http://www.cs.bham.ac.uk/~axs/misc/self-awareness.txt

From Aaron Sloman Fri Sep 24 10:42:55 BST 1999
From: Aaron Sloman <A.Sloman@cs.bham.ac.uk>
Subject: varieties of self-awareness (Was: Re: grooming)
To: PSYCHE-B@listserv.uh.edu

Bernie Baars wrote:

> I thought Ramon Moore's idea about self-grooming and its relevance to
> consciousness and self was really very interesting. Sue Pockett objects that
> the mirror test is often considered to be "the" measure of physical self
> recognition, but I have always been somewhat uncomfortable with that. The
> fact that humans and some primates recognize a mark on their foreheads while
> other big-brained creatures don't ,strikes me as pretty arbitrary. Other
> species may recognize their own pheromones --- who is to say which is more
> fundamental?

Here's something I have often wondered about: many animals use bits of
themselves to do things to other things (e.g food), and treat their body
parts differently from the other things.

E.g. A carnivorous bird may use its claws to hold a captured
animal against the branch it has perched on while it pecks bits of
flesh off.

Do such birds ever peck their own feet?

Animals with long tails can pounce on and bite moving objects?

Do they ever try to eat their own tails?

1.  Is there very precise (genetically determined?) control of
        pecking/biting by visual and tactile cues which ensures that
        no error ever occurs?
        How, exactly?

2.  Do they make errors and then learn from the unpleasant result that
        this sort of object is to be avoided (even though it may
        taste good!!)?
        What exactly do they learn in this case? See next question.

3.  When they make the distinction between bits of themselves and the
    rest of the physical world:

        is there any representation, within the animal, of itself as
            having an extended body of which *this* is a part and *that*
            isn't,
        or is there simply a shallow behavioural rule (as implied by 1)?

        Or are there far more alternatives than those two?

Have ethologists or psychologists investigated these questions?

The kind of "self-awareness" described here involves separating the
*physical* world into self and non-self (possibly while transformting
bits of the latter into the former!), and that is, of course, different
from the kind of self-awareness involved in knowing that one has visual,
tactile or other experiences, which has led to all the philosophical
discussion about qualia.

Both are different from the kind of self awareness involved in knowing
that one is selfish, clever, easily embarrassed, etc.

Or knowing where one is in a pecking order.

Maybe there are other kinds of self-awareness. E.g. can a hunted stag
distinguish states when it is nearly exhausted from states when it is
still fresh, and change its behaviour accordingly (e.g. searching for a
place to hide vs trying to out-run a predator)?

Again, if this occurs, is it triggered by a shallow behavioural rule or
is it part of a more general ability to represent itself as having a
collection of time-varying states which can be thought about
independently of triggering specific responses? I suspect only the
former.

How many significantly different kinds of self-awareness are there?

By "significantly different" I mean being capable of existing in
an animal without the other kinds, and therefore making use of different
information processing architectural features, and possibly being put to
different uses.

All sorts of homeostasis involve a very simple kind of self-awareness.

Biological evolution is a damn good engineer.

As information processing capabilities of organisms evolved and analog
feedback loops were supplemented with many other kinds of control
mechanisms, it is likely that organisms developed many different ways of
acquiring, storing and using diverse kinds of information about
themselves.

Some of those might be restricted to self-representation (e.g. body
temperature? nutrient levels?) and others equally applicable to self and
to something else e.g. information about physical extent, about spatial
location, about direction of movement, about location in a social
hierarchy.

My guess is that there are MANY different kinds of self-awareness, and
when we have distinguished them clearly we'll be surprised to find how
many of them occur in all sorts of animals.

And in things like computer operating systems, word processors,
compilers, spreadsheets, etc.

Cheers.
Aaron
===
Aaron Sloman, ( http://www.cs.bham.ac.uk/~axs/ )
School of Computer Science, The University of Birmingham, B15 2TT, UK
EMAIL   A.Sloman@cs.bham.ac.uk
Phone: +44-121-414-4775 (Sec 4774)       Fax:   +44-121-414-4281
http://www.cs.bham.ac.uk/research/poplog/freepoplog.html