Consciousness

EMBODIED COGNITION

The machines described so far are all disembodied, con-
fined inside boxes and interacting with the world only
through humans. When first put to work controlling
robots, most could carry out only a few simple, well-
specified tasks in highly controlled environments, such
as in special block worlds in which they had to avoid or
move the blocks. This approach seemed sensible at the
time because it was based on an implicit model of mind
that was similarly disembodied. It assumed the need
for accurate representations of the world, manipulated
by rules, without the messiness of arms, legs, and real
physical problems. We might contrast this with a child
learning to walk. She is not taught the rules of walking;
she just gets up, falls over, tries again, bumps into the coffee table, and eventually
walks. By the same token, a child learning to talk is not taught the rules either; in
the early days she pieces together fragments of sounds she hears and gestures
she sees, parses words wrong, and eventually makes herself understood.

The connectionist approach is far more realistic than GOFAI, but still leaves out
something important. Perhaps it matters that the child has wobbly legs, that the
ground is not flat, and that there are real obstacles in the way; maybe it matters
that she has the vocal cords she does and that her parents’ gestures are con-
strained by their limbs.

As we saw in Chapters 5 and 6, embodied or enactive or 4E cognition are names
for the general idea that mind can be created only by interacting in real time
with a real environment  – the idea, drawing on the phenomenology of Mer-
leau-Ponty, ‘that cognition is not the representation of a pregiven world by a
pregiven mind but is rather the enactment of a world and a mind’ (Varela,
Thompson, and Rosch, 1991, p. 9). Andy Clark (1997) wants to put brain, body,
and world together again  – both causally and computationally speaking. ‘For-
tunately for us’, he says, ‘human minds are not old-fashioned CPUs trapped in
immutable and increasingly feeble corporeal shells. Instead, they are the surpris-
ingly plastic minds of profoundly embodied agents’ (2008, p. 43). What he means
by ‘profoundly embodied’ is that every aspect of our mental functioning depends
on our intimate connection with the world we live in. Our ‘supersized’ minds and
our powers of perception, learning, imagination, thinking, and language are all
created by brains interacting with bodies and their environments, both physical
and social.

On this view the real world is far from being a messy complication we can do
without; rather, it provides the very constraints and feedback that make percep-
tion, intelligence, and consciousness possible. Human intelligence is not just
‘recognition intelligence’: it is about using understanding to make autonomous
real-time decisions. Creating machines this way means constructing real, physi-
cal, autonomous agents that move about in the real messy world, working from
the bottom up rather than the top down. There is no point in a driverless car
recognising a collection of pixels as a white van slowing down quickly unless it

‘human minds are not

old-fashioned CPUs

trapped in immutable

and increasingly feeble

corporeal shells’

(Clark, 2008, p. 43)

Input layer

Hidden layer

(there may be

several hidden

layers)

Output layer

FIGURE 12.3 • This artificial neural network (ANN)

has just three layers of units:

the input layer, the output layer,

and a hidden layer in between.

During training, the weights on

the connections between the units

are adjusted until the network

provides the correct output. Such

a network can learn to recognise

faces, to produce sounds in

response to written text, and

many other tasks, depending on

what is connected to the input and

output units.