Embodied Cognition and Consciousness: A Bestiary
A. Claims
1. Our mental models or maps are more partial than we suppose. The
partialness of our models is supplemented—and rendered hard to
notice—by our interactions with the environment (e.g. visual saccades;
keeping written records). It is possible to treat elements in the
environment as supplementary parts of our representational models:
i.e. perhaps our minds are extended outside our skulls. The most
extreme version of this view is that there are no mental models or
maps at all.
2. Our mental models or maps are more action-oriented than we
suppose. For example, instead of creating an internal map of some
geographical space, we represent landmarks and tie these
representations to potential actions at these landmarks (e.g. ‘at this
spot I can go left to get to the library, or right to enter the kitchen’). At
one extreme, there are radically enactive views of cognition which
equate perceptual content with sensory-motor knowledge: seeing is
more like touching than depicting.
3. Some cognitive work is off-loaded onto our
environment. What counts as mental computation
extends outside the skull by utilizing environmental
structures (e.g. playing Tetris or Scrabble, long
multiplication, nautical slide rules). This has been
called epistemic action. At the extreme, certain ways
of thinking are not possible without embedding the
organism into a particular sort of environment:
cognition must, sometimes or always, be understood
as taking place within a unit that includes both the
organism and its environment.
4. Our mental models or maps are non-representational: they are not
‘mirrors of nature.’ This view can have various sources:
a) ‘Neurophenomenology.’ The properties and categories our minds
encode are not ‘read off’ from the external environment but are
constructed through an interaction between the organism and its
environment (e.g. colour properties). Thus our world is partially or
completely a matter of ‘codependent arising’ rather than of
representation. At the extreme end of this spectrum is the view that
there is in principle no such thing as a representation of the world as it
really is and hence, perhaps, there is no mind-independent reality.
b) Neo-pragmatism. Representation does not occur in virtue of inner
mental tokens standing in some intentional relation to the world: rather,
representation is a matter of complex interactions between organisms
and/or their social or material environments. Mentality is a matter of
complex overt behaviour in a context, not inner coding.
c) Anti-cognitivism. All or much of the ‘processing’ by sensory and
motor systems does not involve the computational manipulation of
representations, but instead consists in some sort of complex,
mechanical interaction with the environment—e.g. the forces acting on
a set of legs. And/or our cognitive architecture is—fundamentally, or
thoroughly—‘associative’ or ‘pattern completing.’
5. Our cognitive capacities and predispositions are strongly influenced by
our embodiment (e.g. our relative size, our visual orientation and other
facts about our sensory modalities, our bipedalism, our emotional
makeup, and so on). The notion of ‘embodiment’ may include cultural
as well as biological factors, such as facts about the natural languages
we speak or theories we have inherited. At the extreme, our cognitive
capacities are unavoidably fully determined by our embodiment.
B. Evidence
1. Robotics: Distributed, autonomous and peripheral
sensory and motor systems, rather than the activities of
a ‘central processing unit,’ can ground cognition.
2. Developmental psychology: Brain, body, and local
environment interact in complex ways to determine the
early cognitive success of infants (e.g. learning to walk).
3. Evolutionary psychology: The evolutionary function of
the brain is to be the real-time control system of the
body; it is not ‘designed’ to be a tool for describing and
explaining reality.
4. Psychology of perception: Phenomena such as
change blindness and inattentional blindness show that
perception is attention-dependent and active—i.e. it
does not consist in the passive encoding of patterns
from the environment.
Dr. Andrew Bailey
Philosophy Department
University of Guelph
Guelph, Ontario
N1G 2W1, Canada
[email protected]
5. Dynamical systems theory: Dynamical systems theory—which does
not distinguish between the organism and its environment—is an
explanatory framework for psychology that competes with cognitivism.
C. Consciousness
1. The extended mind paradigm applies to cognition
but not to phenomenal consciousness. It claims
only that non-occurrent beliefs and memories are
outside the skull.
2. The enactive mind paradigm calls into question
the status of consciousness. It tends to treat
conscious perceptual content as non-imagistic,
and as either a motor skill or knowledge of
possibilities for action.
3. The co-dependent arising approach suggests a phenomenological
approach that denies any deep distinction or opposition between the
physical and the phenomenal: the world is not separate from us,
though it is prior to us.
4. Anti-representationalism in general tends mostly to be directed at the
units of mental content, but scepticism about inner mental tokens has
implications for consciousness.
5. The notion of epistemic action is not particularly relevant to
phenomenal consciousness.
6. The embodiment paradigm has little to say on the question of
consciousness: its implications are for concepts not phenomenality.
Sample References
• Beer, R.D. The dynamics of active categorical perception in an evolved
model agent. Adaptive Behavior 11 (2003): 209–243.
• Clark, A. Being There. MIT Press, 1997.
• Clark, A., & Chalmers, D.J. The extended mind. Analysis 58 (1998):10–
• Brooks, R.A., Intelligence without representation. Artificial Intelligence
47 (1991): 139–159.
• Kirsh, D. and Maglio, P. On distinguishing epistemic from pragmatic
action. Cognitive Science 18 (1994): 513–549.
• Noë, A. Action in Perception. MIT Press, 2005.
• Simons, D.J., & Chabris, C.F. Gorillas in our midst: Sustained
inattentional blindness for dynamic events. Perception 28 (1999): 1059–
• Simons D.J., and Rensink R.A. Change blindness: Past, present, and
future. Trends in Cognitive Sciences 9 (2005): 16–20.
• Thelen, E., & Smith, L. B. (1994). A dynamic systems approach to the
development of cognition and action. MIT Press, 1994.
• Varela, Thompson and Rosch. The Embodied Mind. MIT Press, 1991.

Slide 1