[Date Prev][Date Next][Thread Prev][Thread Next]   [Date Index] [Thread Index] [Author Index

Re: Communication between cells...



Interesting. Comments interspersed, below...
----- Original Message -----
To: ***
Sent: Thursday, January 20, 2005 11:16 PM
Subject: Re: [ROSEN] Communication between cells...

At 04:47 PM 1/20/05 -0500, Judith asks:
I don't know why you refer to temporal aspects as "one dimensional", though. Could you elaborate on that?

HP: Time can be represented (modeled) by numbers on a one-dimensional line. One particular time can be represented by one number. That means it is a one-dimensional or a scalar quantity. Rosen's Chronicles are such a time line or time series. As you know, your position in space requires 3 numbers, so it is three-dimensional. If I want to meet you for lunch, I have to specify 4 dimensions, 3 for where and one for when.
Hmmmm, while it's true that time "can be" represented by numbers on a one-dimensional line, that's not a very accurate representation of what I perceive about reality. The way I see it, although it has been said that change is the only constant, but it seems to me that time is the constant that makes change possible. One of the things my father labeled as being inapplicable to natural systems in general, and biological systems in particular, was the physics-based notion of "state". He did not view time as a collection of static moments. The state-based model of time is sort of like a strobe-light, but we know from experience that even though a strobe-light makes things look as if they're moving in choppy "snapshots" like bad animation... they are actually moving smoothly all the time, including in between the strobe's intermittant illumination. I would argue that the one-dimensional measure of time is guaranteed to make this natural system look very different than it really is.
Judith: In any case, there are aspects of this which fascinate me. Like; just because the "information capacity" of electrical signals is small, why would that be any reflection on how important this aspect is in the overall organization?

HP: We have good measures of syntactic information capacity (e.g. Shannon's logN bits). There is no good way to measure the importance of semantic information. Importance is mostly a subjective value judgement. In the cell one bad bit can be lethal, irrelevant, or anywhere in between.
Why do you say that there is no way to measure the importance of semantic information? We naturally do this all the time. Anyone who's grown a veggie garden should be able to recognize the principle. It's true that we can only arrive at a rough guestimate in the general neighborhood of the real value (to the organism, I mean), but this is good enough for humans to have become 'successful' to the point of overpopulation. Spinach needs cool soil temps for the seeds to germinate, tomatoes need warm soil temps. That's semantic information specific to each species, and it gets even more specific with various varieties within species-- and we've figured a lot of that out, too. In a sense, the semantics are what are encoded into the "internal predictive models" my father spoke of in Anticipatory System Theory.
 Judith: What if there's far more information being transmitted than what we define as information? Or perhaps the role of electrical communication is more along the lines of "a synergist"? Something doesn't need to be big to be potent.

HP: I think what you mean is, What if our model does not represent what's really going on? That is always a possibility, but then we expect that eventually the model won't work. The Hertz condition will not be met.
 
What I'm saying is that we have reason to suspect that there's far more going on in inter/intra cellular communication than just the chemical signals. So, it would seem prudent to investigate these potentialities before generating our model. If we don't, then under the circumstances it's a virtual guarantee that the model won't commute well with the system.
Another aspect I find intriguing is you said "electrical signals are important for rate-dependent coordinations" and for activity regarding time in the brain... Molecular/chemical activity is dependent on sequencing, isn't it? Rate, sequence, duration... all of these are time-related aspects. So these two modes are both dealing with different aspects of time.

HP: It is true that everything happens in real time, but nevertheless some things depend crucially on rates of change and other things do not. For example, all the basic laws of nature depend on (and are  expressed as) rates of change (time derivatives). By contrast, all formal mathematics and logic are rate-independent. That is, it makes no difference in its validity how fast you can prove a theorem. Computers will compute the same function no matter at what rate the CPU works. Similarly, the sequence of amino acids in protein synthesized under genetic control does not depend on the rate of synthesis (within wide limits). The meaning of the text you are now reading does not depend strongly on the rate you read it (again within limits). The left side of the modeling diagram is rate-dependent (Nature's causal entailments). The right side is rate-independent (formal inferential entailments). That is why the observer's choice of encoding that must connect the two sides can not be entailed by either side.
Well, this isn't what I was driving at (although I would suggest that the "limits" are all context dependent, which is one of the commutations between those inferential entailments and causal entailments). What I was speculating about was whether these kinds of phenomena in living organisms offer one of the avenues for studying the nature of "time", as a contributor to, or in addition to, the model-based behavior that Anticipatory Systems exhibit... AND whether such a study of "time" in this context might perhaps shed some light on the communications between/within cells, as well. It seems likely to me that this multiple-aspects-of-time involvement is actually associated with "anticipation". One of the aspects of anticipatory systems that really needs further study is the "internal predictive models". Specifically, what information is encoded and how does it get encoded, even in a single celled organism that has no central nervous system.
 
Judith