xi Consciousness is the biggest mystery. It may be the largest outstanding obstacle in our quest for a scientific
understanding of the universe.
xiv The problem of consciousness lies uneasily at the boarder of science and philosophy.
p.4 When we perceive, think, and act, there is a whir of causation and information processing
p.12 what it means for a state to be phenomenal is for it to feel a certain way,
and what it means for a state to be psychological is for it to play an appropriate causal role...
A specific mental concept can usually be analyzed as a phenomenal concept, a psychological concept, or as a combination of
the two.
p.12 [Rene] Descartes held that every event in the mind is a cogitatio, or a content of experience.
p.13 Sigmund Freud and his contemporaries solidified the idea that many activities of the mind are unconscious,
and that there can be such things as unconscious beliefs and desires.
p.22 Psychology and phenomenology together constitute the central aspects of mind.
p.22 Conscious experience does not occur in a vacuum. It is always tied to cognitive
processing, and it is likely that in some sense it arises from that processing. Whenever one has a sensation, for
example, there is some information processing going on: a corresponding perception, if you like.
p.27 Attention. We often say that someone is conscious of something precisely when they
are paying attention to it; that is, when a significant portion of their cognitive resources is devoted to dealing
with the relevant information. We can be phenomenally conscious of something without attending to it, as witnessed
by the fringes of a visual field.
p.27 If one were to try to explain attention, one might devise a model of the cognitive processes
that lead to resources being concentrated on one aspect of available information rather than another... It is clear
that there is a phenomenal and a psychological property in the vicinity of [the concept of attention].
p.28 Awareness can be broadly analyzed as a state wherein we have access
to some information, and can use that information in the control of behavior... Awareness of information
generally brings with it the ability to knowingly direct behavior depending on that information.
p.28 a conscious experience is reportable. If I am having an experience, I can talk about the fact
that I am having it. I may not be paying attention to it, but I at least have the ability to focus on it and talk about it,
if I choose. This reportability immediately implies that I am aware in the relevant sense.
p.29 Newell (1992)... describes awareness as "the ability of a subject to make
its behavior depend on some knowledge,"
p.31 Cognitive models are well suited to explaining psychological aspects of consciousness. There
is no vast metaphysical problem in the idea that a physical system... should be able to deal rationally with information from
its environment, or that it should be able to focus its attention first in one place and then in the next.
p.111 If we have a model that captures the causal dynamics of someone who is learning, for example, it follows
that anything instantiating those dynamics in the right environment will be learning.
p.114 In Consciousness Explained (1991), Dennett puts forward a more sophisticated account
[of the concept of consciousness] that draws on much recent work in cognitive science. The model proposed
here is essentially a "pandemonium" model, consisting in many small agents competing for attention, with the agent that shouts
the loudest playing the primary role in the direction of later processing. On this model there is no central "headquarters"
of control, but multiple channels exerting simultaneous influence... If successful, it would provide an explanation
of reportability, and more generally of the influence of various sorts of information on the control of behavior.
It also provides a potential explanation of the focus of attention... Unlike most authors who put forward cognitive
models, Dennett claims explicitly that his models are the sort of thing that could explain everything
about the experience that needs explaining. In particular, he thinks that to explain consciousness, one only needs
to explain such functional phenomena as reportability and control
p.220 awareness is... a state wherein some information is directly accessible and available for
the deliberate control of behavior
p.225 A natural suggestion is to modify the definition of awareness to something like direct availability
for global control. That is, a subject is aware of some information when that information is directly available
to bring to bear in the direction of a wide range of behavioral processes. [JLJ - available for the satisfaction
of needs, the solution of problems, or for judgments of suitability or quality, or for the activation of scripts]
p.281 Bateson (1972): information is a difference that makes a difference.
p.282 physically realized information is only information insofar as it can be processed. As Mackay (1969)
puts it, "[I]nformation is as information does." ...Information is a difference that can make a difference in transmission.
p.327-328 A popular objection to artificial intelligence (e.g., Searle 1980, Harnad 1989)
is that a simulation of a phenomenon is not the same as a replication. For example, when we simulate digestion
computationally, no food is actually digested. A simulated hurricane is not a real hurricane; when a hurricane is simulated
on a computer, no one gets wet. When heat is simulated, no real heat is generated. So when a mind is simulated, why
should we expect a real mind to result? Why should we expect that in this case but not others, a computational
process is not just a simulation but the real thing?
p.328 I suggest that the answer is as follows: A simulation of X is an X precisely when the property
of being an X is an organizational invariant. The definition of an organizational invariant is as before: a property
is an organizational invariant when it depends only on the functional organization of the underlying system, and not on any
other details. A computational simulation of a physical system can capture its abstract causal organization,
and ensure that that causal organization is replicated in any implementation, no matter what the implementation is made out
of. Such an implementation will then replicate any organizational invariants of the original system, but
other properties will be lost.
p.332 It might be, for example, that a computation that mirrors the causal organization of the brain at
a much coarser level could still capture what is relevant for the emergence of conscious experience... one could argue that
the centrality of computation in the study of cognition stems from the way that computational accounts can capture almost
any sort of causal organization... Whatever causal organization turns out to be central to cognition and consciousness,
we can expect that a computational account will be able to capture it. One might even argue that it is this flexibility
that lies behind the often-cited universality of computational systems.
