Consciousness and the Persistent Vegetative State.

pogonotomygobbleΤεχνίτη Νοημοσύνη και Ρομποτική

15 Νοε 2013 (πριν από 3 χρόνια και 11 μήνες)

65 εμφανίσεις


1




Consciousness and the Persistent Vegetative State.




Neil Levy





Centre for Applied Philosophy and
Public Ethics,

University of Melbourne

Parkville 3010

Australia


Program on the Ethics of the New
Biosciences

James Martin 21
st

Century School

Univer
sity of Oxford,

United Kingdom



neil.levy@philosophy.ox.ac.uk






‘I think there’s a pain somewhere in the room,’ said Mrs Gradgrind, ‘but I
couldn’t positively say that I have got it.’




Dickens,
Hard Times
.




One of the most controversial ethical is
sues concerns the withdrawal of the means of life
(whether life
-
support or nutrition and hydration) from patients, and the most controversial
such cases concern patients diagnosed as being in persistent vegetative states (PVS). PVS
patients are so controve
rsial because they may seem


to family members and bystanders



2

to be conscious. Withdrawing the means of life from them seems to many to be condemning
to death a person. Hence, even those who countenance the appropriateness of abortion and
of
allowing th
ose in irreversible comas to die may baulk at similar actions directed at PVS
patients.

In this paper, I want to examine recent scientific evidence that has been widely
interpreted as lending support to this reaction; as showing that some PVS patients are
conscious. If this evidence is to be interpreted in the suggested way, then PVS patients

apparently

have lives that are of value: of value to them and (therefore) to others as well.


I shall argue that in fact this evidence changes very little. The majorit
y of the paper will be
devoted to an examination of the empirical basis of the claim that (some) PVS patients are
conscious. I will show that this claim conflicts with our currently accepted beliefs about the
neural basis of consciousness. I will therefore

devote much of the paper to Ned Block’s
challenge to these beliefs. I will argue that we ought to reject Block’s challenge. In addition, I
will show that Block’s argument
actually seems to conflict with, rather than support, the
recent scientific evidence

for the claim that PVS patients are conscious. Additionally, I will
argue that even if Block is right, and PVS patients are (sometimes) conscious, the
implications of this fact for our treatment of them are less significant than might be
imagined.


PVS an
d Consciousness: New Evidence
?


A study recently published
in the journal
Science

(Owen et al. 2006)

raised the question of the
consciousness of PVS patients forcefully enough to attract widespread attention from the
mainstream press.
Owen and colleagues p
robed the brain of a patient diagnosed as PVS
using fMRI. They used four probes, two designed to test language comprehension and two
asking her to follow instructions. The language probes came in two kinds:

those containing
only

unambiguous words, and thos
e including ambiguous words. With unambiguous
sentences, activity was observed in speech specific areas (
the

middle and superior temporal
gyri), in the same areas as those observed healthy controls. When the sentences contained
ambiguous words (e.g., “The
creak

came from a
beam

in the
ceiling
”), there was an additional
response from the left inferior frontal region, a region we know is involved in semantic
processing. Once again, similar activity was observed in healthy controls.


3


The instruction probes wer
e, to my mind, more impressive. Once again, there were two
variants. Variant one asked the patient to imagine playing tennis, variant two asked her to
imagine walking from room to room in her house. In each case, she was asked to engage in
the task for 30
seconds at a time. During the tennis probe, significant activity was observed
in the supplementary motor area; during the navigation task activity was observed in the
parahippocampal

gyrus, the posterior parietal cortex, and the lateral premotor

cortex. In

both
cases, the responses were indistinguishable from those observed in healthy controls.


These results have widely been interpreted as evidence that the patient was conscious,
despite her lack of responsiveness.
One neuroscientist was quoted by the
New
York Times

saying that the study presented


“knock
-
down, drag
-
out” evidence

for

conscious activity


(Carey 2006)
. The suggestion, th
u
s, is that some patients diagnosed on behavioral grounds as
lacking consciousness are actually in something closer to locke
d
-
in syndrome: capable of
normal or near
-
normal cognition, but unable to communicate with the outside world. This
interpretation seems certain to reignite the debate over PVS patients, inasmuch as it suggests
that
withdrawing the means of life
results in t
he death of
someone who is, to all intents and
purposes, like you and me.

Does the evidence gathered by Owen et al. support th
is
suggestion
?


Traditionally, patients like the one tested by Owen et al. are said to be in persistent vegetative
states, where s
uch a state excludes consciousness.

A patient is design
at
ed as in
a
vegetative
state if she meets the following criteria: she exhibits preserved sleep/wake cycles

in the
absence of voluntary motor responses or contingent response to stimuli. In other words
, the
patient is unresponsive to words, gestures or other stimuli. She may engage in spontaneous
activity: moving, crying, laughing, and so on, but her activity is unrelated to external stimuli.
A patient who is minimally responsive to such stimuli is said

to be minimally conscious.
Someone is said to be minimally conscious when they are able, at least sometimes, to follow
commands, to answer yes/no questions (by word or gesture), talk or to respond to stimuli
(Laureys et al. 2005).

The patient tested by Ow
en et al. fit
ted the

standard criteria for a
vegetative state at the time of the examination, but upon re
-
examination five months later

4

she exhibited relatively transient response to a mirr
or slowly moved in front of her,
suggesting that she may

have

be
en

transitioning to the minimally conscious state.


It is important to recognize that
PVS has not usually been understood as excluding islands of
preserved cognitive function.
A number of previous studies have shown task
-
specific brain
activation in patients,

without provoking the re
-
examination of the diagnosis which has been
the response to Owen et al. For instance,
Schoenle & Witzke (2004) measured event
-
related
potentials in the brains of PVS patients
, using sentences ending in congruent or incongruent
wor
ds as stimuli. In normal controls, an N400 response is elicited by the incongruent
endings. 12% of VS patients and 77% of what they describe as near VS patients exhibited
the response, reflecting preserved semantic processing in these patients. Unpublished

data
reported by Perrin showed a P300 response


correlated reliably with recognition


to the
patient’s own name in PVS (Laureys et al. 2005).

The evidence from Owen et al. of semantic
processing in PVS
is
therefore
unsurprising.


W
hy
wasn’t this
earlier

evidence of semantic processing in PVS patients not

interpreted as
showing that PVS patients are
conscious? Isn’t semantic processing mental activity, and
doesn’t mental activity imply consciousness? Semantic processing is indeed mental activity,
but ment
al activity need not be conscious; indeed, it generally is not. In both normal and
pathological cases, subjects frequently engage in mental activity that is not conscious.

Automatic action
s



action
s

carried out by processes that do not need conscious supe
rvision
or initiation


are by far the most common, making up perhaps 95% of all
our
actions
(Bargh
&

Chartrand 1999).

In automatisms of this kind, a conscious subject engages in
actions that do not require consciousness. In addition, there are pathologica
l cases, in which
an apparently unconscious subjects acts.
Consider automatism, most familiar in the form of
somnambulism. Subjects in this state may engage in all kinds of activity, some of it very
sophisticated


driving a car while obeying traffic signa
ls, playing the piano, and so on


in
the (
apparent
) absence of
conscious
ness

(Searle 1994; Broughton et al. 1994)
. Their
responsiveness to the environment would easily earn them the status of minimally conscious,
but they are not intelligently responsive.

Instead, their
actions are routine and stereotyped.



5

Of course, there is room for scepticism about the lack of consciousness of such people.
Though they are clearly not
normally

conscious, and they fail to recall their activity during the
episode, they mi
ght never
theless have some

kind of consciousness for its duration. But there
is convergent evidence from normal subjects
, suggesting that these subjects are not
conscious.

In a very great variety of paradigms, subjects can be shown to have
processed
a
stim
ulus, inasmuch as its content is causally effective on their behaviour,
while remaining
unconscious of it. The most obvious example is priming by masked stimuli

(Deheane et al.
1998)
. A stimulus is shown to a subject very briefly (50ms) and then immediatel
y masked by
another. In this paradigm, the subject reports that she has not perceived the first stimulus.
But she has nevertheless processed its content. For instance, given the masked stimulus
“sheet” and asked to complete a word stem task (“she
-
”) she is

more likely than chance to
complete the stem with the primed word than alternatives (“sheep”, “shear”, “shell”, and so
on).


Most (though, as we shall see, not all) researchers think that we are entitled to take subjects
at their word. Since the subjects
in these experiments report that they
we
re not conscious of
the stimulus, we ought to conclude that they were not conscious of it. We can then use th
i
s

data

to isolate the
neural correlates of consciousness

(NCC); “the minimal set of neuronal events
and me
chanisms jointly sufficient for a specific conscious percept” (Koch 2004: 16).

Awareness of the target stimulus in subjects in these studies is correlated with activation of
higher associative cortices (Deheane et al. 2006). Thus, it seems that we can assu
me that this
pattern of activation is, or is part of (from now on I drop this qualification) the NCC.
Further evidence comes from the correlation between other syndromes and activation of
higher associative cortices. Call the neural state corresponding to
activation of higher
associative cortices
S
.
A
bsence of
S

is correlated with failure to report

stimuli in change
blindness as well as

with neglect (Deheane et al. 2006).

It is also correlated with reported
absence of consciousness in somnambulism and seizu
res (Laureys 2005). But if
S

is the
NCC, then PVS patients are unconscious:
Though PVS patients respond to a variety of
stimuli, their cortical responses are isolated from higher associati
ve cortices (Laureys et al.
2005
).



6

On this basis, we can argue tha
t though Owen et al.
may
provide

evidence for preserved
cognitive function

in PVS

that is more impressive than we previously thought, the best
explanation for the
ir

results will hold that this is preserved function
without

consciousness.
There are two poss
ible hypotheses here. First, we might suggest

that the response is elicited
via the same kind of automatic mechanism that elicits priming in normal subjects. In
response to a suggestion along these lines from Paul Matthews, Owen argues that such an
automat
ic response would last only a few seconds, not the 30 seconds
manifested by the
patient

(Hopkin 2006)
. The patient,
Owen suggests
, is following instructions, and it is that
fact, rather than the details of the neural activation, that demonstrates that she
is conscious.
If this suggestion is right, we might turn to a second explanation: perhaps she
is

following
instructions, but doing so unconsciously. After all, following these instructions does not
seem
much
more impressive than the kinds of activities man
ifested by subjects
undergoing
epileptic fugues or in states of automatism. This might be empirically testable: if the patient
is following instructions unconsciously, we should expect her responses

to

be relatively
stereotypical. We might ask her to carry

out a sequence of tasks (imagine playing tennis, then
imagine navigating your house, then imagine writing your name). If she can follow this
instruction without further prompting, we can conclude she is conscious (of course, if she
fails this test, we can
not conclude that she is
not

conscious: her brain damage might account
for the failure in other ways, e
.
g.
b
y causing short term memory lapses). Until we have such
evidence,
however, we
ought to conclude that her failure to manifest
S
indicates that she is

not
conscious.


How Many Correlates


Unfortunately, nothing in consciousness studies can be taken for granted
. All the evidence
we have cited for the claim that PVS patients are not conscious can be contested.

To put it
another way, the claim that
S

is

th
e NCC can be contested, or, perhaps better, reinterpreted
in a way that makes it consistent with the claim that PVS patients are conscious. The
evidence that
S
is the NCC
consists in the fact that instantiations of
S

correlate with
subjective reports of co
nscious awareness.
Thus, the argument
relies upon using subjective
reports as
the “gold standard” for the study of consciousness

(Block 2005)
; attribution of
consciousness of a stimulus to a subject requires her sincere report that is aware of the

7

stimulus
.

But the idea that there is or ought to be a gold standard in consciousness studies
has come under fire from some quarters. If we take this challenge seriously, then the mere
fact of sincere report c
annot be taken to be conclusive. The way is then open fo
r us to
conclude that
S

is not the NCC; or at least not
the NC
of the kind of consciousness that
matters
here.


The attack

upon the gold standard

has come principally from Ned Block, and builds on his
well
-
known
distinction between two
fundamental concepts

of consciousness,
phenomenal

consciousness and
access

consciousness

(Block 1995)
. Phenomenal consciousness refers to its
qualitative character. A state is phenomenally conscious inasmuch as there is something it is
like to be in it. In contrast, informati
on is access conscious if it is available for rational
control; if it is simultaneously accessible to the decision
-
making, planning and volition
al

parts of the mind. Access consciousness and phenomenal consciousness are
, Block suggests,

highly correlated:
the content of a state that has a qualitative feel to it is
typically
globally
available. But they may sometimes dissociate,
he argues
.

Now, if access consciousness and
phenomenal consciousness can dissociate, then there may be two
kinds

of

NCCs,
correspon
ding to each

kind of consciousness
, and it is an open question to which
S

corresponds.


It might appear that Block’s suggestion, that there are two kinds of consciousness and
therefore two different NCCs, is empirically intractable. If
consciousness c
an
on
ly be
empirically studied via

the reports of subjects,
the dissociability of access and
phenomenological consciousness

would seem to be impossible to demonstrate empirically.
A state that is reportable is
access conscious; subjects would therefore be able

to report all
and only their
access conscious

states
. If a subject had a phenomenally conscious state that
was not
access consciousness, she could not let us know (indeed, it seems that she could not
let
herself

know). So phenomenal consciousness without
access consciousness would be in
principle inaccessible

(o
n the other hand, the reverse dissociation
might

be reportable:
subjects might be able to report states that are globally availab
l
e yet which lack
phenomenality).
The question whether
S

corresponds
to just access consciousness, or
to
phenomenal consciousness

plus access consciousness

would therefore seem to
be
unanswerable
.


8


Block thinks that th
is

methodological claim is false, and that therefore phenomenal
consciousness in the absence of access cons
ciousness is empirically tractable.
He
marshals
several kinds of evidence for this claim. Some of his evidence is evidence for the claim that
access and phenomenal consciousness can dissociate; some for the claim that phenomenal
consciousness is empiricall
y tractable by means other than subjective report.


As evidence for the dissociability of phenomenal and access consciousness, Block appeals to
the everyday experience of suddenly becoming aware of a noise, and aware, moreover, that it
has been
going on fo
r some time. If this experience is veridical, he suggests, we must have
been phenomenally aware of the noise the entire time, though we only became access
conscious

of it

some time after we first became phenomenally conscious of it. But is the
experience v
eridical? Block thinks that
a classic
experiment of
George Sperling’s
demonstrates that it is. Sperling (1960) flashed a 3 x 3 array of letters to subjects for a very
brief period. Subjects typically said that they were aware of
all

the letters, but they c
ould
report only about half of them. Were they genuinely conscious of all of them? Sperling hit
upon the idea of signalling to subjects, using a tone
after

the presentation of the stimulus,
which row they should report (a high tone indicated they should re
port the top tow, a
medium tone the middle row and a low tone the bottom row).
Subjects generally succeeded
in naming the letters in the indicated row, but having done so, they were unable to report
any other numbers. Since they could not know, prior to he
aring the tone, which row they
would be required to report, the fact that they were able to report
any

of the

row
s

indicates
that they were conscious of
all

of them; thus, their claim that

they

were so conscious is
veridical. In other words, Block claims,
the subjects in Sperling’s experiment are
phenomenally conscious of all the letters, just as they claim, but do not have access
conscious to all of them at once (Block 1995).


Having shown, to his

own satisfaction, that phenomenal and access consciousness
can
dissociate,
Block turns to the claim that the dissociation is empirically tractable. He
has two
pieces of evidence for the claim

(Block 2005)
. The

first comes from

studies of binocular
rivalry, in which two images are presented simultaneously to a subj
ect, one to each eye. In
this paradigm, subjects see only one image at a time, with the images switching stochastically.

9

Block notes that the neural correlates of both images in the visual cortex are active
simultaneously, and suggests that the simplest ex
planation
is
that both are

simultaneously

conscious. Only one is reportable at a time, but both are conscious.

If that’s right, the
binocular rivalry paradigm gives us evidence o
f phenomenality without access.


Why not think, though, that only
the reported

image
is conscious, even though both are
processed? Block bolsters his case with
his second piece of
evidence
,

from Signal Detection
Theory

(SDT)
.

He
outlines a task similar to that

used to demonstrate semantic priming. This
time, however, the experiment
is an exclusion paradigm. Subjects are instructed

not

to
complete a word stem with the masked stimulus. Subjects generally find this a difficult
instruction to obey, which has been taken as evidence that the
y are not conscious of the
masked word. But Visse
r and Merikle
(1999)
showed that changing subjects’ motivation can
increase their success at the task. They gave subjects a starting credit of $15, and deducted $1
for each mistake they made. In this variant, subjects were more successful at obeying the

ex
clusion

instruction.
Visser and Merikle suggest that the motivation increased subjects’
attention, with the result that a stimulus that might have been unconscious without the
motivation became conscious. But, following Snodgrass

(2002)
, Block
suggests a r
ival
explanation
. He suggests that the stimulus was always
weakly

conscious, and that the
incentive changes the
criterion
for belief without changing the degree of consciousness.
Though subjects remain very unconfident that they have seen the word, they re
cognize that
if

they have seen a word, it was the target word, and therefore avoid using it to complete the
stem. He claims that this interpretation is bolstered by the fact that on inclusion paradigms,
where the subject is instructed to complete the stem
using the masked word, reward and
punishment has no effect on performance.


Now, assume that
S

is instantiated in the brains of subjects in the exclusion paradigm. Since
we know that
S

is instantiated in masked stimulus paradigms, and the exclusion paradig
m is a
variant on a masked stimulus paradigm, this looks a safe assumption. If
Block is right,
S

is
not, as we thought, an NC for
phenomenal

consciousness. In fact, subjects instantiating
S

are
phenomenally conscious of the stimulus, though they fail to re
port it. Instead,
S

is an NC for
access

consciousness, the kind of conscious which allows for reporting. It follows that the
failure of PVS patients to instantiate
S

is not evidence that they are
phenomenally

unconscious,

10

and the claim that Owen et al. hav
e demonstrated conscious awareness in the PVS state
looks far stronger.


The Significance of Consciousness


The claim that access and phenomenal consciousness can dissociate in the suggested manner
is a disturbing one. If it is true, phenomenal
consciousn
ess may be far more widespread
than previously thought, and this has implications which extend beyond
PVS. One way to
bring out these implications is by considering the topic of pain. Most of us, laypeople and
scientists alike, implicitly accept the gold s
tandard: if someone does not report being in pain,
we conclude that they are not in pain. But if phenomenality can dissociate from access
,

we
cannot be so sanguine People might be in pain without being able to tell us, or themselves.
How do analgesics work
? Do they block
pain
, or merely
access to pain
?

Block himself suggests
that there may be evidence for phenomenally conscious pain which is not access conscious;
evidence from ‘hidden observer’ experiments and, from reports of patients administered a
genera
l anaesthetic that the operation was painful (Block 1995).


It is generally assumed that PVS patients do not experience pain. But if Block
i
s right, we
should rethink this assumption. We know, after all, that at least some of them respond to
painful stimul
i. But since these cortical responses are isolated from higher associative
cortices, it has
been assumed
that they are not conscious of these stimuli (Laureys et al.
2002). But with
S

no longer being supposed to correlate with phenomenal consciousness, we
cannot make this assumption.
PVS patients may lack access conscious, but there is no reason
to believe that they are not phenomenally conscious of noxious stimuli, of spoken word
probes, and the other paradigms used to test for preserved cognitive capaciti
es.


Suppose that this is the case: suppose, that is, that
PVS patients are phenomenally conscious
of their environment, but lack access consciousness. How should this affect the moral
debate over withdrawing life support from them? In fact, it would affec
t it surprisingly little.
Phenomenal consciousness is, plausibly, a necessary condition of being a moral patient; a
being whose welfare matters morally. If PVS patients are sentient, then it matters what we
do to them. We can benefit them by causing them p
leasure and harm them by causing them

11

pain. To that extent, their moment
-
by
-
moment states are of potential
value

and
disvalue

to
them: they can suffer (on the assumption


contra

Carruthers (2004
)


that the awfulness of
pain consists in its phenomenality)
. However, their
lives

are of little positive value to them. In
the absence of access conscious and the sophisticated mental states that depend upon

it
,
such as the ability to conceive of oneself as a being persisting through time, to recall one’s
past, to

plan and to have preferences for how one’s life goes, PVS patients are only



and
very minimally


m
oral patients. It matters to them what state they are in, at any moment in
time, but not how their lives go over time. Since their continuing existence is
not of value to
them, continuing to live is not in their interests. We do them no wrong by allowing them
(painlessly) to die.


That is not to say that the finding that PVS patients are phenomenally conscious would not
be of practical import.
Given current
treatment regimes and methods, PVS patients may
frequently be in pain. We do them wrong if we fail to treat their pain


if they are really
phenomenally conscious. But are they really phenomenally conscious?


Defending the Gold Standard


Let us turn now t
o two questions. First, do Block’s arguments
bolster the arguments of
Owen et al? Second,
just
h
ow strong are Block’s arguments?


If Block is right, then PVS patients might be phenomenally conscious even though they fail
to instantiate
S
. But it does not f
ollow that Owen et al. can appeal to Block’s view to support
their claim. Indeed, if Block is right, the results reported by Owen et al. are
at least as

mysterious as on the traditional view. If
S

is the NC of access consciousness, then we should
expect PV
S patients, who fail to instantiate
S

to be unable to use their phenomenally
conscious states in cognition. Yet the results of Owen et al., and in particular the navigation
and imagination results which have most impressed observers,
would seem better evid
ence
for
preserved
access

consciousness

than for
phenomenality. If the patient were (only)
phenomenally aware of the experimenters’ instructions, she should fail to follow them;
indeed, perhaps even to process their semantic content. To that extent, Block’
s claim is no

12

help to Owen et al. at all.

Nevertheless, Block’s

argument supports the
speculation

that PVS
patients m
ight be phenomenally conscious; that is, if his argument bears scrutiny.


There are, however, good reasons to be sceptical of Block’s clai
ms. All the phenomena he
cites are susceptible to explanations consistent with the gold standard. Consider Sperling’s
experiment. Block takes the fact that subjects can report the content

of any of the rows, but
not of all of them
, to show that they are ph
enomenally conscious of all but only access
conscious of the row they are asked to report. Yet access is a
dispositional

notion: content is
access conscious if it is “
poised

for
free

use

in
reasoning

and for direct ‘rational’ control of
action and speech”
(Block 1995: 382). Content is not access conscious when it is reported,
but when it is
reportable
.
It follows from the fact that subjects are able to report any row that
they are access conscious of all the rows.

If Block is to show that Sperling’s subject
s
experienced a dissociation between phenomenal and access conscious, he must show that
they
remain

phenomenally conscious of the array after reporting the indicated row. This he
has not done (of course, they may remember being phenomenally conscious of th
e entire
array, but that is different
from
continuing to be phenomenally conscious of the entire
array).


What of his empirical evidence for the tractability of the supposed dissociation between
access and phenomenal consciousness? As we saw, Block’s claim

here rested largely on
Snodgrass’s
SDT

interpretation of masked stimuli paradigms; that is, on the claim that
subjects in these experiments
are

phenomenally conscious of these stimuli, but are
unconfident regarding what they have seen. Yet Snodgrass’s rei
nterpretation of these
experiments is mysterious. As we saw, he claims that SDT explains why motivated subjects
perform better at exclusion paradigms. In fact, on SDT the real mystery is why
unmotivated

subjects do so badly on these paradigms. The subject,

recall, is asked to complete a word
stem using anything
other

than the masked stimulus word. Now, suppose the subject has
perceived the stimulus (say “
sheep
”) but lacks confidence that she has seen it. Given,
however, that she seeks to complete
the stem w
ith a word other than

that

presented,
shouldn’t she avoid it in any case? No matter how low her confidence that she saw
the word
“sheep”, her confidence that
she saw

“sheep” should be higher than her confidence tha
n

13

that she saw
any other word.

Hence, she
should avoid completing the word stem with
“sheep”.


Snodgrass, and once again Block following him, claims that SDT theory is preferable
because it gives a superior explanation for why motivation improves performance on
exclusion paradigms, but leaves perf
ormance on inclusions paradigms unchanged. But, even
leaving aside the mysterious performance of
unmotivated

subjects on exclusion trials, this
claim seems to be false.

Suppose th
e traditional
hypothesis
,

that motivation heightens
attention, and thereby re
nders

other
w
ise

unconscious stimuli conscious,

is
true
. Would we
expect the observed differences between inclusion and exclusion paradigms? We might.
If
subjects generally complete word stems with the first thought that comes to mind


which is
what the pr
iming experiments

seem to

show


on both inclusion and exclusion studies, then
we would get the observed pattern of errors on exclusions.
H
eightening attention would
allow subjects to correct their errors on
these
trials.
But there are few

or no
errors on
inclusion

trial
s,

and therefore little to correct. Hence,
giving people an extra incentive would
not have any effect

on inclusion trials
.

It is worth noting here that there is widespread
agreement that whether a subject is conscious of a stimulus is suscep
tible to top
-
down
influences, as well as bottom
-
up. That is, it is not only the features of the stimulus


its
content, the length of presentation and its context


that influence whether a subject is
conscious of it, but also her
expectations
(Deheane et a
l. 2006). The motivational
manipulation can therefore be parsimoniously understood within th
e traditional
framework.


Of course, Block may simply deny that the widespread claim that top
-
down factors can
influence
whether or not

a stimulus is conscious.

He
can point out that evidence that top
-
down factors have this effect comes from subjective report, but since subjective report
requires access consciousness, we cannot take it as a guide to phenomenal consciousness.
Block may therefore suppose that top
-
down
factors only affect access consciousness,
without altering phenomenality; alternatively (or perhaps equivalently) he may say that top
-
down factors work in the same way as the motivation in the exclusion paradigm, influencing
only the
subject’s criterion fo
r belief.



14

Block

faces a further

problem, however. He must explain the

winner
-
take
-
all feature of
consciousness: the fact that subjects asked to report the degree of visibility of a stimulus
hardly ever use intermediate grades (Deheane, Sergent & Changeux
2003). If phenomenal
consciousness can come in degrees, then why don’t subjects report this fact?
The natural way
for him to proceed is to hold that though phenomenal consciousness comes in degrees,
access consciousness

doesn’t

(at least, not in this exper
iment)
. Subjects only report the
presence of a single percept because they only have access to one of the stimuli, both of
which they are

nevertheless

conscious of. Thus we have phenomenality in the absence of
access. But,
at best, Block gives us no reason

to believe this claim.


Why can’t subjects in exclusion paradigms report the stimulus regarding which they lack
confidence (or even that they thought they saw
something
)? Block doesn’t say. If motivation
simply changes the criterion for belief, then the
stimulus ought to be
weakly

access conscious.
In fact,
nothing

Block says gives us any

reason to suppose, even on the SDT
explanation of
how motivation alters performance on the exclusion paradigm, that access and
phenomenality dissociate.
Indeed, it seems

that SDT itself is committed to the correlation of
phenomenal consciousness and access consciousness: subjects fail to report the stimulus
because they have weak access consciousness of it, and they have weak access consciousness
of it
because

they are on
ly weakly phenomenally conscious of it. Moreover, the explanation
endorsed by Block
, according to which subjects are phenomenally aware of the stimulus but
fail to report it simply because they lack con
fidence in their perception,
seems to entail that
acce
ss conscious is preserved
prior

to the motivational manipulation. It might be weakened,
but we need not assume even tha
t; it might be fully preserved (we can account for the
weakness via the fleetingness of phenomenal consciousness, without supposing any
c
orrelative weakness in access). Since Block gives us no reason to suppose that access was
absent prior to the motivational manipulation, he is unable to account for the

winner
-
take
-
all
feature of consciousness
, here or elsewhere
. Worse


for him


we have
at least as much
reason to suppose that
access
is perfectly correlated with phenomenality in the exclusion
paradigm as to suppose that they dissociate.
So Block
appears to be
wrong: the exclusion
paradigm gives us

no reason to think that phenomena
lity is e
mpirically tractable in a way that
bypasses access.



15

The same point can, it seems, be made about all of Block’s cases in which access
consciousness and phenomenal consciousness supposedly dissociate. Even setting aside the
fact that there are plausible alt
ernative explanations for all of them, nothing Block says gives
us any reason to suppose that in any of them the degree of phenomenal consciousness is
greater than the degree of access. He has failed to show that the two come apart;
a fortiori
, he
has fail
ed to show that the alleged dissociation is empirically tractable. The SDT explanation
of motivated performance on the exclusion paradigm, Block’s major piece of evidence for
the empirical tractability claim, is in fact multiply vulnerable. Without this ev
idence, however,
we have no reason to

reject the gold standard (at least, not yet: empirical results which would
allow us to do are conceivable). Nor, more pertinently, do
we have any reason to accept that
instantiations of
S
correspond to access conscious
ness without phenomenality. Thus, the
claim that we know that PVS patients are unconscious
because
they fail to instantiate
S

remains a strong one.


None of this shows, of course, that access consciousness and phenomenal cannot dissociate.
Indeed, given ou
r reliance upon the gold standard, there seems no way to rule out the
possibility of such a dissociation. Since we can only report, even to ourselves, states of which
we are access conscious, we cannot rule out the possibility that there are phenomenally
c
onscious states of which we are not access conscious. However, in the absence of evidence
for

the claim, we are entitled to treat it as a sceptical hypothesis
, akin to the claim (for
instance) that some of the people we see around us are
actually

zombies (
which also seems to
be empirically intractable). In the absence of evidence, we are entitled to the presumption
that states, of ourselves and others, that
seem

not to be conscious actually
are

not conscious.
On the basis of this assumption, and the further

piece of evidence that these unconscious
states correlate with the absence of brain state
S
, we are entitled to conclude that
S

is the NC
of phenomenal consciousness. And on that basis, we ought to conclude that PVS patients
are not phenomenally conscious
.


16

References


Bargh, J. A., & Chartrand, T. L. 1999. The
unbearable automaticity of being
.
American

Psychologist

54: 462
-

479.


Ned Block. 1995. On a confusion about a function of co
n
sciousness.
Behavioral and Brain
Sciences

18: 227
-
287.


Ned Block. 20
05. Two Neural Correlates of Consciousness.
Trends in Cognitive Sciences

9: 46
-
52.


Broughton, R. Billings, R. Cartwright, D. Doucette, J. Edmeads, M. Edwardh, F. Ervin, B.
Orchard, R. Hill and G. Turrell. 1994. Homicidal somnambulism: A case report.
Sleep

17:
253
-
64.


Carey, B. 2006.
Vegetative Patient Shows Signs of Awareness, Study Say.
New York Times
,
September 7.


Carruthers, P. 200
4
.

Suffering without subjectivity.
Philosophical Studies

121: 99
-
125.



Deheane, S., Naccache, L,Le Clec’H G., Koechlin E
., Mueller M., Dehaene
-
Lambertz G.,
van de Moortele P.F., Le Bihan D.1998.

Imaging unconscious semantic priming.
Nature

395:
597
-
600.


Deheane, Sergent & Changeux 2003.

A
neuronal network model

linking Proceedings of the
Natioanl


Stanislas Dehaene, Jean
-
Pierre Changeux, Lionel Naccache, Jérôme Sackur and Claire
Sergent. 2006.
Conscious, preconscious, and subliminal processing: a testable taxonomy.
Trends in Cognitive Science

10: 204
-
211.


Hopkin, M. 2006. ‘Vegetative’ patient shows signs of conscious thou
ght.
Nature

443: 132
-
133.


Koch, C. 2004.
The Quest for Consciousness: A Neurobiological Approach
. Roberts & Company.


Steven Laureys, 2005. The neural correlates of (un)awareness lessons from the vegetative
state. TICS 9: 556
-
559.


Laureys S, Faymonville
ME, Peigneux P, Damas P,Lambermont B, Del Fiore G, Degueldre
C, Aerts J, Luxen A, Franck G, Lamy M, Moonen G, Maquet P. 2002. Cortical processing of
noxious somatosensory stimuli in the persistent vegetative state.

Neuroimage
17: 732
-
41.


Laureys S, Perr
in F, Schnakers C, Boly M, and Majerus S. (2005). Residual cognitive function
in comatose, vegetative and minimally conscious states.
Curr Opin Neurol
,
18
:
726
-
733.


17


Owen,
A.M.,

Coleman,

M.R.,
Boly,

M,

Davis,

M.H.,
Laureys,

S. &
. Pickard
, J.D.

2006
.

Detect
ing Awareness in the Vegetative State.
Science

5792: 1402.


Schoenle
, P.W &
. Witzke
, W.2004.

How vegetative is the vegetative state? Preserved
semantic processing in
vegetative state

patients
: e
vidence from N 400 event
-
related
potentials
.
Neurorehabilitati
on

19: 329
-
334.


Searle
, J.R. 1994.
The Rediscovery of the Mind.

Cambridge, Mass.: MIT Press.


Snodgrass, M. 2002.
Disambiguating

conscious

and

unconscious influences: Do exclusion
paradigms demonstrate unconscious
perception?
American Psych
ology

115: 545
-
580.


Sperling, G. (1960) The information available in brief visual presentations.
Psychological
Monographs

74: 1
-
29.


Visser
, T. A. W. &
Merikle
, P. M. 1999. Conscious and unconscious processes: The effects
of motivation.
Consciousness and
Cognition

8: 94
-
113.