They used to have lots of competitions with tasks like: describe, in 25 words or less, what makes Aunty Ethel's Home-Baked Cookies taste so great. (These days those types of competitions are considered too difficult, and competitions have tasks like: What is the missing letter in the following: "Aunty Ethel's Home-Baked _ookies".)
Following in this great tradition, I set myself the following task: describe what consciousness is, in 25 words or less.
And the answer I came up with, in just twenty words, is:
Consciousness is a system for deciding whether or not to do the thing that you were going to do next.There are some implications of this description that merit further exploration:
The second conclusion is sufficiently radical that I will deal with it first. Conventional wisdom holds that the brain is very distributed in its operation, and there isn't any particular part of the brain "in charge". The notion of "pontifical neuron" is identified as a notion too ridiculous to be true, held up as an obvious straw man to be discarded in favour of some more obscurely complex of theory of who or what is "in charge" of what the brain does.
Let us for the moment suspend our sense of ridicule, and allow the possibility that there might be an identifiable pontifical neuron somewhere in brain which is "in charge". It does not follow that this one pontifical neuron is doing all the work. Rather, the activity of this one special neuron represents the very endpoint of a complex decision process, and there could easily be hundreds of millions of neurons involved in creating the input information which the pontifical neuron uses to calculate its decision about whatever it is that it is deciding.
One objection to the pontifical neuron theory is that the pontifical neuron is too fragile: if it dies, then consciousness in the afflicted individual would suddenly cease. But this objection can be overcome if we suppose that evolution has managed to invent a special method for implementing the functionality equivalent to a single neuron using a group of neurons. And this method is the following: use a special neurotransmitter for the output of these neurons, and allow that neurotransmitter to be transmitted in a leaky fashion, so that the transmission of the information in that neurotransmitter goes from one group of neurons to another group of neurons without going through any single point of failure, yet at the same time being at least approximately a single unified channel of information. For lack of a better term, I call such a group of neurons a super-neuron.
There appears to be more than one such super-neuron in the human (and mammalian) brain, however there is one that corresponds most likely to the pontifical neuron as described above, and that is the substantia nigra, which is a group of neurons whose output neurotransmitter is dopamine, and whose outputs project onto the striatum. This group of neurons is known to fire a signal whenever consciously initiated motor actions are performed. (This does raise a question about non-motor conscious actions, in particular redirection of attention, which I will discuss further below.)
There is a very attractive metaphor for conscious "yes"/"no" decision-making from the political arena, which exists whenever a yes/no decision is made on the outcome of a vote. For example, in the final moments of deciding whether or not to enact a new law, that new law is either passed or not passed by the legislative assembly charged with the creation of laws in that country. There may be very complex information processing stages involved in the formulation of a new law and deciding the pro's and con's, and perhaps suggesting and voting on amendments, but at the very end of the process, it all comes down to a simple yes or no. Somewhere in the legislative building there is a guy counting bits of paper (or something equivalent), and the result of that count determines if the law is going to pass. On the one hand it is obvious that the vote-counting constitutes only a tiny portion of all the information processing involved in passing a new law, but on the other hand it is a very crucial part of the process, and any failure or corruption of the vote-counting role would undermine the integrity of the whole system.
If the primary role of conscious is yes/no decision-making then it follows that something other than consciousness must formulate the proposed actions before consciousness decides yeah or nay. In one sense this formulation of actions is part of the process of conscious decision-making, but it does appear to lie outside the domain of the subjective component of consciousness, as suggested by experimental results of Libet et al.
In fact, even before an action is proposed, it is necessary to recognise in the first place that circumstances require the formulation of a proposed action, and that this action will require the assent of consciousness (with all that that involves in terms of cost and delay) before it goes ahead. The recognition of this requirement appears to be performed via another super-neuron, the locus ceruleus, which projects axons that transmit norepinephrine onto most of the brain. Suggesting indeed that the task of formulating proposed actions is carried out not by any specific brain area, but by any brain area whose task and expertise is relevant to the problem at hand. The norepinephrine tells those brain areas two things:
A good metaphor for this method of answering questions is a teacher trying to get an answer to a question out of a classroom of pupils. The teacher first asks the question, and asks if any pupil knows for sure what the answer is. It may be that some pupils can guess the answer, but are not sure that they are right. The teacher could ask that the pupils supply an answer even if they are not confident that they are right, perhaps starting with those most confident, and moving onto wilder guesses if nothing suitable comes up. When a possible answer is supplied, the teacher then asks the class to raise any possible objections (a pupil may not know the right answer, but they might recognise a wrong answer for what it is). Finally the class votes to reject or accept each suggested answer, until one is accepted, which will then be presented to the teacher as the class's answer to the question.
This is an extremely heuristic technique for answering questions, but then life is a very heuristic business. It is not always necessary to come up with the best answer to a question, but there are many circumstances where it is necessary to come up with some answer. (For example: "There is something strange in my bedroom, what do I do next?")
I have mentioned that the substantia nigra appears to correspond to the pontifical neuron, even though it appears to be active specifically during motor responses, and what we recognise as consciousness appears to do more that just control motor actions, in particular when it is engaged in what we call thinking.
There is also the phenomenon of attention, which is something that we are subjectively aware of being able to consciously control, and which can be directed without necessarily any involvement of motor systems (although often they are involved anyway, like when we turn our head towards something that we are paying attention to).
Attempting to combine conscious control of attention and conscious control of motor behaviour into a single theory of consciousness raises two questions:
The first question is probably best answered by laboratory research on real brains, so I will leave it for others to answer.
As for the second, attention is involved in both the input to and the output from consciousness, and in some sense "closes the loop", allowing consciousness to "control itself". That the direction of attention is consciously controlled is a fact that we know subjectively. A famous joke tells that there is a school of thought among philosophers that dogs are not conscious, and that the opposing school consists of those philosophers that own dogs. We might ask the second group of philosophers if they also believe that dogs can consciously control the direction of their attention. If the answer is negative, then the conscious control of attention may be uniquely human, or perhaps restricted to ourselves and some of our close relatives.
The control of attention by consciousness relates attention to the output of consciousness. How does it relate to the input? The answer is that the very purpose of attention is to influence which set of inputs is most likely to provide the stimulus to which a proposed response needs to be formulated and consciously decided upon. Note that I use the word "influence" – attention does not absolutely determine which stimulus gets a conscious response, because a sufficiently strong enough stimulus outside the set being attended to will win the battle anyway. Nevertheless, to the extent that consciously controlled attention determines the input to consciousness, the loop is closed, and an ongoing stream of conscious thought is possible.
There is another somewhat different interpretation of non-motor conscious activity, and this is the idea that much conscious thought involves hypothetical decision making, i.e. thinking about what one might do in a situation that will or might occur in the future. Of course we can tell ourselves that we have decided to do something, and then find that when the moment comes we don't actually do it. But giving consideration to a future or hypothetical situation at leisure is likely to influence our future behaviour, because we have time to discover more considerations relevant to the decision when the time comes finally to actually make it.
A full description of non-motor conscious activity is probably going to involve both the role of attention and the concept of decision-making in advance.
With power comes responsibility, at least it does if a system is intended to fulfill some purpose, and with responsibility comes accountability.
I have explained what consciousness does, but so far I have said nothing about how it does it, and who or what if anything makes sure that it is performing its task correctly, or at least as effectively as possible.
Subjectively, so-called qualia and feelings are involved in conscious decision making. To some extent we can explain feelings as being derived from past experience, i.e. generalising about results of past decisions. However there is also that elusive phenomenon called free will. Free will appears to be a means of doing what is not necessarily expected to produce the greatest immediate satisfaction.
We should be suspicious of the idea that free will (whatever it is) is completely free, as that would seem to imply a complete lack of accountability. A suggestive metaphor is that of a dog on a long leash, still under some sort of control, but given enough freedom to get the job done (it must be a working dog then).
The leash that applies to free will is sufficiently long that it is virtually impossible to observe under laboratory conditions, at least not without holding some poor subject under imprisonment for long periods, which would be somewhat unethical. Nevertheless, we can observe ways that free will almost inevitably fails if it persists unrewarded for long periods, such as the dieter whose diet always fails, or the drug addict whose addiction always overcomes them (although drug addiction is a somewhat unnatural example).
The theories of behaviourism (now somewhat discredited) claimed that rewards and punishments directly controlled learning processes, and the usual objection to this theory was that it ignored the role of free will. However, the supporters of the concept of free will often go to an opposite and indefensible extreme, suggesting that there can be no mechanistic link at all between the actions of free will and rewards and punishments that follow from those actions. Our subjective feeling about this issue is generally that the promise of reward or punishments (i.e. good and bad feelings) in some way "influences" free will, but does not "determine" it. This description is unfortunately not very satisfactory, and it seems to lie outside normal concepts of causation that apply to physical phenomena in general.
The solution that I suggest is that free will represents the ability of somewhat arbitrary strategies for conscious decision making to form and persist, but conditionally on rewards exceeding punishments over the long term. The metaphor that strikes here is that of a business in a competitive economy: you can run a business any way you like, as long as profits exceed losses over the long term. In the business world competitive forces also have a tendency to erase the effects of specific causes of profit or loss, i.e. if business is good, then your competition will copy you, and eventually eliminate your competitive advantage, and if business is bad for everyone in a given type of business, then, as some producers go out of business, reduced supply will increase output prices and decrease input prices, allowing others to remain viable.
There may be similar forces modulating the effect that rewards and punishments have on free will, in particular it may be the difference between expected "raw" punishments and rewards and actual raw punishments and rewards that determines the profit/loss accountability applied to free will. This helps to explain the very emotional effects of relief or disappointment, which can be regarded as a reaction to the non-occurrence of expected punishment or reward respectively.
So there it is, consciousness explained, in 25 words of less, with a few more words to clarify. I have proposed the existence of something approximately equivalent to the much vilified pontifical neuron, and I have included in my discussion conscious decision-making, attention, conscious thought and free will.