S. Laureys (Ed.)
Progress in Brain Research, Vol. 150
ISSN 0079-6123
Copyright r 2005 Elsevier B.V. All rights reserved
CHAPTE
ro
an
d, U
ntly
: aft
bject
e hu
of th
gdo
eve
an a
ot a
se in
will grow as brains grow. Hence, consciousness is most likely to be a continuously variable property of the
lated that could determine an ‘‘assemblage,’’ i.e., a
neuronal assemblies. Such assemblies, where 10
million neurons can synchronize in activity over
on (iii) the availability of facilitating ‘‘modulato-
lar proteins, function as the physical correlate of
‘‘mood’’. These chemicals will originate from
It is this two-way iteration of chemicals, of which
11DOI: 10.1016/S0079-6123(05)50002-5the assembly size is a mere index, that can be
viewed as ‘‘consciousness’’ (Fig. 1).

Corresponding author. Tel.: +44 1865 271852;
Fax: +441865271853; E-mail: susan.greenfield@pharm.ox.ac.ukmerely 230ms, have been established for over a
decade (Grinvald et al., 1994). It has previously
widespread sources in the brain, and also from
the rest of the body, including the immune system.quantifiable measure reflecting the combined tem-
poral and spatial dynamics of constantly changing
ry’’ chemicals that, through synthesis/breakdown
and phosphorylation/dephosphorylation of cellu-brain, in both phylogenetic and ontogenetic terms. Here, we describe how modern techniques may be
utilized to determine the physiological basis of consciousness.
Introduction
The only way in which the physical brain can ac-
commodate the ebb and flow of a continuously
variable conscious state, would be at the interme-
diate level between macro brain regions and indi-
vidual neurons: the level of highly transient
assemblies of brain cells that wax and wane in
size, from one moment to the next. As yet, this rate
of turnover of assemblies and their size at any one
time has not been quantifiable. Perhaps in the fu-
ture, however, a unit of measure could be formu-
been suggested that these transient, three-dimen-
sional configurations of large-scale assemblies
throughout the brain, which need not respect con-
ventional anatomical boundaries, will correlate
with different degrees of consciousness at any one
moment (Greenfield, 1997, 2000a, b). These as-
semblies will vary in size from one moment to the
next, according to (i) the strength of the trigger (a
hub of neuronal circuitry, analogous to a stone in
a puddle) that initiates their transient synchrony,
and also (ii) according to the ease with which the
neurons will be synchronized, in turn dependentA neuroscientific app
Susan A. Greenfield

Department of Pharmacology, Mansfield Roa
Abstract: For a neuroscientist, consciousness curre
the phenomenon is distinct from self-consciousness
extreme emotion, but still be accessing a sentient, su
does not appear to be an exclusive property of th
formation in either the anatomy or the physiology
animals, no phylogenetic Rubicon in the animal kin
is crossed as the brain grows in the womb, no single
at birth, when consciousness might be generated in
view of consciousness might be therefore that it is n
but that it is a consequence of a quantitative increaR 2
ach to consciousness
d Toby F.T. Collins
niversity of Oxford, Oxford, OX1 3QT, UK
defies any formal operational definition. However,
er all, one can ‘‘let oneself go,’’ when experiencing
ive, conscious state. This raw, basic subjective state
man brain. There is no obvious qualitative trans-
e central nervous system of human or non-human
m. Similarly, there is no clear ontogenetic line that
nt or change in brain physiology, and certainly not
ll-or-none fashion. A more plausible, and scientific,
different property of the brain, some magic bullet,
the complexity of the human brain: consciousness

12Circulating
peptide(s)
Action on
Rest of body
Degree of sensory
stimulation
Extent of
associations
Formation of
Competing
assembly
Neuronal
assembly size
Availability of
other modulators
Availability
of amines
Fig. 1. A possible mechanism for the generation of conscious-
ness. The two sets of concentric circles represent two transient
assemblies of tens of millions of brain cells: the largest assembly
will dominate at any one moment in the brain, and determine
that moment of consciousness. The degree to which cells are
recruited, and hence the degree of consciousness, will be deter-
mined by a variety of factors, such as vigor of sensory inputs,
pre-existing connections (‘‘associations’’), and degree of com-
petition (‘‘distractions’’), as shown by the smaller assembliesImaging of these transient assemblies would
provide a net index of consciousness, a much
sought after, and as yet elusive, ‘‘correlate of con-
sciousness’’ that is both necessary and sufficient
(Greenfield, 2000a). It is indeed possible to gener-
ate potentially falsifiable hypotheses as to how
different factors, in turn expressible simultaneous-
ly in both phenomenological and neuroscientific
terminologies, will result in neuronal assemblies of
varying size and hence varying degrees of con-
sciousness. In neuroscientific terms, let us there-
fore assume that a subjective state is a transient
neuronal assembly of varying degree, recruited by
the activation of a hard-wired hub, or network of
neurons. Such a hub, once activated by some in-
put, could trigger a temporary formation of a
much larger assembly of brain cells, rather like a
highly durable stone that can elicit far more ex-
tensive, yet highly transient, ripples in a pond. In
receive instruction. The spatial resolution is high
starting to form. Signature chemicals, such as peptides, will be
released from the transient assembly. The type, number, and
concentration of these chemicals will thus represent the salient
assembly in the brain, and convey that information to the rest
of the body, via the circulation. In turn, chemicals released from
the immune system and the vital organs will modify the work-
ing assembly of neurons, as will other chemicals such as hor-
mones and the ‘‘amines’’ that are released in relation to arousal.
Consciousness is therefore dependent on the whole body work-
ing cohesively (see Greenfield, 2000b).and there is little or no long-term exposure damage
reported. On the other hand, the time delay in an
area of neuronal activity and its detection can be in
the region of 6–10 s. Although this can be math-
ematically corrected, this temporal delay makes it
difficult to dissect the sub-second activity of the
brain, but instead is used to detect gross regions of
activation.
To combat the problem of time resolution, an
imaging technique based on changes in brain con-
ductivity has been developed for use on human
volunteers, called functional electrical impedancesuch a way, different mental states could be cor-
related with different sizes of neuronal assembly,
and as such provide a bridge between phenome-
nology and physiology (Table 1).
The appeal of the predictions made in Table 1 is
that they are testable, i.e., that different parame-
ters, such as amine availability, age, and stimula-
tion, could be differentially varied, and the
ensuing, corresponding assembly size, noted. The
primary problem, however, is with the limitations
of the techniques currently available.
The neuroscience portfolio: brain imaging
The most obvious means of monitoring the for-
mation of transient neuronal assemblies in the
conscious brain is by imaging technology. How-
ever, within the range of methodologies available,
no particular approach is ideal. First, there is
functional magnetic resonance imaging (fMRI).
This is a very powerful non-invasive method of
visualizing regions of activity within the intact
brain of human or animal patients (Hirsch, this
volume). Areas of neuronal activity have a higher
demand for oxygen, which is carried by the hem-
oglobin molecule. A powerful magnet is able to
detect the changes in the magnetic field that are
generated by a hemoglobin molecule when it re-
leases oxygen. The great advantage of this tech-
nique is that subjects do not need to be injected
with tracer substances, as with positron emission
tomography (PET), and since they are not re-
quired to be under anesthetic, can be fully coop-
erative and can perform necessary tasks and