We can see this difference with respect to vision. The visual system is well under-
stood, with something like ten separate parallel pathways from the eyes to different
areas of the brain. About 85% of cells take the major route through the lateral genic-
ulate nucleus (LGN) in the thalamus to primary visual cortex (V1) in the occipital lobe
and then, with increasing numbers of diverging pathways, to V2–5, MT, and many
other areas with varied functions. The rest go via the superior colliculus in the thal-
amus to various other cortical and subcortical areas. It has long been known that
damage to the eyes, thalamus, and V1 produce blindness, so these early parts are
necessary for conscious vision (though see Chapter 8 on blindsight), but may not be
sufficient. Conversely, patients with activity in V1 but no connections to higher areas
may report no visual experience, and this applies to other senses as well.
Patients in a ‘persistent vegetative state’ (PVS), are described as awake without
awareness, somewhere between ‘coma’ (in which a patient has closed eyes and is
unresponsive to any stimulation) and a ‘minimally conscious state’ (in which there
is some responsiveness or inconsistent signs of consciousness). Belgian neurolo-
gist Steven Laureys (2005) tested several patients with what would normally be
painful electrical stimulation. This produced activity in the brainstem, thalamus,
and primary somatosensory cortex, but not higher up the pain matrix in parietal
lobes and anterior cingulate cortex. Similarly, with loud sounds primary auditory
cortex is activated, and with flashing light primary visual cortex, but in neither
case is there activity in higher association areas (Di et al., 2008). Laureys concluded
that PVS is due to disconnection between the primary sensory areas and the fron-
to-parietal network and that ‘neural activity in primary cortices is necessary but
not sufficient for awareness’ (2005, p. 558).
Other evidence comes from studies in which cells in V1 are shown to adapt to
invisible stimuli, and from the fact that V1 is suppressed during dreaming sleep
even though vivid visual dreams are reported. Studies using single-cell recording
in monkeys show that cells in V1 cannot tell the difference between movement
caused by eye movements and that caused by movement in the scene, whereas
cells higher in the visual hierarchy can – as they must if you are not to think the
world has moved every time you move your eyes. From this and other evidence,
Koch concludes that ‘While V1 is necessary for normal seeing – as are the eyes – V1
neurons do not contribute to phenomenal experience’ (2004, p. 105).
This might seem a curious conclusion: how can V1 be both necessary for normal
seeing and not contribute to phenomenal experience? The underlying assump-
tion here seems to be that most of what goes on in the nervous system is uncon-
scious, and ‘only a fraction of all sensory data pass into awareness’ (Koch, 2004,
p. 170). All the unconscious stuff is a necessary precursor of conscious experi-
ence but isn’t directly responsible for consciousness. This is clearly a Cartesian
materialist description – relying on that magic difference between conscious and
non-conscious processes – and it leaves untouched the question of what it might
mean for the physical activity of neurons to ‘pass into awareness’.
The default position in NCC research is simply to ignore these questions. For exam-
ple, a recent meta-analysis of whole-brain fMRI studies contrasting conscious
with subliminal visual processing concluded that the NCCs of visual conscious-
ness comprise ‘a subcorticalextrastriate-fronto-parietal network encompassing
inferior and middle occipital gyrus; fusiform gyrus; inferior temporal gyrus; cau-
date nucleus; anterior insula; inferior, middle, and superior frontal gyrus as well
