Chapter 3

How do you see something? You might not have given it much thought before, not even when taking an eye exam. Why would you—or anyone for that matter— question such a highly self-evident experience? A sense with such perspicuity seduces our intuitions into convictions about its accuracy; we think what we see is what is actually there, as we see it. But is it? Crick starts off the chapter by revealing how oblivious and disinterested the lay person is about how vision works. The question, "How does my brain work when I see something?", although simple, reveals itself to have a gluttonous appetite for a complex diet of information from psychology, physiology, and molecular and cell biology. Despite all this information, Crick emphasizes a paradoxical point; we still have no idea how we see anything. 

With such a surfeit of information failing to provide an adequate answer, one begins to get the impression that– whatever answer lies behind the question, "How do I see an apple?"– that answer contradicts the principles of Occam's razor. To demonstrate how interesting vision is, and the difficulty in fully understanding how it works, Crick initially presents three surprising aspects on the present (at the time) knowledge of the visual system; 

i. knowledge on it was enormous: knowledge accumulated over years from experiments, and theoretical work—involving both animals and humans—was significant.  As previously mentioned, disparate fields had made significant contributions and progress in understanding how vision works. In the psychology of vision, for example, different conditions under which the rapid succession of still pictures produced smooth movement had been probed and elucidated. Further, in the physiology of vision; the structure, behavior of the eye, and relevant parts of the brain had been parsed and understood. Lastly, in the molecular and cell biology of vision, the nerve cells and their constituent molecules had also been described. Such knowledge, by scientific standards, was extensive.

ii. Still, in spite of all that progress, no one–scientists and psychologists included–had a clear idea on how we see anything.  The simplest question: "How do I see color", still lacked an answer. What are the processes involved when one recalls the image of a familiar face, a familiar movie? Such questions, according to Crick, could not even be outlined with the current (at the time) knowledge on vision.

iii. How you think you see things is largely simplistic, or in many cases, plain wrong. Many of us have had our intuitions on vision contaminated by watching television. It is tempting to think that our vision works similarly to television i.e. our eyes are the camera and they project what is happening on to the retina, which acts like a screen. We, the viewers, watch this screen and are able to tell what is going on in the world. The viewer, in most cases, is a little man or woman somewhere inside our brain who is following what is going on, just like someone sitting in a cinema watching a movie. Crick calls this the Fallacy of the Homunculus (homunculus is Latin for "little man"). 

Interestingly, Crick admits that the Fallacy of the Homunculus requires an explanation, since seemingly everyone feels this way about how they see the world. However, before expounding on its mysterious existence and origins, he provides his Astonishing Hypothesis as a counter argument against it; "it's all done by neurons". With the Astonishing Hypothesis in mind, the problem of seeing —according to Crick—starts looking somewhat different. He states that the task then becomes more about finding brain operations or brain structures that work in an analogous manner to the mental picture of the homunculus. But before setting out to uncovering what they are, he first describes the brain and what the act of seeing entails. 

To the lay person, an even more ignored or absurd question compared to "How do you see anything?", is the question of "Why do you need to see anything?". One can only imagine the amusing responses to such questions, ranging from a simple "I just open my eyes''  to the more insightful ''To know what I am looking at". However, the answer to the need for vision is more nuanced than it appears. A global and biological reason is to increase survival and leave descendants for the continuation of the species. However, Crick adds that the primary function of vision has to do with addressing proximal goals. You need to know what is in front of you so you can decide, in real time, what to do about it. This holds great significance for both predators and prey in the animal kingdom, particularly in the context of the wild. He mentions that according to John Allman, a neurobiologist at Caltech, mammals—unlike reptiles, have a more pronounced need to preserve heat due to their constant activity and high body temperatures. This becomes even more important for smaller mammals, who have a much larger surface area relative to their volume i.e. they lose heat easily. Partly due to this, mammals had to be smarter in order to be able to quickly locate food. This required not only development of a larger neocortex, but also an increase in visual acuity. However, he states that most mammals do not have especially good vision, except the primates (monkeys, apes and humans). 

In order to derive meaning from the light entering your eyes as you gaze at an object before you, your eyes must possess the ability to perceive the wavelength, intensity, and motion of that light. With these parameters, the goal is to find out not only what is out there, but also what it is doing, and/or what it is likely to do. This requires experience and prior knowledge, most of the time. Three general remarks that Crick makes to set the stage so that we can understand just how vision works:

1. You are easily deceived by your visual system – your visual acuity decreases significantly away from the center of gaze. You see clearly what you are paying attention to. If you hold your hand up and move it closer to your eyes, at some point, the increasing resolution will enable you to see your fingerprints. However, as you move your hand away, such resolution is lost and the prints are not detectable at a considerable distance. Despite such loss in detail, you still feel like you see the world as it is, in good clarity. There is no part of your brain that sounds the alarm that some vital information has been lost. To demonstrate this in a more direct way, google a picture of a Kanizsa triangle and look at it. Even though there is no triangle, your vision tells you that there is, by making the background outside the "triangle" appear darker, even though they are the same shade. Your brain extracts the triangle out of the three pac-man faces, and deceives you into seeing a triangle that is not there. The triangle is not real. 

2. The visual information provided by your eyes can be ambiguous - there are many different interpretations of the objects that you see in the real world before you. However, it is rare for people to notice this. An obvious example is the Mona Lisa. From one angle she is smiling, but from another she is sad. In the age of the internet, it is easy to see such images, where a face appear to be facing this way in one instant and the opposite way in another instant. Look at the Necker Cube in Figure 2. The Necker Cube, without any colored faces, is a series of twelve lines that are the same length. For some reason, these lines appear as a 3D cube. Interestingly, it is a bistable 3D cube, and its orientation inverts when you look at it long enough (see blue colored faces). The brain is uncertain which it prefers, and only detects one at a time, not an odd mixture of the two. To avoid ambiguity in a lot of objects you see in real life, your brain combines information (shape, color, movement, etc) and settles on the most plausible interpretation. Therefore, a lot of assumptions have to be taken into account before you see anything (believing is seeing?). 

3. Seeing is a constructive process - unlike a video camera, your brain does not passively record the information coming in; it actively seeks to interprete it. As the above examples have shown, your brain is highly involved in intepreting what you see. It settles on the interpretation that it sees as the most veridical solution - the one corresponding to "what is out there" (as supported by other senses, such as smell, sound, touch etc). To see the constructive process in real time, close one eye, hold a pen in front of you and—while staring ahead—move the tip of the pen away from your nose. At some angle, it will disappear and you will not be able to detect it. This is called the blindspot, and it is always there. However, there is no blank space in our blindspots, our brains "fill in" the information for us and we see something instead of nothing. Crick expounds on this process in Chapter 4. 

Taking these three remarks into account, along with the information they have provided, Crick stresses an important point about vision; What you see is not what is really there; it is what your brain believes is there. So maybe, a little skepticism (although not practical in everyday life) about what is in front of you is something that needs to be considered. How then do we actually see?

Just as a movie in a computer is not stored inside a computer as the movie one watches upon pressing "play", what we see with our eyes cannot be stored inside neurons as small pieces of visual scenes. Computers use symbols i.e. a combination of 0's and 1's to store and represent information. Using this analogy, Crick suggests that what we see as images and motion pictures is also represented and stored as symbols in our brains. As symbols, the brain can manipulate and store the information about objects at many different levels, and bring about the symbolic representation of that object when that particular object has been encountered. This is not a straightforward manner, and requires processing at multiple levels of the brain and corroboration from other senses, as we have seen. As Crick puts it, when something has been symbolized explicitly, the information can easily be made available so that it can be used, either for further processing or for action. So, how would vision work? By having neurons firing in a way that symbolizes such information fairly directly. Just as computers have a particular arrangement of 0s and 1s for each specific object, specific neurons firing in a specific way can be the way that our brains represent, and those objects. Such a process requires explicit multi-level, symbolic interpretations of the visual scene in order to "see" it. 

In summary, firing of a neuron (or neurons) is highly correlated with some particular aspect of the visual world. 

Blogger’s thoughts: So far this has been the most interesting chapter for me, because I have personally grappled with this topic for the past two or three years. Not only does Crick make the case that vision is elusive and deceptive, he states towards the end that there is no way for us to confirm that what we see is indeed veridical. Taking this to its logical conclusion, as Donald Hoffman has, you do not see reality. Had I not read "The Case Against Reality" by Donald Hoffman, I would have found this chapter to be mind boggling. In the book, Donald Hoffman goes a step beyond what Crick is saying here. According to Donald Hoffman, whatever reality is, it is not what you see. What you see is just an adaptive fiction. Worse than that, you do not even have the ability to see reality, even with all the current tools and technology. I still have not encountered a more radical view in all of science, and that is saying a lot. I highly recommend his book. Maybe I should review it next. Let me know. 

It is interesting to know that Crick and Hoffman had a discussion about the relationship between vision, veridicality, and the existence of reality, with Crick standing his ground that even though we cannot see reality, it exists and we have ways of knowing that it exists. This whole chapter reminded me of the argument of whether or not a falling tree makes a sound when no one is there to hear it. When you think of what sound is —vibrations in the air that propagate to your eardrums, then one gets a little uncomfortable talking about sound when there are no eardrums present to detect it, and brains to interpret it. One can go a step further and imagine being in a comma for two years and having the longest dream while in that comma. How would you know that you are not in actual reality? What is reality exactly, when you can have experiences in your brain and your body treats them as real? I guess, even in Science, one cannot totally avoid referencing The Matrix. 

With regards to how we see, I do not see how Crick's methodology is any different from that of other neuroscientists at the time. They had traced the sequence of events as light entered the eye to the retina and to the brain. I am guessing Crick will focus more on the particular firing sequence of neurons and the order in which they do so. But even so, fMRI had already been invented four years prior to Crick releasing the book. fMRI allows us to visualize the activity of neurons, in real time, as subjects perform certain tasks e.g. looking at images. With that, the activity of the neurons at certain brain regions can be identified and associated with certain states of consciousness. Let us not rush, maybe things will be clear as we delve into Chapter 4. Read on.....