“Champions know that success is inevitable; that there is no such thing as failure, only feedback. They know that the best way to forecast the future is to create it.” – Michael J. Gelb
The way in which our mind processes visual information is a miniature version of the way in which it processes all kinds of information, with multiple layers of processing and continuous feedback loops being used to construct an internal mental reality which most effectively helps us to maximize rewards from the external world. That is, consciousness is merely a system for using feedback to help us create the future.
In I am a Strange Loop, Douglas Hofstadter discusses the reflexivity of our minds using many powerful analogies, including those of video and audio feedback, and describing the brain as a library of dormant symbols, which only come alive when triggered by a relevant external event. He describes consciousness as “the dance of symbols in the brain” (a phrase I love), making it clear that the dance is composed of symbols and not neurons. That doesn’t mean that neurons are not ultimately responsible for the dance: rather, one of his central arguments is that confusions about the physical brain and it’s relationship to consciousness are merely confusions about different levels of description. Systems, and the brain is an extremely complex system, can always be described at many different levels.
For instance, if I start my car, I might say, “I just turn the key, and the engine starts”. Alternatively I can say, “the turning of the key sends an electrical impulse to the spark plug, which fires ….. “, and, moving several descriptions down, I could describe the movements of billions of atoms in the car cylinders. However, for most of us, most of the time, “I turn the key and the engine starts” is a good enough description, which contains within itself, a multitude of nested and referenced descriptions and definitions which are needed to understand it’s meaning. Our perception of consciousness is just the top-level description of what’s going on in our brain.
In Mind Hacks, Tom Stafford and Matt Webb, provide interesting and useful “middle level” descriptions of many of the systems that our brain uses to navigate the world, and I will focus here on some of those used in visual perception. Visual perception, like all brain functions, is incredibly proactive (and interactive) with the outside world, and although providing us with a constant and seamless perceptual experience, is actually an internally created model based on (quite often) very incomplete inputs from the outside world – much of what we think we see is in reality extrapolated (or interpolated) by our brain, based on small chunks of information filtered through many systems as our eyes jump from place to place seeking threats and opportunities in our environment.
Importantly, our eyes are set in a moving body (which helps us to estimate depth and movement of other objects). Only the central fovea of our retina has high-resolution vision (where the highest density of receptors are), helping us to accurately see only what is directly ahead (an area the size of the full moon when we look at then night sky). Everything else on our peripheral vision is really a blur, with an important (but completely imperceptible) blind spot just outside this central area.
The reason we don’t perceive the blurred or missing areas is that our eyes are constantly moving (around 5 times per second), often in the direction of something in our peripheral vision which has excited the brain’s attention. These movements, called saccades, are not perceived by us as the eye does not move smoothly but rather jumps between different fixations, with the brain ignoring the movements and creating the illusion of smooth and unbroken perception between these fixations (something you can see if you watch someone else’s eyes, or use eye tracking equipment). [You can see demos of eye movement, and map your blind spot, at the referenced links.] These movements are responsible for many classic visual illusions (see the rotating snakes illusion below).
The brain makes a lot of assumptions in interpreting data from our eyes, and shadows play an important part in interpreting position, shape and movement too. When we see shadows, our brain assumes that light comes from above and that objects move and light sources don’t (pretty sensible assumptions for most of our history on the planet), which occasionally leads us astray, but then a fast moving shadow could be a fast moving object so it makes sense to take avoiding action! In addition to shadows, our brains use binocular cues, perspective, intensity and motion to judge distance and movement.
To give one very specific example, the startle reaction (blinking, flinching, twitching) is an automatic one buried in one of the more primitive parts of the brain. We respond to something that appears to be on a collision course within 80ms, which is far too quick for any sophisticated processing, and even too quick for our brain to combine information from both eyes. It is triggered by symmetrical expansion of dark areas – that is, if we see something darker than background, which is increasing in size but not changing shape as this happens, then it is very probably moving directly towards us (rather than obliquely) and we need to take action. [Or think of the startle response to a long thin shape in the grass, which happens long before our brain works out that it is a piece of wood (to use an example I have referenced before).]
Our visual perception, like our minds, is built on multiple layers of feedback systems. helping us to navigate the world safely and productively. To truly understand any aspect of human behaviour, we need to consider all these different feedback mechanisms and broaden our focus away from the highest levels of deliberate thinking to encompass other levels of mental processing. All behaviour consists of multiple interconnected feedback loops.
I am a Strange Loop by Douglas Hofstadter (2008)
Mind Hacks by Tom Stafford and Matt Webb (2004)
http://ilab.usc.edu/bu (for eye tracking and visual attention demos)
http://serendip.brynmawr.edu/bb/blindspot/ (to map your blindspot – read the background material first)