Saturday, November 6, 2010

are my eyes playing tricks on me?

Seeing as I've been studying neuro ALL WEEKEND for an exam on Monday, the very thought of which ABSOLUTELY gives me a bad feeling in the stomach, and I am frankly sick of it right now, I may as  well make a pseudo-helpful review post.

The topic being VISION! and optical illusions!  Well, here's just one optical illusion as an example.

Chevreul's Illusion:

Chevreul's illusion consists of vertical stripes of varying luminance.  The luminance is actually constant within each stripe, however - so why does it seem like the borders of each stripe is slightly different in luminance from its interior?

Let's see if it is possible to explain this using as few technical terms as possible.

So starting at the (relative) beginning, there are several different layers of cells in our retinas.  Because our eyes are just that cool.

One type of cell is the photoreceptor (you know - rods and cones, for seeing in black and white, and color).  The photoreceptors convert light energy into changes into membrane potential.  Changes in membrane potential mean voltages change across neuronal cell membranes and neurons fire, and...after all this communication business, signals get to our brains and stuff happens.  In this case, we see.

Well, photoreceptors communicate with the other layers of cells in the retina.  The cells at the end of the line, so to speak, are the ganglion cells.  Ganglion cells are the only cells that bring information from the retina to the brain (via the optic nerve).

Each ganglion cell has a receptive field - receptive field is an area of the retina that, when stimulated with light, changes the cell's membrane potential.  Most ganglion cells have a concentric center-surround receptive field organization.  Meaning, there is this bulls-eye type thing that makes the cell respond differently based on if the light goes in the center, or in the surrounding area.  Diagrammed something like this (image taken from wikipedia):

Furthermore, in retinal ganglion cells, the center-surround organization of the receptive fields leads to a neural response that emphasizes the contrast of light-dark edges:

(picture taken from Neuroscience: Exploring the Brain, 3rd edition by Bear, Connors, Paradiso); preview from google books here.

So looking back to Chevreul's illusion, you can see how a ganglion cell could get confused - and respond more if it's receptive field is at the border of two stripes rather in the middle of one stripe.

(Okay, sorry for all the buildup and then a really lame partial conclusion).

There's a good explanation with a diagram and everything for a different optical illusion, the Hermann grid, here.  I was just too lazy to draw out a diagram for the Chevreul illusion with ganglion cells superimposed all over different parts of it.

And so ends my ridiculously confusing account of one type of optical illusion.  I think never again should I attempt something like this.  Okay, back to real studying...


  1. Hey, keep going! That helped me a bit to prepare for a seminar on Chevreul illusion :)

  2. Glad it helped! Maybe I'll try to do these types of posts more often :)