In recent years, the
movement dedicated to applying neuroscience to traditional philosophical
problems has been gaining momentum and has become very influential. Cognition
and the Brain is a collection of 12 essays clustered around five themes:
data and theory in neuroscience, neural representation and computation,
visuomotor transformation, color vision and consciousness. The authors attempt
to give an overview of the basic questions about human cognition, many of which
have been studied for millennia, with the conviction that these questions will
be answered only by a philosophically sophisticated grasp of what contemporary
neuroscience is teaching us about how the human brain processes information. It
is an attempt to bridge the gap between philosophy and neuroscience by applying
the methods of neuroscience to traditional philosophical problems and using
philosophical thinking to illuminate issues in neuroscience. I find the book
very informative about the current state of neuroscience and the attempt well
motivated. Unfortunately, the book is for the most part, not accessible to the
non-expert in philosophy or neuroscience.

The book starts with
Andrew Brook’s and Pete Mandik’s (editors) assertion that the answer to certain
traditional philosophical problems about the mind will be answered only
by a philosophical ‘grasp’ of what contemporary neuroscience is teaching us and
that the evidence for this (admittedly bold) proposition is overwhelming. By
appealing for example, to the problem of perception they state: ‘the
philosophical problem of perception has been transformed by new knowledge about
the vision systems in the brain’ (p.3). But even if hypotheses such as that we
have two complementary visual systems are proved to be true what makes us think
that the answer to the traditional philosophical problem of perception lies in
the neurosciences? One main difficulty for instance (neurophysiologically
speaking), is that it appears that there is nothing intrinsic in the brain that
constitutes the difference between a red and a green quale — a patch of color
inherent in a sensory state. It appears so to speak that there is no
neurophysiological difference between these two states; to the best of our
neurophysiological knowledge, that is, there are no anatomical differences in
cells in the visual cortices that correlate with colour differences[1].
As V.G. Hardcastle (1995) has said, the best neurophysiological theory might
not be able to rule out a case of inverted spectra[2].
Such theories then would not be able to say in what the instantiation of a
particular color quale consists.

The first section comprises issues that pertain to
questions such as whether cognitive function is localized, the modularity of
mind, and introspection. In Ch.1, Valerie Hardcastle and Matthew Stewart argue
against localizationism, namely the view that brain and cognitive functions are
localized. They site evidence to the effect that even a system as simple and
biologically basic as oculomotor control involves contributions from units
dispersed widely across the cortex. And they point out I think correctly, that
the brain’s capacity to recover function by using new areas when damage to an
area affects function holds the same implication. They conclude by arguing
against Fodor that if there are any modules in the brain there are precious
few. Neuroscientific research, they argue, shows that back projections for
instance, are to be found everywhere, which means that there cannot be
encapsulated modules.

In Ch.2, Thomson, Lutz and Cosmelli argue that
introspection can and should play an important role in neuroscience. They
propose neurophenomenology as a neuroscientific research program whose aim is
to make progress on the issues associated with the notorious explanatory gap,
namely our lack of understanding as to how consciousness arises from physical
processes in the brain. Unfortunately, I cannot give a detailed account here of
the main ideas of neurophenomenology. The authors however, draw a useful
distinction between the explanatory gap as a scientific problem and as a
philosophical problem. They claim that the former persists because experimental
investigations of the neural correlates of consciousness usually focus on one
or another particular feature of the experience and accordingly try both to
control as much as possible any variability in the content of subjective
experience and to minimize reliance on the subjects’ verbal reports. This
approach then seems too limited for investigating the labile and self-affecting
character of conscious processes. They plausibly suggest that the
experimentalist should try to investigate these sorts of complex subjective
neurodynamical processes by making extensive use of the first-person
perspective. But what about the traditional philosophical problem? The authors
say that even such phenomenological investigations aim not to close this gap
(namely, how is it that the physical/objective gives rise to subjective
experience) but rather to bridge it by establishing reciprocal constraints.
Well, the discovery of reciprocal constraints is progress alright but the
traditional philosophical problem remains unresolved and it is hard to see how
it can be resolved. The section closes by Victoria McGeer’s interesting
suggestion that we should treat self-reports as expression of the underlying
and affective systems rather than thinking that that their job is to describe
what’s going on in peoples’ heads.

In Section 2, Chris Eliasmith suggests that in order
to understand how the mind works we should move beyond past approaches of
understanding the mind such as symbolism and connectionism. His approach
unifies representational and dynamical descriptions of the mind. Encoding and
decoding relations are used to define representation. Eliasmith argues rather
convincingly that we need to improve on connectionism by incorporating to its
bottom-up approach to cognitive modeling top-down constraints on the designs
and analysis of its models. In Ch.5, Rick Grush attempts to bridge the gap
between cognitive neuroscience and a Kantian conception of the mind. He argues
to the effect that various results and theories in cognitive neuroscience are beginning
to suggest that time is an interpretive elaboration supplied by the mind/brain
and not contents merely received from without. In Ch. 6, Sean Kelly addresses
the puzzle of temporal experience and suggests that there is a
conceptual/phenomenological distinction to be made between the pure visual
experience of motion and the visual experience of an object as moving, and that
neuroscientists have tended to conflate the two. In response, he recommends
that we look for a certain kind of short-term visual storage that could serve
as the basis for the experience of a single, unified, temporally extended
object moving through space. 

In Ch.7, Pierre Jacob starts from the Milner
and Goodale (1995) studies on visual-form agnosia[3]
and their very interesting hypothesis that we have two complementary visual
systems, namely vision for perception and vision for action. He uses
psychophysical experiments in humans based on visual illusions in order to make
suggestions about how visuomotor representation and conscious visual perception
relate. In the following Chapter, Pete Mandik focuses on Jacob’s claim that
spatial perception and motor output are interdependent and attempts to
characterize this interdependence. He proposes a view which favors the positing
of mental representations mediating between perception and action. Ch. 9 &
10 focus on color vision. Paul Churchland, once one of the most well known
eliminativists about consciousness talks now about consciousness as a real
phenomenon in need of scientific explanation. He presents a nicely illustrated
case where by exploiting shifts in experienced color due to tiredness and
habituation, experiences of color can be brought about not only where the
colors do not exist in nature but also where they could not exist (according to
a long-held color theory — Hurvich, 1981). Churchland admits that it remains
an a priori possibility that our colour qualia may vary independently of
the physical realities of the Hurvich network but warns us not to confuse a
priori issues with closely related empirical issues. In Ch.10, Zoltan Jakab
argues that color experience arises from processing that distorts the stimulus
features that are its canonical causes in numerous ways, thereby in large
constructing our world of perceived color. Our perceived color similarity, says
Jakab, is a systematic misrepresentation of the corresponding stimuli. So are
colors physical properties after all? According to Jakab, ‘colors are physical
properties of environmental objects, but perceived colors (i.e. appearances…)
are products of the mind’ (p. 372). 

The last section of the book focuses on the
problem of consciousness. In Ch. 11, Jesse Prinz observes correctly that there
are other problems in consciousness studies besides the hard problem and that whereas
some are not tractable none is easy. But he simply sidesteps the hard problem.
Prinz argues that consciousness arises when mechanisms of attention allow
intermediate-level perceptual systems to access working memory. He suggests
that consciousness has a uniform material basis but although he may be right
that finding the neurocomputational basis of consciousness would be worthy of
an invitation to Stockholm the traditional mind-body problem which poses the
most difficult problems in the science of the mind will remain unresolved:
however useful such discovery may indeed be it doesn’t admittedly give any
inside into what explains the occurrence of our subjective feels, namely to why
physical states are accompanied by experience at all.

In the concluding chapter, Andrew Brook does a
good job in dividing the theoretical landscape of views on consciousness and he
attempts to confront Chalmers’ conceivability argument and the spectrum
inversion thought experiment to the effect that consciousness is amenable to a
physicalist explanation. Brook rejects views such as McGinn’s (2004), namely
that we are cognitively closed with respect to the problem of consciousness
i.e. we cannot solve it even in principle, simply by saying that such views are
‘wildly premature’ (p. 415). The discussion on the conceivability argument is
not satisfying either. In a less than two pages long discussion of the
argument, Brook simply assumes that ‘once representations are in place, there
ain’t nothin’ left over to be left out’ (p.416). Well, whether there is
anything left out we don’t know, it remains to be discussed. What we do know is
that the explanatory gap is closely related with the conceivability argument.
And whenever there is conceivability there is always a need for an explanation,
which is always called for whenever it is conceivable that things could have
gone the other way. One for instance, wishes to know why is it that H2O should boil at 212°F at sea level or why it expands when freezing.
There is a need for an explanation here since it is conceivable that it may not
happen or that it could have gone the other way: It could have been that water
doesn’t boil at 212°F at sea level and it
could contract rather than expand when freezing. The explanation in this case
has to do with the very constitution of water and not merely with relations or
correlations between entities. After all physical and functional facts are
filled in the explanatory gap disappears (moreover one can predict that a
liquid with the same constitution would behave accordingly — the higher order
properties of this liquid can be deduced by its constituents). But this
situation is quite unlike the case of consciousness. Even if all physical and
functional facts are filled in it appears that the idea of a zombie is still
coherent and this is because such explanations do not yield an understanding of
the constitution of consciousness.