Some ad hoc thoughts about theories, models, and graduate school.

I had a great opportunity to read and study with a close friend D through 
his PhD and post doc in biophysics. He was (is) blind and needed a lot of 
reading, library, and editing work that was beyond his reader and 
secretarial stipends, and beyond the reach of a lot of his readers.

This work experience and my friendship (conversations, etc) during the work 
really transformed my understanding of science.

I remained a reader and later a lab tech over the course of a post-doc 
period that tried to empirically demonstrate a mathematical model or some 
relative of it. That part failed, but again I learned a lot.

Since I was doing the computer imaging (from fluorescence microscopy) work, 
I was there for the failure. A success would have shown a particular change 
in the false color image time series and that particular change did not 
appear. We were working in a friend's new lab. This meant the methodology 
was flawless, with state of the art tools at the time (mid-90s), and his 
technical help was invaluable. Fluorescence microscopy was all the new rage 
then. Now it is just standard fair.

This failure was a disaster for D's career. But it was one of those 
half-full, half-empty dilemmas, or so I argued to cheer him up. The failure 
said this model was wrong. In a more generous intellectual world, that could 
be considered a success in elimenating a particular direction.

But the technical fields are not generous. You can't use a failure to 
support the next grant unless you are already established and run a large 
lab where failures can be tolerated as routine blips in the work flow.

The above experiences have multiple barings on economics as a discipline. I 
started thinking about this last week during the Rag-off scandal. The thread 
on graduate school, Cult Mentality of Academia was another line of thought.

Part of the reason why economics is so intolerant of diverse views is that 
career success depends on positive results derived from known (acceptable) 
models. Since the public policy apparatus is rigged (via capitalist hegemony 
over research money) to limit what is `acceptable' there is no (paid and 
easy) way out of the universe of standard economic theory and its models. 
This probably explains why a lot of critical work goes directly to the 
publishers, internet, blogs, and so forth and skips the academic review 
channels.

In another direction, the above mentioned math model was derived from real 
time series data that showed periodic pulses decreasing over time. The known 
underlaying reason was essentially a diffusion theory. We could derive the 
general shape from the real time series. The method of derivation was a set 
of fourier transforms that produced a big pulse followed by a fast 
diminishing sin curve. This class of curves has a lot of fluid dynamics 
applications. In particular they can be linked to extra-cellular flows of 
substances that travel by diffusion.

Diffusion is a heavily modeled phenomenon in physics, chemistry, and 
biology. Because of its mathematics, it is a rich field for models. Go here 
to get the general math idea:

http://web.unideb.hu/zerdelyi/Diffusion-on-the-nanoscale/node4.html

Glance through Fick's law section 1 and go to section 2.2 Boltzmann's 
transformation, Parabolic law. Ignor the math and watch figure 2.1. The 
illustration provides a basic model for a periodic time series of diffusion 
in pulses---just imagine a continuous process that doesn't return to the 
beginning partition. The oscilloscope image of a beating heart is pretty 
much the same general shape.

This general model is very useful for tracking the diffusion of active 
substances and the timing of cell and tissue reactions that start with a 
strong reaction that diminishes in time as the substance triggers a reaction 
and diffuses away. This general cycle then is repeated with a new wave of a 
trigger or signal compound. The key feature of 2.2 to note is a homogeneous 
plane. This can be generalized into a three dimensional cylindrical tube 
with a homogeneous diffusion direction.

We were dealing with the cellular level, but the basic diffusion model for 
atoms works pretty much the same for small molecules in low viscosity 
mediums. For the model purposes the extra-cellular fluid was considered to 
have the physical characteristics of water (low viscosity, laminar flow) at 
normal room temperatures.

The important thing is to write the equations with clearly identified and 
quantified components most of which are already known and can be plugged in 
leaving as few variables as possible. Most of these quantified components 
were found in physical chemistry and or cell physiology and were well 
established.

The model was constructed to produce a differential gradient in the 
horizontal axis (perpendicular to flow direction producing a heterogeneous 
distribution) of a small growing section of a plant root, i.e. a cylindrical 
tube. The alteration of a standard model was this horizontal gradient. It 
would appear to be a change of color from one side of the root to the other 
under UV confocal micrscopy using fluorescent Ca2+ (glow in the dark 
calcium). A transverse gradient did not appear in over 600 runs.

In other words there was a predicted event that didn't happen on a 
reasonably large sample. The most famous experiment in physics that didn't 
work was the Michelson Morley experiment.

My buddy was too distressed over the failure to even think it might not have 
been a complete loss.

I always thought, well hell it mathematically worked, it was just 
empirically wrong. That should explain something beside the fact the model 
was wrong. But I had nothing hanging in the balance, no serious ego 
involvement. The ideas, methodology, and model were not mine in any creative 
sense at all. I was just along for the ride.

It was only much later that I realized I might have got more intellectual 
background and stimulus out of D's graduate and post-doc days than he did 
precisely because I was not on the chopping block. Whatever holes in my 
knowledge were never going to show and if they did, I could alway find out 
what to fill in without the terror of a credibility crash. I was free to 
learn as much as I could tolerate before saturation and fill in a lot of 
blanks in a general science education.

CG








_______________________________________________
pen-l mailing list
[email protected]
https://lists.csuchico.edu/mailman/listinfo/pen-l

Reply via email to