Consciousness

may explain why we have these two different visual systems.
The result is that a great deal of what we do is done quickly
and accurately, and independently of what we consciously
perceive.

Can we now conclude that one of the streams is con-
scious while the other is a zombie, as Ramachandran and
Blakeslee (1998) claim? Although Milner and Goodale were
initially cautious about making this distinction, they subse-
quently took the same view: while the ventral stream pro-
vides ‘a conscious representation of the world’ (Goodale,
2007, p. 626) and ‘the sole route to phenomenal visual
consciousness’ (Milner, 2008, p. 177), ‘the visual products
of dorsal stream processing are not available to conscious
awareness – [. . .] they exist only as evanescent raw mate-
rials to provide the unconscious moment-to-moment sen-
sory calibration of our movements’ (Milner, 2012). Similarly, Nancy Kanwisher
(2001) suggests that ‘the neural correlates of the contents of visual awareness
are represented in the ventral pathway, whereas the neural correlates of more
general-purpose content-independent processes [. . .] are found primarily in the
dorsal pathway’ (p. 98). Note that these formulations make several assump-
tions: that consciousness has contents, that the contents are representations,
and that there is a difference between some areas or processes that are con-
scious and those that are not. We have begun to question every one of these
assumptions.

The principles of embodied, or distributed, cognition
encourage us to question the very distinction between
vision for (conscious) perception and for action. On this
view, perception always happens (to a greater or lesser
extent) through and for action. Hurley’s (1998) book Con-
sciousness in Action and Alva Noë’s (2005) book Action
in Perception both understand conscious experience as
dependent on, or even constituted by, embodied action,
whether actual or potential. More recent experiments
with patient D.F. support this view. German psychologist
Thomas Schenk found that she can accurately grasp reflec-
tions of objects, but only when there is an actual object
there to be grasped. That is, her apparently intact vision-
for-action actually relies on haptic (touch-based) feedback
from objects in the environment to scale her reaching
actions and grip size to those objects. This converges with
findings from visually normal participants in Schenk’s experiments and oth-
ers (e.g. Bingham, Coats, and Mon-Williams, 2007): for example, if the size of
a grasped object is changed while a reflected visual object stays the same
size, people are very well able to recalibrate their reach actions and grip sizes
using haptic feedback.

These debates, about perception and action, consciousness and unconscious-
ness, have never been more heated than in discussion of a strange phenomenon
known as blindsight.

Perceptual

orientation

matching

Visuomotor

‘Posting’

D.F. Control

FIGURE 8.10 • Polar plots illustrating the

orientation of a hand-held card

in two tasks of orientation

discrimination, for D.F. and an age-

matched control. On the perceptual

matching task, both were required

to match the orientation of the

card with that of a slot placed

in different orientations in front

of them. On the posting task,

they were required to reach out

and insert the card into the slot.

The correct orientation has been

normalised to the vertical (Milner

and Goodale, 1995).

Perceptually different

Physically same

Perceptually same

Physically different

FIGURE 8.11 • Diagram showing the ‘Titchener

circles’ illusion. In the top figure,

the two central discs are of the

same actual size, but appear

different; in the bottom figure,

the disc surrounded by an

annulus of large circles has been

made somewhat larger in size so

as to appear approximately equal

in size to the other central disc

(Milner and Goodale, 1995).