. Now light from a nearby point converges on a single point on the retina. Light from a different nearby point will converge on a different point on the retina. Now the distant objects that you were looking at become out of focus.

 

[nitsuj]

Here is my basic question: how does knowing the place on the retina that the light beam struck allow you to calculate the angular position of the object?

lol the constancy of your experience. there is no "calculating".

 

[SongDog]

In simple terms, for a perfect lens, the azimuth and elevation on the retina of the points which are illuminated with images of your two cubes are simply a scalar multiple of the angles on which they were incident upon the centre of the lens. The scalar multiple is a function of the lense shape and of its refractive index, but that isn't really relevant to your question.

For real (imperfect) eyes, the variations fill textbooks. There's a reason optometry and opthamology are separate specialties...