Re: Dennett: Consciousness

From: Harrison, Richard (
Date: Wed Jan 10 1996 - 12:41:15 GMT

On Dennet and Kinsbourne (1992); Time and the observer: The where and
when of consciousness.

Their headings were quite useful so here's another section by
section summary...

D & K compare the way two models of consciousness treat subjective

1. Two models of consciousness

1.1 Cartesian materialism. Is there a "central observer" in the

Where there is a mind there is a 'point of view', a "locus of
subjectivity". However, within the brain there is no single point
where all information funnels in. D & K propose that when, for
example, different parts of the brain respond to different aspects of
a visual stimulus (shape, colour, motion, recognition etc) there is
no single place where it all comes together to give our 'stream of
conscious' (concearning that stimulus).
The models;
    1. The Cartesian Theater (CT). All sensory information converges to
     one point in the brain from which our stream of consciousness is
    2. The Multiple Drafts Model (MDM). There is no central point.
     Localised discriminations in different areas of the brain are
     not re-presented for a more central discriminator (thus avoiding a
     homunculus). Our stream of consciousness is a number of parallel
     streams of conflictng and continuously revised contents (drafts)
     where no single one is the true version. Also the objective temporal
     properties of particular states do not necesserily determine the
     temporal properties of subjective experience.
1.2 Some "temporal anomolies" ot consciousness

Under various conditions people report experiences in which the
temporal ordering of the elements in their consciousness, or the
temporal relation of those elements to concurrent activity in their
brains, seem to be anomolous. D & K propose that their model can
account for these anomolies.

A. Colour phi. Involves apparent motion. When two dots are briefly
lit in rapid succession a single spot is seen to move. If the two
spots are different colours, a moving spot is seen that changes
colour from one to the other abruptly at the mid-point. The question
is; How can we fill in the mid-point and colour change before the
second flash occurs?

B. The cutaneous "rabbit". When a subjects arm is mechanically tapped
in different places (e.g. 5 taps on the wrist then 2 near the elbow
then 3 more on the upper arm) they report the taps feeling as if they
are steadily moving up ther arm. Again, how did their brains know
that after the wrist there would be elbow taps? (NB If only the wrist
is tapped then the taps are all experienced in one place). Perhaps
the brain delays conscious experience untill after all the taps have
been percieved.

C. "Referral backwards in time". Direct electrical stimulation of the
brain can induce sensations around the body. Libet compared the time
for this to occur with the time taken to induce similar sensations in
a more usual way (electrical impulse to the hand). Whereas cortical
stimulation took up to 500 msec to produce a reported experience, hand
stimulation led to a far quicker response (despite the longer
distance the message needed to travel).

D. Subjective delay of consciousness of intention. In other
experiments Libet asked subjects to make "spontaneous" decisions to
flex one hand at the wrist while noting the position of a clock. He
found the conscious decisions lagged up to 350-400 msec behind the
onset of "readiness potentials" recorded from scalp electrodes.

In each example there is an anomoly between the subjective sequence
of the stream of consciousness and the objective sequence of observed
events in the environment and nervous system. D & K say this doesn't
necesserily challenge the basic principle that causes preceed
events but that erronous presuppositions make these phenomena appear

2. The models in action: Diagnosing the tempting errors

2.1 The representation of temporal properties versus the temporal
properties of representations.

We must distinguish between the features of the representings from
the features of the represented (e.g. an oil painting of an artist
making a charcoal sketch). There is no reason to suppose that the
beginning of the representations represents the beginning of the
represented. What is important is not when particular events are
represented but their temporal content (e.g. how long ago did x
So long as the experiences come between an upper constraint (occurs
in time to control the appropriate behaviour) and a lower constraint
(the earliest time at which infomation can arrive at the system) it
does not matter in what order the representations occur.

2.2 Orwellian and Stalinesque revisions: The illusion of distinction

D & K challenge the central CT idea that there is a place in the
brain where consciousness happens and all revisions come via this
location by looking at a thought experiment.
They consider two ways of making you remember events that didn't

    1. Insert a memory of something that didn't happen after it was
     supposed to happen - an Orwellian (from the Ministry of Truth in
     Orwell's 1984) revision.
    2. Stage show trials of false events with bogus testimony,
     confessions and evidence - a Starlinesque ploy.

The difference comes down to a memory (post-experiential) versus a
perceptual (pre-experiential) distinction. If a CT model is correct
then the distinction can be maintained, but if it is wrong then this
is not necesserily the case.
They show that with the phi colour phenomenon both explanations,
that your memory plays tricks on you (Orwellian) or that your eyes
play tricks on you (Starlinesque), are consistent with whatever the
subject (even if this is you) says, thinks or remembers.
Both versions of the CT model can account for all the data. Of verbal
reports - one says they are innocently mistaken, the other that they
are accurate reports of experienced "mistakes". Of nonverbal effects
- one says they are the result of unconsciously discriminated
contents while the other says they are the result of consciously
distributed but forgotten contents. And of subjective data - mistaken
experience and immediately misremembered experiences would feel the
The MDM does not involve any attempt to arbitarily draw a line
between conscious and unconscious. It makes explicit the idea that
once a disciminationis made it does not have to be made again. In the
phi colour example, the brain adjusts to the conclusion that is
drawn making the new interpritation available for the control of
subsequent behaviour. The new information needs only to include a
reference to the past time.

D & K argue that their model stops the unanswerable questions that
the Cartesian model leads to from arising.

3. The Libet controversies re-examined.

3.1 Libet's experiment allegedly showing "backwards referral".

As well as methodological problems in Libet's experiments there are
problems with his interpretation.
Backwards referral (representations contain a temporal reference) is
 more plausible than backwards projection (representations actually
projected backwards in time to some CT).
D & K say Libet argues the Starlinesque viewpoint and Churchland (the
critic) makes the Orwellian countermoves.

To be honest I found this section confusing although a central
question seems to be; Is there a conscious fact of the matter? (?)

3.2 Libets claims about the "subjective delay" of consciousness of

The concept of absolute timing of experience were further
investigated in Libet's experiments involving timing conscious intent
and readiness potetials. The principle finding that consciousness of
intent lags upto 500 msec behind the relevant brain events. Hence
Libet's conclusion that real decision making occurs prior to consciousness
with conscious intent at most having an "executive veto" role in
decision making.

A problem with this methodology is that the self-monitoring is a
process that changes the task stopping the desired "spontaneous"
nature of the task.

3.3 Grey Walter's experiment: A better demonstration of the central
contention of the MDM.

Libet's experiment created an artificial and judgemental task that
robbed the results of their hoped for significance. Grey Walter
(1963) testeed the hypothesis that certain bursts of cortical
activity were the initiators of intentional action. Subjects looked
at slides and pressed a button when they wanted to see the next. The
button was a dummy, what moved the slides was cortical activity.
Subjects reported that they were surprised when the slides moved
just as they were "about to" push the button. D & K propose a MDM
explanation. The delay between cortical activity and decision is due
to expectations being set up by a decision to change the slide
being tuned to expect feedback after a certain time period or report
back with alarm under other condition. The fact about where the
alarm is interpreted as happening in the subjective sequence shows
nothing about when in real time the consciousness of the decision to
press the button first occured.

4. Conclusion

The representation of sequence in the stream of consciousness is a
product of the brains interpetative processes, not a direct
reflection of the sequence of events making up those processes. In
the same way as space perception does not use space-in-the-brain,
time perception does not use time-in-the-brain. D & K suggest this is
true in the wider context not just in the short timescales they have

Although it is important to learn where and when various
informational streams converge (etc) the details we find tell us
nothing of consciousness. The temporal sequence in consciousness is
a matter of the temporal content represented not the timing of the

This archive was generated by hypermail 2b30 : Tue Feb 13 2001 - 16:23:58 GMT