comments 4/10 07
This paper is very nicely written and exciting to read. For a scholarpedia article, I don't think it should be longer than it is.
One major problem with the article in its present form, however, is this: Rather than being a review of different self models, it is reallyt af review of SMT. Either the article should be called "SMT" as a subsection to "Self Models", or it should be rewritten to allow for mentionings of other approaches. A focus on SMT as a kind of "example" of a self model could be ok, but in the present form, the focus on SMT is much too strong.
reply 14/10 07
I am afraid the second half of this comment may be based on a misunderstanding. This is an article on the originally philosophical concept of a self model and some associated empirical research programs, which are now developing. It is not a review of different philosophical or scientific approaches to the problem of self-consciousness or "models of the self" in a more metaphorical sense. A "self model", as used here, is a representational entity, typically, but not necessarily, conceived of as physically realized in a single biological brain or a single machine - and not a theoretical model, or a product on the level of academic discourse within a scientific community or a particluar philosophical stance on self-consciousness. I agree that it would be good to have an (unbiased) article on self-consciousness per se (as well as one on teleofunctionalism!). But it seems the comment above is based on an equivocation of the term "model".
However, it would be completely outside of the scope of ScholarPedia to even attempt a review of purely empirical research programs of self-consciousness. For those interested in purely philosophical discussions, here are two brief, but quite informative and substantial reviews of philosophical approaches to self-consciousness, one from the angle of classical analytical philosophy, and one from a phenomenological perspective, with both carefully ignoring SMT and the notion of a "self model" altogether:
reply, 24/10 07
I accept the distinction between "self model" and "self-consciousness per se" as you describe. If you make the distinction more clear in the actual article (others may be confused in the same way), I'll accept it.
reply, 25/10 07
Good point. I have changed the "that" in the second sentence into "this specific", that should very efficiently make the distinction right at the beginning, and for every reader.
Comments on Self-Models
Overall, a very rich and interesting article. As a general comment, it would be very helpful to add to the introductory section a short paragraph laying out what will follow in the article—a road map for the reader. Likewise, it would be very helpful to add a short concluding paragraph reviewing the major points in brief. I realize that the article is already quite long for this format, but it would help the reader not familiar with this style of work to follow along. Below are several more specific comments:
1. Could you please specify, preferably early in the article (section 2?) what a model is? That concept needs to be clarified. Is it just having a representation? If not, what separates it from representational content that is about a single domain?
2. It would be helpful to say a bit more about the constraints the differentiate nonconscious from conscious self-models. There is some material in section 2, but it could be fleshed out a bit more.
3. In the paragraph on functional analysis, a word on why self-models which “achieve sensorimotor integration, generate complex, flexible and adaptive behavior, and attend to and control your body as a whole” must be conscious. Again, this is a request to stress the constraints on conscious, as opposed to nonconscious, mental states.
4. There is no need to completely restate the paragraph on the robotic starfish in the explanation of 2nd-order embodiment. It is enough to refer back to the original paragraph.
5. Is 3rd-order embodiment simply being conscious of your self as embodied? Could a disembodied brain have this sort of awareness? Why not? It wasn’t clear to me that this captured the “body” element at play in many uses of “embodied” in consciousness studies. Or does 3rd-order embodiment always presuppose the presence of the other 2? Please clarify.