Brain in a vat, thoughts from embodiment.

The philosophical example goes;

If you put a brain in a vat and connect all the inputs necessary, would the brain be fooled that it actually wasn’t a brain in a vat, but a normal brain in a normal world?

All kinds of fun philosophical issues follow. Embodiment however, could firstly argue that since it is only a brain, it could not function at all because brain is body -there is no separating. The argument would then be that all the inputs is a misleading assumption behind the question. We would obviously not have all the inputs (bar for a moment that input/output type stuff is difficult to maintain under this perspective). However, for argument’s sake, let’s accept both the word input (and all its assumptions) as well as that a brain is connected in such a way that it may as well have been a part of a body and in a world. This does however take the fun out of the question since we are basically saying that it already is fooled to be a normal brain in a normal world. The curiosity however is that, from an embodied perspective, you are more or less forced to clarify the example to the extent where it isn’t an exciting question.


It is only really exciting to begin with because growing up we are taught that the brain is separate from the body, we may even be taught the the mind is separate from the brain -so the example feeds off of common sensical, traditional, dualism -brain is something different from body, and/or -mind is different from brain. Embodiment doesn’t allow this separation, which forces a restatement of the question -in a way that answers it implicitly. Neat, right!?

Ontological and epistemological definitions of affordances (as per previous post). (2/5)

The ontology of affordances (based on my previous post), then, defines affordances as physical properties inherent to the object/agent that may be acted upon only by other compatible objects/agents.

The specific affordance to-be-explained is derived from the specific physical properties with the object/agent and they are necessarily constrained/restricted by both the body of the object/agent and the physical properties of the environment. For example, our legs are able to move in some ways but not others, we are restricted in the movement of our legs by a) the physical properties of the make-up of our leg (the knee puts the most obvious restriction) and b) the physical properties of the make-up of the environment in which it is currently in (living in a gas allows relatively free movement of the leg -compared to living in water, for example, but gravity will “restrict” -probably more accurate to say control here- us in one sense, whereas, say direct physical constraint -someone holding your feet down- restricts us in another sense). Here, thus, it should be obvious to see that Physics, Chemistry and Biology are necessarily implicated as the basis upon which determines what is a restriction and what is a constraint.

The most important part; defining affordances minimalistically ontologically, avoids many of the ontological consequences faced when defining affordances as relationships (by leading to some form of idealism), although, as I will argue, affordances within objects/agents depend on each other epistemologically. I believe this is also the consequence by using the definitions of realisation and actualisation for the epistemological reliance of affordances.

Realisation and actualisation is the, how we come in contact with, how we gain knowledge of, what affordances do. What we do. How we do them. Since (if I’ve got this right) (radical) embodied cognitive science posits that, consciousness, cognition, memory (and many other representationalist terms) are not properties of the brain -but things we do– then I think it appropriate to the central ideas of rECS. I use “radical Embodied Cognitive Science” instead of embodied cognition due to the well argued taxonomy that Chemero presents in his book. Radical does not get a capitol letter however, to make the point that the theory is not radical in and of itself (like Chemero argues) but is merely a distinction from Embodied Cognition. This distinction seems to me necessary because of Chemero’s arguments.

Affordances rely on the mechanisms of realisation and actualisation.

Realisation is to do with what Wilson & Golonka discusses on their blog, that which is perceptible necessarily contains information, if I understood it correctly (energy array etc., their definition is brilliant and me rewriting it would not do it justice, it also serves my purposes well).

Actualisation is to do with the coupling, when we act on the perceived affordances.

Objects exist when we are not there to perceive them; realisation, but not actualisation. It should be obvious that once we have perceived an object and some of its affordances, the realisation is retained by virtue of the compatible affordances of both the affordances of the object and the affordances of the agent. As of yet however, I believe the affordances of the agent are necessary (we can realise the affordance we need in an object in order to actualise the affordance perceived of our body).

Failing when doing; I don’t see this as an issue, why would it be? This type of reasoning belongs to Evil Philosopher type arguments, in that, because we “get it wrong” then it somehow reflects on the actual mechanism of perception and/or action. I do not believe this is so. Direct perception gives us the information that we are able to perceive and act upon, but in my mind there has to be a perturbance or something not yet perceived to disrupt our ability to actually carry out, actualise, affordances. And, objections like that seem to assume that we are perfect beings. As I see it, our sensory modalities are limited, we are not the pinnacle of “creation”, we will get things “wrong” -but like all other philosophy of mind objections it doesn’t have a bearing on ontology, solely epistemology. We evolved to perceive to survive and reproduce, not to gain a perfect perception of the environment. And that’s ok. Doesn’t have a bearing on affordances since they are defined ontologically without the requirement of being accurate.

Ontology; Affordances, thus, are not defined as realisation and actualisation, but as (simplified here) physical properties reflected by Wilson & Golonka’s definition of information.
Epistemology; Affordances rely on the mechanisms of realisation and actualisation. All three are necessarily constrained by physical properties of themselves individually as well as each other. Objects and agents can be realisable but not actualisable, both in presence and in absence of each other; actualisable only in presence of each other.

Reading list for Embodied Cognition

[Edit 19/7 2013: I am getting quite a lot of traffic to this post, so I thought I’d point you to my thesis reference list instead as this post is a bit messy and incomplete.]

I am collecting my readings on Mendeley, in a group called Embodied Cognition (should be the only one so far..). I figured I needed somewhere to collect all readings I go through, however, since I have yet to find a way to add books to the group, and thought I may as well put them here in case anyone else has any utility for it. Will update the post as I’ve read articles/books etc. Also, please feel free to comment with additional readings that you’ve found valuable in understanding EC.

(in the order I read them)

Added on 14/3 2013
Larry Shapiro – The embodied cognition research program (article)
Louise Barrett – Beyond the brain (book)
Alva Noë – Out of our heads (book)
Wilson & Golonka’s blog (all entries)
Wilson & Golonka – Embodied cognition is not what you think it is (article)
Tim van Gelder – What might cognition be if not computation (article)
Montagne, Laurent, Durey & Bootsma – Movement reversals in ball catching (article)
Pfeifer & Bongard – How the body shapes the way we think (book)
Gerd Gigerenzer – Rationality for mortals (book)

Added on 16/3 2013
Haller & Krauss – Misinterpretations of significance (article)
Ziliak & McCloskey – The cult of statistical significance (article)

Added on 19/3 2013
Anthony Chemero – Radical embodied cognitive science (book)

Added on 8/4 2013
Semin & Smith – Embodied grounding (book)
Gibson – The ecological approach to visual perception (book)

In progress;
Pan, Bingham & Bingham – Embodied memory: Effective and stable perception… (article)
Holmes & Heath – Goal-directed grasping: The dimensional properties of an object… (article)
Mann, Dicks, Cañal-Bruland & van der Kamp – Neurophysiological studies may provide… (article)
Gray, Sims, Fu & Schoelles – The soft constraints hypothesis: A rational analysis approach… (article)
Hayhoe & Ballard – Eye movements in natural behavior (article)
Hayhoe – Vision using routines: A functional account of vision (article)

Russel and Norvig 1995   (article)
Pfeifer and Scheier 1991   (article)
Pfeifer and Scheier 1999   (article)
O’Regan and Noë 2001   (article)
McFarland and Bisser 1993   (article)
Monteliore and Noble 1989   (article)
Thompson 1996   (article)
Bird and Layzell 2002   (article)
Schelling 1969   (article)
Epstein and Axtell 1996 (article)
Bo{r/v}et & Pfeifer 2005   (article)
Bartlett 1932   (article)
Ashby 1956   (article)
Freeman 1991   (article)
Clancy 1997   (article)
Neath and Suprenant 2003 (article)
Dewey, ?. (1896). ?   (article)
Titchener, ?. (1895). ?   (article)
Kahneman and Tversky, 1996   (article)
Gilovich, Griffin and Kahneman, 2002,   (article)
Tversky and Kahneman, 1986   (article)
Wason and Johnson-Laird 1972   (article)
Thriver 2002   (article)
Cosmides 1989   (article)
Wundt 1912, 1973   (article)
Shaffer and McBeath, 2002   (article)
Fillenbaum, 1977   (article)
Sweetser 1990   (article)
Sher and McKenzie, 2006   (article)
Shaffer, 2004   (article)
Fodor and Pylyshyn   (article)
Chomsky   (article)
Kuhn 1962   (article)
Feyerabend 1963, 1965   (article)
Titchener, 1895   (article)
Titchener and Lange   (article)
Dewey, 1896   (article)
Fodor, 1981   (article)
Gibson, 1979   (article)
Barwise and Perry, 1981, 1983   (article)
Brooks (1991, 1999)   (article)
Clark (2001)   (article)
Thelen and Smith, 1994   (article)
Thelen 1995   (article)
Kirsh and Maglio 1994   (article)
Clark 1997   (article)
Adams and Aizawa (2008)   (article)
Beer 2003   (article)
van Rooii, Bongers & Haselages (2002)   (article)
Markman and Dietrich 2000a   (article)
Markman and Dietrich 2000b   (article)
Dietrich and Markman 2003   (article)
Grush, 1997,   (article)
Grush, 2004,   (article)
Turvey, 1981   (article)
Michaels and Carello, 1981,   (article)
Heft 1989   (article)
Heft 2001   (article)
Turvey 1992   (article)
Michaels 2000   (article)
Read 1996   (article)
Dennett 1998   (article)
Cosmelli, Lachaux and Thompson 2007   (article)
Thompson and Varela 2001   (article)
Bickle, 2003,   (article)
Churchland, Neurophilosophy, (book)
Thelen and Smith 1994 (article)
Pfeifer and Scheier 1999 (article)
Edelman 1987  (article)
Searle 1980  (article)
Schwanen and Plugel 1991  (article)
Barsalou 1999  (article)
Glenberg 1997 (article)
van Orden, Holden and Turvey 2005 (article)
Montessori 1967 (article)

Only read articles available on Mendeley. Books available on loan, from me, if you fancy a visit to Lund, Sweden, otherwise they’re available in bookstores online.

Exploding boxes and affordances

If I have understood this correctly, there is an issue with two boxes being identical but only one having a specific affordance -in this case being “pick-up-able” or “touch-able”. Both boxes, to us, are perceived to have the affordance but this is not the case. Therefore, we can’t rely on perception to decide which affordances are available to us.

I am not sure this makes sense. The central point of the masses of words beneath is, does it not assume us to be constant naive explorers of the world? And is this a fair assumption? (Now you don’t have to read all the details, you’re welcome.)

It strikes me however that, the consequence of trying to explore a possible affordance, gives information on which affordances are available to us. Because we are explorers/seekers, we navigate our environment and find out what is, and isn’t. We rely on perception to do so. Is ice walk-on-able? Sometimes. How do we go about finding out? We poke the ice infront of us with a stick, indirectly finding out if the ice is walk-on-able. So we use secondary mechanisms to find out if something has an affordance, if it is not directly perceptible to us, but this relies on that we have seen evidence of it not being able to be relied on by direct perception of our environment. The assumption in the issue may be that we are constant naive explorers, which we aren’t.

Is it not true that both boxes still persist in holding the affordance, only that, in one case we end up exploded and the other we don’t? A big enough box provides the affordance of sitting, regardless of what the consequence of sitting on it is?

It seems to me that this is a classic case of philosophy of mind issues -where we can’t rely on direct perception to perceive what is “actually out there”. For all its worth, if we are to be assumed to be naive explorers, I posit that we will always sit on the exploding chair, touch the exploding box, walk on what is perceived as a solid and rigid surface and so on -because the visual properties, surface and rigidity and so on, lend us to perceive the existence of such an affordance.

So it depends on reliance then. If we rely solely on direct perception (making us constant naive explorers) then objects retain their perceived affordances, regardless of which affordances actually are available to us. If two objects are identical in all perceptible ways, then we can not rely on direct perception to know which affordances are available to us. However. We are not constant naive explorers. We see others interact with objects, get burned on stoves that look like they are turned off. So what do we do? We quickly touch the stove, or look at the knob or hold our hand above the stove.

As for objects, without the involvement of interaction with other things, it is my view that they retain, many, finite number of affordances. These affordances will be available to other objects, but not all affordances to all other objects. Which ones are, will only be realised when another thing, with its own affordances, interact with it, perceive it. With this said, affordances are not the actual relationship between things, it is the fit between one objects affordance and another objects affordance. If they are compatible, then to the specific object, the other object has the specific affordance. They form a relationship, but importantly, both have their affordances retained in the non-presence of each other, if, the affordances are available when interacting with each other.

Example, the structure of DNA contains four characters, only A can bind to B but not C and D and vice versa, in A and Bs absence of each other they both retain the affordance of being able to bind to one other and they both do not have the affordance of binding to C and D. In A and Bs presence of each other they can bind to each other, and thereby realise both of their affordances. In essence, they retain their separate affordances in absence because they can be realised in presence.

No wonder social relationships are so damn important evolutionarily, seeing others being blown up by boxes would surely rule out to me going near any box even similar in visual makeup to the one that exploded!

Ugh.. always feel I lack knowledge when I finish a thought. Either way, these are thoughts associated to and