From: William Benzon (firstname.lastname@example.org)
Date: Thu 26 May 2005 - 17:26:37 GMT
on 5/26/05 8:49 AM, Scott Chase at email@example.com wrote:
> --- Robin Faichney <firstname.lastname@example.org> wrote:
>> Thursday, May 26, 2005, 1:41:18 PM, Scott wrote:
>> <Book recommendations>
>> Thanks for that.
>>> people engaged in a common activity will collapse
>>> each other, neurally speaking. They are almost as
>>> wrt neural states.
>> Can you give me a reference for this?
> It's what I haphazardly remember from a recent reading
> of William Benzon's book _Beethoven's Anvil_. He has a
> pertinent definition of "Ensemble State Collapse" on
> page 61 of the hardcopy where those engaged in what he
> calls "musicking" or musical activity have their
> neural states as a whole collapse into one sort of
> like a single person. I think he would be the one to
> ask for explication, because I'll probably botch it up
> pretty badly. There's some loose ends I need to tie up
> between some of his concepts (like the persona concept
> I just had an exchange with him over) before I can
> "get a grip" so to speak.
It's a very tricky business. I adopt Walter Freeman's account of the
nervous system in terms of complex dynamics. That means we think about the
state space or phase space of the whole system, which is a function of the
number of elements, the number of states each element can take, and the
dependency between elements. The more elements and the more states per
element, the large the phase space; dependency between elements reduces the
size of the phase space. We don't really know how to estimate the size of
the phase space of a human nervous system, but it it clearly quite large.
Now, what's the size of the phase space for a group of people interacting
with one another? The number of elements is much larger, i.e. ten times as
large for a group of ten, and so forth. But, as these people are
interacting with one another, that introduces dependencies between their
respective nervous systems, thus reducing the overall size of the collective
phase space. The argument I make -- which I sketch below -- is that when the
group is involved in making music, the collective phase space is no larger
than that of a single individual.
I explicitly do not argue that the neural states of the individuals are the
same. Such an assertion doesn't even make sense because there is no way we
can compare the states of two brains, not at this micro-level of detail. If
it were something like sleeping or waking, that we can compare, but such
states are not defined a high level of detail. But we're looking at states
defined at the level of the neuron or even the synapse. At this level it is
impossible, even in principle, to ascertain whether or not two brains are in
the same state.
Why? Because we have no way of establishing a 1-to-1 correspondence between
the elements in two different brains. While human brains are grossly alike,
there is no reason to think that they have the same number of neurons,
either in total, or for comparable regions. So, we cannot put the elements
of two brains in 1-to-1 correspondence. Without such a correspondence, there
is no way to compare the states of individual elements.
Here's the passage from my book (Beethoven's Anvil, Basic Books, 2001, pp.
Let us expand our description of the brain to encompass social systems of
two or more people. When thinking about the dynamics of individual brains,
we are thinking about how complex brain states evolve from one to another.
Social dynamics is about the evolution of states of the collective neuropil.
As far as I know, no one has considered this problem. Weıre going to have to
make things up out of whole cloth.
As an extreme case, imagine that we are dealing with a pair of individuals,
Frick and Frack, such that each brain has the same number of possible
states, Q. These two people are in different locations and completely
unaware of each other. There is thus no dependency between what happens in
these two brains; they are completely decoupled. If Frick is in state
number 23,587, Frack could be in any one of Q states and the reverse is true
as well. Since Frick can be in any one of Q states, it follows that we have
Q^2 (Q times Q) possible states for the pair.
Once Frick and Frack start interacting, however, dependencies develop between their respective brains; their actions constrain one another. The number of possible states for the pair is no longer Q2. I donıt have a general strategy for how to estimate the number of states possible to the ensemble. But, I would like to consider an extreme case, where one person is attempting to imitate the other exactlya scene, for example, that was brilliantly realized in the Marx Brothersı classic Duck Soup. In that case, I suggest, the number of states possible to the ensemble approaches the number possible to one member of the ensemble acting alone. We can say that that interaction has the maximum possible coupling strength. Less demanding interactions will have weaker coupling.
I am not claiming that, during imitation, Frick and Frack have the same
brain states. That is meaningless, for there is no way to compare the states
of two brains. As Freeman has noted, brains are unique, reflecting unique
histories. Rather, I am asserting that the demands of imitation force the
two brains to depend on one another so that the state-space available while
performing that task approaches that of a single unconstrained human brain.
Imitation, of course, has been extremely important in recent cultural
theorizing; it is at the heart of memetics and of Merlin Donaldıs very
influential book Origin of the Modern Mind. One might, of course, object
that, however closely two individuals might match their physical motions,
their minds are free to wander. To which I respond that their minds arenıt
that free: it takes quite a bit of concentration to imitate someone. In any
event, my analysis is quite informal.
The same argument holds for a group of three or more people. If they are
completely decoupled, the number of possible states in the ensemble is Q^n,
where n is the number of people in the group. But, to the extent that all of
the members of the group are doing the same thing, the number of possible
states will, as in the case of Frick and Frack, approach Q, the number of
states possible for an individual brain. What, now, do we make of an
orchestra of musicians performing Beethovenıs Fifth Symphony? Given that
that composition can also be realized by a single musician performing
Lizstıs piano reduction, it would seem that the number of states in the
orchestra must approach the number possible to one individual.
The musicians in the symphony orchestra, however, are not imitating one another. Yes, the musicians playing the same part are doing the same thing. But the piece as a whole is scored for some twenty-plus highly interdependent parts. Thus we have a new principle:
Ensemble State Collapse: The size of the collective neural state space of a
musicking ensemble approaches that of a typical member of the ensemble.
This is closely related to the equivalence principle we described above.
That principle was about the physical continuity of the coupled system: one
neuromuscular system or several? This one is about the size of a systemıs
If this seems counter-intuitive, remember that everyone in the ensemble
hears all of the parts. They differ in what they are doing, but, of course,
what they are doing is constrained to the sounds they are hearing. The major
components of the brain dynamics of each musicking individual will be
entrained to the music itself. The differences, of course, will reflect the
different motor dynamics required of each person to make her contribution to
the music. In order for her part to fit in, each musician must actively
track, actively intend, the full musical texture. In his book African
Rhythm and African Sensibility, John Miller Chernoff notes that even the
most skilled African drummers, often find it difficult, or at least strange,
to play against a multi-part rhythm when even one of the interlocking parts
is missing. They need the whole gestalt.
This was distributed via the memetics list associated with the
Journal of Memetics - Evolutionary Models of Information Transmission
For information about the journal and the list (e.g. unsubscribing)
This archive was generated by hypermail 2.1.5 : Thu 26 May 2005 - 16:40:43 GMT