Flynn’s Cat – Part I: An Exploration of Embodiment

© 2017 Simon’s Cat Limited

It is fair to say that a wide range of topics invoke discussion under the banner of embodiment, too many to discuss in a few short posts. There are however, a few themes that while not exhaustive, are prominent in the literature:

Does the body and world form part of my cognitive processing as opposed to merely causing it?

Does my body determine in some way, how I understand my world?

Can my cognition be explained solely by my interaction with the world, without appealing to representations or computational processes?

Beginning with the question of whether cognition, or at least parts of it, extend beyond the brain, a common problem from systems analysis arises, namely, where is the boundary of the system, and how might its parts be either decomposed or clustered together to aid investigation, or are questions of boundary and decomposition, themselves part of the problem.

These issues are not unique to cognitive science.  They arise when attempting to analyse any system through decomposition. Attributes bleed into one another with no clear boundaries or hierarchies, and whether something is viewed as a component of a system, or causal to a state within the system, depends on where the boundary of that system is drawn, and whether that component is inside or outside the boundary. Where the boundary is drawn is dependent on third-party observation and what the requirements of that third-party are.  In other words, in this case, where we draw the boundary depends on what we mean by cognition. If the defining features of cognition are the information processing role inherent within it, then there is no reason that this cannot extend into the body and into the world.

Much of the concern, it would seem, that this approach throws up for traditional cognitive scientists, arises from the heterogeneity of a brain-body-world system and whether it would ever be amenable to scrutiny given that the external could consist of anything?  In other words, would it ever be possible to study a brain-body-world system? But here, Wilson’s ‘Wide Computationalism’ (1994) dispenses with this problem by seeing physical diversity as irrelevant. What is important is the information that is carried and the role that information plays in the larger computational system. Computationalism doesn’t specify the location of the computations, and the computational description of a process abstracts away from the medium it is found in. 

The question then perhaps is whether this gets us anywhere? The mind could indeed extend, but does it, and would any widening of the boundary be criticised by traditionalists as merely playing with words? To differentiate itself, there must be something different in the interaction between an organism and an external artefact that truly mark it as a cognitive system rather than an arbitrarily drawn boundary.

As I type this, I am both writing and thinking at the same time. My laptop provides the medium through which a neural/typing/reading coupling is unfolding to explore a topic in ways I might not otherwise come upon. The cat sitting beside me, spots my cursor moving on the screen. She locks onto its every movement as I drag it across the screen. Her eyes move. Her head moves as she follows it. Eventually her paw reaches out to the laptop screen. For these few moments, is the cursor on my screen, part of the cat’s mind?  Are the words on the laptop part of mine? Are the nature of our interactions with the external artefacts the result of an embedded algorithm inside us  somewhere sandwiched between perceiving and movement, or are those artefacts, at least temporarily, part of our minds?

It can be argued in both cases that the external artefacts are non-cognitive in nature. Neither the cursor nor the laptop would typically be viewed in themselves as cognisers. But must the constituents of cognitive processes themselves be cognitive? If the answer is yes, then the boundary of the mind is around the brain. But can cognition include non-neural constituents and still make sense. A sentence typed by me on my laptop screen forms part of a cognitive process that produces a neural state in me, which in turn produces another interaction with my laptop. If a neural state correlates with something in the real world, then if the part of cognition that is done outside my head has parity with the process were it done inside my head, then it could be argued that it is part of the cognitive process. In this case the movements and other attributes of the cursor are part of the cat’s mind, the words on the laptop are part of my mind and the boundary must widen to include them.

But does it really matter where the boundary is? Does it get us anywhere new? In this regard, the debate on whether cognition extends beyond the brain into the world or whether it is cranially bound but the brain and parts of the world together produce cognition is in danger of descending into a mere argument around vocabulary. In addition, the arguments on whether there needs to be a boundary between neural and non-neural parts of cognition also perhaps misses the larger point that cognition seems to emerge from the interactions and cannot be understood through decomposition into constituent parts. 

So again I need to ask, is there something different in the interaction between an organism and an external artefact that truly mark it as a cognitive system rather than an arbitrarily drawn boundary? Is there a difference in what the cat and the cursor are doing, and what I and the laptop are doing?  A cat reacting to a cursor on a screen, at least as far as output is concerned, can be explained equally well by effect or by extended mind. But what about me and the laptop? From brain to laptop, back again to brain, and so on, as I examine my thoughts. The brain alone has difficulty doing this, so instead it produces words on a screen in order to enhance its ability. The words on the laptop relieve my brain of the burden it would otherwise need to assume. This is not just a simple interaction between an organism and an external artefact; it is the adoption and generation of an external artefact by the brain to enhance its own abilities, and it is this that distinguishes the interaction from other external influences that I might encounter. My cognition is not merely influenced by an external artefact, I have self-generated how I perceive and interact with external artefacts to aid my own cognition. This extension with the purpose of self-augmentation is in my view, what marks extended mind as an interesting departure from traditional cognitive science.

A more interesting question perhaps is, what does the cat perceive in her interactions with the cursor? This is the subject of the next post…

Flynn's Cat - Part 2 > 

References

Wilson, R. (1994). Wide Computationalism. Mind, 103(411), pp.351-372.