Autopilots and Cartesian theatres
On first inspection, it seems to us that we are surrounded by the “external world” which we apprehend through our physical senses. What we see, hear etc. is exactly what is “out there” and we have freedom to respond to those sensations as we wish. So if we wish to switch the light on then we can cause our hand to reach up to the switch.
Looking more closely, it becomes apparent that things are not as simple as that. We are aware that we have a large number of unconscious automatic control systems over which we have no direct control. From the regulation of our heartbeat to the ability to walk without being conscious of every muscle movement. We also know that some of these systems can be modified by training. For instance, we can choose to learn to ride a bicycle or to walk on a tightrope. We can also, with practice and other psychological methods, become less angry or less aggressive when presented with particular circumstances. Although the behaviour of the autopilot at any given time is not directly under our control, the settings of the autopilot and the ways it will behave in the future are. We can “build our character” in a way which we wish to and we identify our “self” as the one who chooses what sort of character or skill we want and then to go for it. Or not.
What we perceive through our senses is not a direct image of the external world but rather a model of that world which is deduced indirectly from our sensations by means of a large amount of computation performed by the autonomous systems which we have been provided with. What we perceive depends, not only on what is “out there” but on our beliefs and desires. This is most clearly revealed in the case of illusions when what we see is not what is there, or not only what is there.
Illusions caused by the autonomous sense data processing
1. The brain can be easily persuaded to “see” something with a clarity which is not present in the received information. For instance the duck/rabbit illusion shown in the Illustration. The information provided in the picture is not sufficient to identify the creature and the brain has to make a best guess as to what the object is. Because the brain has had many examples of ducks and rabbits, and these would seem to provide a good fit to the data, the brain presents the output of this guess to consciousness. However, there is still a problem in that a duck and a rabbit both fit the received data to a similar degree so the brain has difficulty in coming to a conclusion. The best guess will abruptly change from one to the other on the basis of very small changes of viewpoint or of belief. We do not see an indistinct picture or just meaningless lines on the page, we see one or the other, at any one time, without ambiguity.
2. Another illustration of the fact that what we see is the result of much signal processing is the “hidden Dalmatian picture” shown in this illustration. To begin with it just looks like a set of random blobs. However, after a while, or if it is pointed out, a Dalmatian can be seen. Once it has been seen then it instantly emerges every time the picture is looked at. Again, this is the result of the brain coming up with a best guess from sparse data using unconscious data processing.
3. Similar illusions operate in the realm of action and responsibility when a subject believes that they are responsible for a certain outcome when they are not or vice versa [20, Chapter 19]. This implies that the brain is making “best guesses” which it presents to consciousness not only about outside data but also about our own actions. We have only indirect and incomplete knowledge about our own doings.
4. This processing takes time and we see the world, not how it is, but how it was several hundred milliseconds earlier. Evidence for this is provided by Libet's experiments [30] and also from the illusion called “colour-
In this case the subject is shown two lights. First the blue light on the left is momentarily illuminated. This is followed by a pause and then the red light on the right is switched on momentarily. What is perceived is shown on the right. The brain “fills in” the missing data as being a best guess, even though it is wrong. However, what makes this interesting is that the intermediate values are perceived before the red light on the right is switched on and, therefore, before the observer knows that the red light is going to be switched on or what colour it is. This effect can be explained by there being a delay between the lights being switched on and their being perceived. Despite how it seems, the best guess is only made when all the data is available. To determine whether this is a true explanation requires more data [20, p156] but it seems plausible.
Taking into account these observations, a good analogy would be that of an airliner pilot who is aided by many automatic systems including an autopilot. Another would be that of a space probe which had landed on Mars and was being controlled from the Earth. The operator is presented with the data from a variety of systems and could send control signals to which the probe would respond. Due to the time lag, it would be be necessary for the probe to have some autonomy so that it could, for instance, take evasive action if it was about to hit an obstacle. Moreover, in order to reduce the information which would be needed to be transmitted back to Earth, a certain amount of processing of the directly perceived data could take place on the probe and the Earthbound controller would only see the results of the processing, not the raw data itself. Nevertheless, with a suitable interface, the operator could be presented with the illusion that he was on Mars looking out of the windows of the space probe and having direct control of its locomotive and other functions.
Continuing the analogy, the actions of the space probe on Mars are not under the direct control of the operator, there is no time for that, but the responses will have been programmed or conditioned by the operator at some time in the past. Moreover the programming can be changed if it is seen that the rules which the probe is working to are not as effective as they could be.
A boundary between the “self” and the “outside world”
In case of a space probe, it is relatively easy to draw a boundary between the probe which is being controlled and the person doing the controlling. However, if this analogy is used to describe what is happening within a person then things become much more hazy.
For instance, it is generally believed that the locus of consciousness is in, or at least associated with, the brain. Therefore, arms and legs, together with the sensors and muscles which they contain, could be seen as being part of the external systems which are being controlled by the brain rather than part of the essential individual.
That this is widely intuited is demonstrated by the observation that most people would not imagine that replacing a leg with a prosthetic one would change their character or in some other way make them a different person.
Similarly, if someone is given an organ transplant, whether it be a heart or a limb, few would consider them to be a different person from who they were before the transplant took place.
The limb or the heart is considered as part of the equipment with which we are endowed which allows us to survive and operate within the Earth's environment in a directly analogous way to the space probe giving such capabilities to the operator.
However, suppose we try to move the boundary further towards the brain and maybe into the brain itself. As we do so, we initially find a large number of unconscious and autonomous systems which are able to respond to sensory inputs in a way which, hopefully, aids survival and efficiency.
Are these systems part of the person or are they part of the associated machinery? Would replacing a part of the brain with an artificial substitute or a transplant change who the person is? If so, which parts of the brain would cause this effect? A retina? An optic nerve? The V1 visual centre?
Is the person responsible for what these unconscious systems do any more than they are responsible for what the heart does? Or what their cerebellum does?
If our freedom to choose is limited, how does this reflect on culpability and personal responsibility? Where does our autopilot stop and the real “me” start? If we keep stripping off the non-
Many people speak of the person as being more than a monolithic entity. For instance, Plato, in one of his dialogues says this [1]:
St. Paul also speaks of the “higher” and “lower” nature or the “spirit” and the “flesh”. eg. Romans 8.
Another reference to our having two natures is given in the story of the two wolves, attributed to the Cherokee[2].
A similar perspective is described from the psychological point of view in a book which investigates “self-
[1] Plato, “The Republic”
[2] http://www.firstpeople.us/FP-
[3] Walter Mischal, “The Marshmellow Test”