This is really an interesting discussion and I would like to once more contribute some further thoughts.
Within the picture of reality that is assumed to be exclusively only information processing, not more and not less, awareness of a world outside and independent of consciousness, is seen as a highly sophisticated, self-referentially functioning model of the world. Within that model, the observing subject, the person, the “I” is thought to be also “merely” a part of that highly sophisticated modelling procedure.
Consciousness and its Qualia in this picture are seen as supervening on complex information processings outside consciousness (thus, in the brain). In other words, consciousness is seen as a virtual “thing”, it has no physical substance, it is immaterial (like the landscape in a computer game is immaterial, admittedly a bad example, since it presupposes the existence of consciousness – rather than explaining it). The contents of consciousness are seen to be governed exclusively by deterministic information processing, namely the one we call “computationally irreducible”: one has to run the computation to obtain the result, there is no other way to obtain computationally irreducible results, they cannot be predicted in principle.
Some people may go so far to say, because consciousness is a virtual “thing”, not material, therefore it doesn't even exist. Usually it is formulated as “consciousness is an illusion”. Besides the fact that an illusion is an existing “thing”, what these people probably mean is that the contents of consciousness are illusory, since they give no reliable and definite picture about what the external world is fundamentally like.
In my opinion this poses the serious question whether or not the paradigm of an exclusively information processing world and the paradigm of the mind/consciousness as a model-building process about that world is itself just an unreliable model. But in what sense would the existence of illusory models in the mind doubt the paradigm of universal information processing rather than merely “confirm” it?
For that question to answer one had to differentiate between the truth that humans indeed facilitate models about the external world in their minds and the contents of those models. Analogous to the claim that consciousness is an illusion (we all know that it isn't), the contents of the models which are part of the mind's approach to predict / grasp / understand what is going on in the external reality are often really illusions (as surely everybody has already experienced at some point in life).
Now, to compress my serious question posed above: is modelling an external world really the reason and cause for consciousness to exist? Or may that assumption (that model!) be incomplete in certain respects? Are there exceptions / inconsistencies to the informational processing model paradigm?
For answering these questions, let me assume that the paradigm would be true. Now, as a matter of fact, I don't like that paradigm and I don't believe in it. My own model is something that I believe can transcend the model paradigm. How can this happen if the model paradigm would be true? According to the latter my own model must be simply the result of an irreducible computational act I have no access to. Therefore, the intermediate result is that irreducible computation is able to compute something that denies the informational processing model paradigm.
Now, for my model to at all being logically possible and true, it would need something that could exist outside, beyond the informational processing model paradigm. Well, the latter assumes such a realm to be factually existent, namely the virtual, immaterial realm of consciousness mentioned above. But in the informational processing world view, this realm is thought of as a deterministic result of the underlying physical dynamics of the brain, and hence is overall information processing in nature, so my own model can't be true yet. But it will "become" true in the course of what I aim to further explain.
The motivation for assuming that consciousness has at all emerged from some death information processing in the external world is that it is assumed to enhance the survival of a huge collection of cells (human and animal bodies) as long as possible. So it must have an evolutionary advantage for a huge collection of cells to cooperate rather than each cell fighting for its own survival.
This may be true, but it really does not explain why in a genuinly information processing world my physical behaviour – which in the information processing paradigm is thought to be determined by computationally irreducible processes – should evolutionary be more adapted when that process is accompanied by some sophisticated model about the world called “consciousness”.
The trick for the information processing paradigm to be convincing is that it uses the - seemingly - logical undecidability about whether it should be considered true or not. Because it assumes that cells as well as huge amounts of cells have no clue that in reality their coalitions into bigger organisms is also pre-determined by blind information processing, as also consciousness and all the rest of history is assumed to be.
Besides the fact that what should be proven by that paradigm is then presupposed to already exist at the level of cells, namely that information processing naturally leads to model-making for the sake of better coping with the external world (what is surely in part true for human beings), the paradigm gives no reasons why the functioning of a pile of deterministically behaving cells should at all develop some consciousness.
The only reason that it gives is anthropomorphising our subjective impresssion to have some control over what is happening in our lifes to the level of cells. Thereby it is assumed that consciousness has some real control over what is deterministically happening with the organism, and that is a strikingly contradiction, since in a world that is seen as a finite automaton, certainly a very complex one, that automaton is a causally closed thing.
Therefore, attributing some causally effective power to the phenomenon of consciousness within the information processing paradigm must be an incorrect modelling and the correct modelling should be that consciousness is merely an unexplainable byproduct of the mentioned paradigm.
But since that paradigm aims to “explain” the existence of consciousness in a non-mystifying way, it tries to give consciousness a certain purpose, namely model-facilitating. When I go on a walk in nature, or when I relax in the open air bath with a beer in my hand, there seems no model-making involved other than to please myself. Does pleasing oneself give an evolutionary advantage? For a healthy person perhaps, since fresh air and fun is surely healthy. But if I have a huge beer belly and nonetheless want to enjoy another couple of beers, isn't that against longer living?
Moreover, there are people that do not wish to live long, how does this come about in a world where only things are valued which give an organism an advantage for living as long as it can? In my opinion, if there is any physical reason that cells have developed to cooperate with each other, this reason then has to be seen in blind selection of the fittest configurations to survive. There was no model-making involved at this stage.
The result of all of this is in my opinion that consciousness can deliberately act against the information processing paradigm with its ambigous assumption that consciousness should make a difference in the course of these information processing affairs.
My consciousness can do this by showing that if it is true that consciousness is a virtual, immaterial “thing”, emerging from underlying physical information processsing, then by definition it cannot causally influence any of these underlying processes, since it is defined as immaterial, virtual. Thus it could not exhibit any advantage for the survival of the organism, and therefore should be considered as superfluous in a mere information processing paradigm that assigns an evolutionary advantage for consciousness to exist.
The relevant point here is that the mentioned paradigm is defined as a causally closed system and that definition does not allow for any superfluous extra-ingredients that are non-material in nature. But nonetheless there does exist that “extra-ingredient”, which in my opinion shows that the information processing paradigm cannot be a complete description of reality and at the same time be consistent. This is also true for the assumption that reality is a kind of non-deterministic finite automaton, since every such automaton is equivalent to a certain deterministic finite automaton.
This is admittedly a very lengthy comment, but I think it may bring some clarity into the discussion and with that I do not mean the discussion here on that site, but generally the discussion about how to handle the phenomenon of consciousness scientifically.