Drawing conclusions about illusions and the failure to perceive them

Over the past few months, based partly in an ongoing interest in the brain science behind eyewitness errors, partly on being immensely impressed with the related artwork by Salvador Dali I saw in Berlin and Spain last year (e.g., at left), and partly thanks to the need to generate ever-new drawing projects for my 4-1/2 year old granddaughter, I've been teaching myself to draw rudimentary optical illusions (sometimes while sitting in lengthy committee hearings at the Legislature waiting for a bill to come up). In furtherance of that effort I purchased a slim book by a fellow named Robert Ausborne titled  "How to understand, enjoy and draw optical illusions," which was a good starting point for a ham-handed beginner.

One thing I've learned through this process is that, for whatever reason, I cannot see certain types (but not all types) of color-based illusions, in particular so-called "afterimage" illusions, which is when you see the negative or complementary color, i.e., the color across from it on the color wheel, after staring at an initial image then looking at a neutral color. (See examples here and here.) I understand them. I can even create afterimage illusions others can see (it's easy on Photoshop following Ausborne's instructions: With the image in an active layer, perform the following commands: Image > Adjustments> Invert). But as I wrote to the author, "I have discovered that, no matter how hard I try, I cannot see after image illusions … EVER … having now looked at dozens of examples, and comparing notes with my wife (who does see them). I’m not color-blind and my eyes react to illusions based on shading tricks, but I don’t ever get an afterimage." He graciously replied:

The question you ask is a good one.   While I am not a scientist, but as a long suffering, pestering enthusiast I can offer up anecdotal evidence.

The most obvious demographic quirk I've noticed about illusions is age.  The young quickly realize almost all illusions with no trouble at all.  They seem to walk into them guileless.  The older a witness is the harder it becomes.  You don't see that many optical illusion shows in Florida.

As for After-image illusions.  My own research has turned up several factors which affect the ability to "see" them; such as heredity, eye color, color blindness, and a host of other vision problems, including bad lighting.   I have never found a definitive answer and suspect that nobody really knows.  Perhaps you were born with an extra supply of photo chemicals, and just don't run dry that easily.  Perhaps your eyes have an Indianapolis 500 Pit Crew and the photoreceptors are replenishing too quickly.  Perhaps your eyes are making tiny movements which are undetectable, thus making it difficult to expose a single patch of neurons to the illusion.  Perhaps you are an alien plant, and have just inadvertently given humans a sure fire test to discover and weed out your kind.

As an artist who draws illusions for a living I can tell you that it is possible to oversaturate oneself with an illusion.  Most illusions do not stop working while I draw them, and with some illusions it can be like trying to paint a leaf while it's falling; as things tend to wander out from under my eyeballs.   But with some illusions I can become burnt out; I just can’t see them easily anymore.  I run the risk of ruining the illusion, making it too obvious by far in order to stimulate my own burnt out senses.  The reason for this phenomenon may be that we are human; we learn.  The brain adapts.  Just as the brain can make your nose invisible to both eyes, it can make an illusion invisible, once it figures out you don't really need to "see" it.   Perhaps you are just more adaptable than the rest of us.

I appreciate the author's response, and he's probably right nobody knows. I'm still curious about it, though.

I've been interested in how we see color ever since learning with fascination and borderline envy about synesthesia and how differently synesthetes see and interact with colors (which are often associated with numbers, music, or other mathemtaically based facets of life). Then, in researching eyewitness identification in a criminal-justice context, I became acutely aware of just how little of the world around us our eyes actually see and how much is filled in by our memory. That's why eyewitness accounts are extremely reliable when people previously knew the person they're identifying, but exceptionally unreliable when trying to identify someone they'd never seen before. Writing in the June 30, 2008 New Yorker on an unrelated topic (itching, to be precise), Dr. Atul Gawande described the nuts and bolts of vision mechanics that explain why that's the case:

The images in our mind are extraordinarily rich. We can tell if something is liquid or solid, heavy or light, dead or alive. But the information we work from is poor - a distorted, two-dimensional transmission with entire spots missing. So the mind fills in most of the picture. You can get a sense of this from brain anatomy studies. If visual sensations were primarily received rather than constructed by the brain, you'd expect that most of the fibres going to the brain's primary visual cortex would come from the retina. Instead, scientists have found that only twenty percent do; eighty percent come downward from regions of the brain governing functions like memory. Richard Gregory, a prominent British neuropsychologist, estimates that visual perception is more than ninety percent memory and less than ten percent sensory nerve signals.

So memory is filling in most of the image you see when you "see" something, which is why eyewitness errors are much more likely when identifying strangers. As for the optical illusion: No wonder most people see afterimages if our memory is generating 80-90% of our visual perception. Your brain is busy filling in all the gaps in the image you're looking at, and when it's taken away it can't immediately shut down that extraordinarily complicated function. Who knows why I'm an exception, but there always is one. I'd still like to know.