Let's get on with it

[From Bruce Abbott (951014.1400 EST)]

Bill Powers (951014.0720 MDT) --

Systems that behave the same way, from the standpoint of an external
observer, are not necessarily organized the same way inside.

I agree.

Structures
that appear anatomically the same do not necessarily accomplish their
functions in the same way.

True as far as it goes. Yet, is there any reason to believe that the
control system by which I move my arm and the one my dog Muffie uses to move
her front leg might have something in common, not only in their structural
organization but also in their functional organization? This is not to say
that the two structures are identical--after all, there are important
structural and functional differences as well. No, they do not
_necessarily_ accomplish their functions in the same way, but observation
and testing can reveal whether they do or not. That wonderful little
diagram in B:CP showing the gross anatomy of the lowest-level
muscle-length/force control system was first worked out in other animals,
yet turns out to apply rather well to the homologous human system, despite
there being some differences.

Control systems are organized to create

certain outcomes, but the fact that the same outcomes keep occurring
does not mean that the same actions were used each time to create them.

True, but I'll bet the same control system, having the same anatomical
locus, initiated those actions in accordance with its error signal, and that
is the point being debated here, not whether the same control system always
produces the same actions.

If a bird builds a nest and a mouse builds a nest, there is no reason to
conclude that birds and mice share a common nest-building ancestor.

True enough: there must be other evidence before one can draw such
conclusions. One must distinguish those cases whereby similar natural
requirements have produced independent, analogous solutions, which may
depend on very different biological machinery, from cases in which the
solutions emerged early and persisted (although perhaps with modification)
over evolutionary history.

Nests have several functions, and an adaptive control system
organization will find a way to accomplish those functions quite
independently of how any other system might have accomplished the same
functions. If it happens that two organisms use the same method to
accomplish the same function, the reason may be simply that there is no
other easily-found method that will accomplish that function. One can't
conclude that there is some specific structure inside the organisms that
makes them both use the same method and hence get the same result.

Yes, and that is precisely why one needs to observe both function AND
structure before one can reach any secure conclusions on the matter. The
bee's vision system and the rat's are clearly examples of parallel
evolution; the rat's visual system and ours, on the other hand, show clear
signs of having been derived over the course of evolution from the same
protostructure. Human and rat evolution have resulted in modification and
elaboration of this basic system along different lines (for example, the
human eye contains photoreceptors that make color vision possible, whereas
the rat eye does not), and one can expect important differences between
these two species to be found in the neural circuitry involved in analyzing
the visual input. Yet the basic organization of the rat's eye and ours is
the same, both species have optic nerves which partially cross at the optic
chiasm, both optic tracts then run to the lateral geniculate nucleus of the
thalamus and synapse with neurons located there, and those neurons in both
species send their axons to the occipital lobes of the cerebral cortex,
where they terminate on cortical cells in such a way that the spatial
organization of the primary visual area lays out point-for-point like a map
of the retinal surface of the eye. Is it logical to believe that this
parallel organization is an accident of parallel evolution, and that the
structural similarities have nothing to tell us about probable functions
these structures serve? I think not. Yes, one must be careful not to
assume too much--over the course of evolution old structures do sometimes
take on new functions through modification of their detailed connections,
but you seem to be suggesting that one can learn nothing at all from the
study of other animals that will help us to unravel the puzzle of human
brain function. Am I overstating your case?

Getting back to nesting, birds and mice may have independently evolved (not
learned) control structures that lead, under appropriate conditions, to the
building of nests. In both cases the nest is a solution to a problem posed
by the animal's environment in conjunction with the animal's individual
needs. The mere fact that both build nests is not sufficient to conclude
that the same brain structures are involved. Yet all birds of a given
species build their nests in the same way, using the same stereotypical
movements, although there is considerable flexibility in program to allow
for such variables as what nesting material is available. The same may be
said for rats. Having discovered that particular areas of the bird's (or
rat's) brain organize this system in its particular way, it would be no
surprise to find that the homologous structures of related species are
similarly organized to serve a similar function. Knowledge of this system
in birds may not apply to rats and vice versa, but the question of whether
it does is an empirical one that can be settled by appropriate research.

The problems I see in your arguments and explanations is that behind
them there seem to be large numbers of assumptions that are contrary to
these principles. Moreover, these contrary assumptions are widely shared
in your profession, to the point where they are taken for granted. The
difficulty I'm having is in raising doubts about these assumptions
without appearing to claim that they are false. By trying to cast doubt,
I am not trying to claim that I know your statements are wrong, but
trying to show that there is a level of investigation that is called for
to find out whether the assumptions are justifiable. If stimulating a
rat's amygdala results in a rat's showing visible behaviors which, if
they occurred in a human being, would be called fear, this is not an
indication that the rat is experiencing fear or that the effect of the
stimulation on the rat is the same as it is for the human being. It's
not an indication to the contrary, either. It's an open question, which
can be resolved only by further investigation. The rat's amygdala has
been evolving just as long as the human's, and by now could be
performing functions with no parallels in the human, and accomplishing
similar functions by quite different means. COULD be.

You may be surprised, but I absolutely agree (except for the first sentence,
see below). If we differ on this issue, it would appear to be in the degree
to which we believe that the data support these assumptions, or perhaps in
our beliefs as to the degree to which researchers have been sensitive to
these questions and have sought out evidence with which to addrress them.
My experience has led me to conclude that in ceratin areas of research there
is reliable empirical evidence that supports the assumptions on which my
arguments have been based. But I agree that your concerns about
interpretation are certainly cogent and ought to be uppermost in the minds
of everyone doing this kind of research. It is all too easy to be mislead
by surface similarities. In my opinion, however, in the areas of research
about which we've been debating there have been serious efforts made to
address these issues through empirical work.

I guess we'll have to put aside some arguments if we're to make progress
in developing PTC applications to behavior. I can see the logic in your
answers to my questions and comments, but can't seem to get across the
framework from which my questions arise.

I'm wondering if the problem isn't highlighted in something you said earlier:

Nests have several functions, and an adaptive control system
organization will find a way to accomplish those functions quite
independently of how any other system might have accomplished the same
functions.

Perhaps our difference lies in the degree to which we believe that adaptive
reorganization can restructure the functions of the brain. By "adaptive" I
assume you mean systems capable of being reorganized until some organization
is found that establishes control over some perceptual variable, as opposed
to parallel evolution in which different structures may evolve to serve
similar functions. To some unknown extent the functional organizations of
your brain and mine may differ because our experiences have led to our
developing different systems for controlling particular variables and even
to different variables being brought under control. We can go further: in
young children damage to brain structures in one hemisphere (e.g., Broca's
area in the left frontal lobe) often leads to changes in the functinal
wiring of the other hemisphere so that the function reemerges there. The
brain's wiring is surprisingly plastic, at least up to a point. But this
fact should not lead one to conclude that the brain is infinitely plastic,
that any function can appear anywhere as a result of rewiring guided by
experience (reorganization).

Following destruction of Broca's area, the language functions served by
Broca's area will reappear in the young child, not anywhere, but in the
homologous structure of the right hemisphere, and in the adult it will not
appear at all. Furthermore, if our two brains were examined and compared
after our deaths, neuroanatomists would find the same recognizable
structures in both: medulla, pons, cerebellum, hypothalamus, and so on, and
the same nulclei in the same relative positions within each of those
structures. Yes, there would be some differences in the sizes of certain
structures and in the number of nerve fibers comprising the same neural
tracts, and at some points there would be detailed differences in neural
interconnections wrought by our differing experiences. In few of those
structures would there be any significant difference in what had been the
structure's function. Perhaps you have learned to play the Saxiphone and I
have not. Clearly this difference in abilities would indicate that you
possess control systems, formed through practice, that I lack. Yet the
basic structures within which those systems developed would be essentially
the same in you and me, and serving similar functions.

The problems I see in your arguments and explanations is that behind
them there seem to be large numbers of assumptions that are contrary to
these principles.

Large numbers of assumptions? Now there's a vague referent! Contrary to
these principles? Which principles? Contrary in what way, to which
specific assumptions? Perhaps a short list, giving the assumption, the
principle to which it is contrary, and a brief explanation why you think the
two are contrary might help. I believe I have shown that the assumptions I
make are justified by the data and not at all contrary to any principles of
PCT. What am I missing? I think the suggested listing could help
enormously to clarify your position for me.

Regards,

Bruce