Saturday, January 12, 2013

Why am I suspicious of the Ecological Approach to touch (haptic perception)?

I have never done haptic research, but it is a rapidly emerging part of Eco-Psych these days. I have seen many talks at conferences, and overall I think the work is quite interesting. Touch is understudied in psychology, as a whole, and it is good that it is getting this increased attention. That said, I have been getting increasingly uncomfortable hearing these talks at Ecological Psychology conferences, because something often seems a bit off. Something seems, to me, to contradict fundamental tenants of the ecological approach. I have never been able to convince any of these researchers to even see the problem I see, nevertheless take it seriously. They don’t have any particular reason to cares what I think, and that’s alright. However, when preparing the previous post on one of the foundational papers in the discipline TSRM, 1981, I found a passage in which Turvey, Shaw, Reed, and Mace explain the problem (30 years ago). Most people in the field seem to care what those people think (and what they thought before), so maybe my concern will make more sense if I let them explain it.

In the following quote (TSRM, 1981, p. 242-243), they explain why it is absolutely crucial that vision be understood as reliant on optical information outside the organism….

-----Start quote------

The intentionality of visual perception can work only by explaining how organisms can “come into psychological contact” with objects with which they are not in physical or, more aptly, mechanical contact. Solving this problem of perceptual “action at a distance” is the function of Gibson’s theory of ecological information for perception. As Gibson (1975, p. 310) once wrote in reply to a critic:

"When Boynton (1975, pp. 300--l) asserts that 'we are not in visual contact with objects, or edges, facets, faces or textures, we are in contact only with photons' this assertion is loaded with epistemology. It is a strictly philosophical conclusion. I disagree with it. There is a misunderstanding of the metaphor of visual contact, one that goes back to Johannes Muller, and it is one that I discussed repeatedly in The Senses Considered as Perceptual Systems (1966). It leads to the doctrine that all we can ever see (or at least all we can ever see directly) is light.”

The philosophical assumption underlying virtually all research on vision, and underlying all criticisms of Gibson, is that visual contact must be reduced to a physical or mechanical contact of the sort described above. Thus the intentionality of vision is claimed to be only apparent, and is reduced by assumption to causality of an absurdly simply sort. For centuries students of visual perception have been asserting that all that organisms ever see directly is light because (they claim) only light comes into contact with the ocular apparatus of organisms. The fact that critics of Gibson (e.g., Ullman, 1980) repeatedly ask how it is that optimal information gets “into” the organism shows that this simplistic doctrine of physical contact is still being invoked as the material basis of psychological contact.

------End Quote------

So… at least when we are talking about visual perception, Gibson, and TSRM are insistent that the information (the patterns in ambient light that specify the state of the organism's surroundings) are not in the organism. This is considered crucially important to creating a sound foundation for our discipline. It is also crucial for professional posturing, as they assert, roughly:
One of the biggest and most important differences between Ecological Psychology and the so-called Cognitive approaches, is that we are not talking locating the functional basis of perception inside the organism, and they are. When they do this, they got caught in many philosophical quagmires. Our insistence on the existence of information external to organism avoids those quagmires, and render many of the classic Problems of Perception moot.
I imagine that the equivalent version of this for haptic perception is to insist that the higher-order properties of 'things that can be felt' exist out-there, in the object itself. Sometimes the object must be moved by the organism to detect these properties (e.g., you often can't tell the difference between a functional sword and a decent copy without wielding it), but center of mass and other such properties are 'out there' in the object, and the resistance the sword gives to certain types of movements is similarly 'out there', when the one tries those movements.

However, that is not the way the field seems to be going (it has gone that way at least sometimes in the past, but that is not the direction it seems to be headed in now). At the last ICPA in particular there was a wide range of talks trying to determine where, inside the organism, the information might be. People were looking at different structures at joint points, as well as looking at the elaborate interconnections created by connective tissues across the whole body.

Because my intuitive thinking is in line with the above quotes, I am pretty suspicious when people talk about haptic information being specified inside the organism. Within the ecological approach, certainly the default assumption think that the haptic information, like the optic information, would need to be outside the organism.There might be convincing arguments to allow an exception in this case, but I have not heard them. How will we go this route without falling prey to the Problems of Perception?


  1. Isn't another component of the ecological approach the idea that the internal/external boundary is unimportant? It seems to me that mechanical transformations of the myofascial field specific to object properties (or, properties of the combined SE system) is analogous to transformations of the array specific to object properties (or...). I don't see any grounds for assuming information is external and can't think if any ecological research that relies on that assumption (in the strong sense of ruling out a priori internal sources of information).

  2. Henry,
    I am open to having my suspicions assuaged. I am not sure this is the wrong way to go... I am just suspicious. Do you interpret TSRM and Gibson on visual perception differently than I do?

  3. Surely no one could object that the myofascial tissue is involved in perceiving certain object properties, just as rods and cones are involved perceiving other object properties. But we are (I think unanimously) insistent that the information supporting vision is not in the rods and cones, nor any other part of the body. Even ignoring the specification problem, there seemed (I thought unanimously) to be the belief that allowing the information to be inside invited dualism - at the least, it invited the argument that what the organism knows is ultimately something about itself, not something about the world.

    Did we just grow beyond such concerns somehow? Was I totally mistaken about the original issue? Or did someone work it all out to show there was no problem, and I just missed a crucial paper somewhere? This research is all very good, I'm just surprised no one else seems to think there is something odd going on in how we speak about the results.

  4. Actually, you're right, there is a bit of a disanalogy here. We certainly don't sample the myofascial array as we do the optic array, for example. I do see your concern. Hmmm...

    Perhaps the resolution lies in distinguishing between the information sources and the properties they provide information about. I think the concern with dualism is with regards to the latter, not the former. The organism doesn't know about the optic array, it knows about the objects and events that structure it. Yes, the haptic array is internal, but it provides information about moments of inertia, which properly belong to the extended A-E system. Properly speaking, even though the information source is internal, what it allows the organism to know about is not.

    Even if this is right your points are worth discussing. Is there something special about the haptic system, in that rather than relying on energy arrays in the environment, we carry the array around with us?

  5. Henry, I think you're in the ball park here. You have to keep two things separate; the dynamic property to be perceived and the kinematic information used to perceive it. The property is rightfully external to the organism. However haptics is a modality in which the kinematic consequences of that dynamic property occur within the organism. The organism faces the same problem, however; detect the information to perceive the property.

    Keeping this distinction clear is key, I think.

  6. Henry,
    "Carry the array around with us" also sounds weird... but maybe a bit better. There is certainly a difference wherein active touch creates the information (in a very concrete sense), but we typically do not create the information when using other perceptual systems.

    We do create some of the information used in other modalities, but indirectly. See for example, Gibson's discussion of how there is information to perceive the flexibility of surfaces... when an object is placed on it and we see the surface react. In such situations the information is in the light, but only because we did something that caused the information to be in the light. (And it would have been there equally if someone, or something, else had caused the information to be in the light.)

    In contrast, for haptic perception to be veridical, the information must almost always be self-produced. (This didn't have to be the case in an absolute sense, but the experimental data is pretty darn convincing.)

    Still, when Andrew says "the kinematic consequences of that dynamic property occur within the organism", I'm not exactly sure how to react. Certainly there are kinematic consequences within the organism. But what part of the visual-perception story are those consequences equivalent to? My intuition is that they should be equivalent to the bodily consequences of light - neurons firing, etc. (It seems to me that) Andrew's intuition is that the kinematic consequences should be equivalent to the optic array - the patterned light present whether the organism is there or not.