On this shortest week of the year, we slow down, cook lots of good food and burn the lights of memory and lived-holiday traditions. Whether or not your holiday behavior is intelligent, well, that depends on how much rum is in your eggnogg…
Recall in last week’s update, we reviewed some new developments in an intelligent, supramolecular assembly language, including an equation to measure intelligence of a system.
If intelligent, adaptive behavior is a measure of maximizing future options and pointing the system toward the resources to get there, it would certainly help to have a gas gauge and some kind of a memory to make comparisons with. But what are we trying to do with all this?
The answer seems to have been given decades ago, by a couple of amazing brothers, Frank and Otto Schmitt. Both of them obtained PhD’s from Washington University, hung out with Nobel laureates like Compton and AV HIll, and became pioneers of biophysics and neuroscience.
The younger of the two, Otto, coined the term “Biomimetic,” and his 1937 thesis created the now famous Schmitt Trigger, which is found in nearly every microprocessor in use today. The Schmitt Trigger is basically used for converting analog signals to digital, as a comparator, and a de-noiser. Any kind of biological or electronic feedback system probably involves some variant of the Schmitt Trigger.
Otto was a skilled gadgeteer and engineer. His older brother Frank paved the way for him at Washington University in St Louis, where Otto essentially played hooky his entire senior year of high school, and never did receive his diploma. But based on his prowess in building equipment for Frank’s lab, Otto was granted early admission to the university.
Frank and Otto spent summers at Woods’ Hole, where they studied the biophysics of squid and crab neurons. Otto’s 1937 thesis was an attempt to reverse engineer the electrical signals from crab neurons. Eccles, Hodgkin and Huxley later won their 1963 Nobel Prize in Physiology and Medicine for similar studies in the giant squid axon.
Frank went on to MIT, where he chaired the first NIH-sponsored study group on Biophysics in 1955, and founded the Neuroscience Research Program (NRP), which set the national agenda for neuroscience research for decades. Otto eventually settled after World War II at the University of Minnesota. Francis describes both his and Otto’s career paths in his biography, Never-Ceasing Search.
Otto had spent his war years at the Airborne Instruments Laboratory, developing Magnetic Anomaly Detectors used by the Navy to detect German submarines. As a result of his Defense-related work, during the 1950s, he became involved in early man-space flight efforts. The same year Eccles-Hodgkin-Huxley received their Nobel prize, Otto spoke at a Bionics conference in Dayton, saying this about the term he was later credited with creating, “biomimetics,” and how it might apply to manned-space flight:
“Presumably our common interest is in examining biological phenomenology in the hope of gaining insight and inspiration for developing physical or composite bio-physical systems in the image of life.”
Alex, What is Hysteresis?
While Otto’s 1938 paper, “A Thermionic Trigger” outlines the circuit’s design and applications, little detail is given on how he teased those hints out of crab nerves. For that, we have to go to a 1940 article, “Electric Interaction Between Two Adjacent Nerve Fibers.” which he wrote, while on a National Research Council-funded fellowship with A.V. Hill at University College of London. What Otto was going for with his Trigger, was to create a synthetic neuron.
By that point, based on his thesis work, as well as observations of Hodgkin’s work of the previous three years, it was clear that signals propagate along nerves in a wave-like fashion of excitation. What was unclear was how this proceeds without causing noisy interference between adjacent nerves.
Otto would later compare nerve signals to the coaxial cables he was familiar with from his Navy work with radar and signals. What was required was a way to compare signals, set a threshold for signaling, as well as a way to make sure the nerve didn’t respond to adjacent noise:
“The possibility of such an interaction between separate, active and resting, units is of interest from several aspects. (i) Normally, local currents set up in the vicinity of an active region do not, and obviously must not, excite adjacent fibres. Some mechanism apparently is present by which, not only the further propagation of the impulse in the active fibre, but also its “isolated conduction” is ensured. (ii) A subthreshold effect of an action potential on an adjacent fibre must be expected, however, since some part of the local current is bound to penetrate the surrounding tissue.”
The result of Otto’s biologically-inspired tinkering was a self-adjusting, memory, comparator, and de-noiser, all in one elegant, square “hysteresis” curve, which is the symbol used for his trigger in cicruit designs.
Fastforward 75 Years
What is amazing about Schmitt’s Trigger is how far down into the molecular and atomic biological underpinnings this intelligent, self-adjusting circuit appears to have reached. As noted last week, recent advances from Japan in creating biomimetic structures have stemmed from studying the biophysics of microtubules, key components of the cell’s skeletal and signaling mechanisms.
Several of the AFOSR-sponsored grant publications and patents highlight similar aspects of microtubules, including references to them as “programmable switches,” as well as showing a square electronic hysteresis curve. See, for instance, Figure 3 of “Multi-level memory switching properties of a single brain microtubule.”
In 2014, the Japan-based team spent a year at MIT, re-visiting nerve conduction studies, this time using biophysical tools that allow atomic-scale measurements of both the nerve membrane, but also of microtubules within the nerve. The data obtained are compelling, pointing to multiple levels of memory and feedback going on before, during and after nerve firing, and will appear this coming year.
But because microtubules appear to exist at the interface of physical and bioelectromagnetic forces, it is not surprising that they also have mechanical memory properties. A recent paper, “Why Microtubules Run in Circles: Mechanical Hysteresis of the Tubulin Lattice,” points out their role as both sensors and shapers of cellular forces, structures, and functions.
“Curved states can be induced via a mechanical hysteresis involving torques and forces typical of few molecular motors acting in unison. This lattice switch renders microtubules not only virtually unbreakable under typical cellular forces, but moreover provides them with a tunable response integrating mechanical and chemical stimuli…”
The relationship of microtubules to molecular motors and viral metastability was described in last week’s update. In the same way, mechanical metastability of microtubules allows them to “remember” force-induced curved shapes. These can arise, for instance, when molecular motors carry viruses and other packages along the microtubules. These properties resemble man-made shape-memory materials.
Maxwell, lo these many years
The point is, mechanical, and electromagnetic forces are working hand in hand. The interchangeability and exchangeability between intracellular, intramolecular forces, energy, resource use and intelligence almost smack of Maxwell’s now famous equations, recently highlighted as a cover story for IEEE’s Spectrum. The article points out that Maxwell’s equations’s did not immediately just leap into the fore–it took 50 years, and an autodidact Scottish telegrapher like Heaviside to reframe the equations into the form most high-school physics students now recognize:
Today, we learn early on that visible light is just one chunk of the wide electromagnetic spectrum, whose radiation is made up of oscillating electric and magnetic fields. And we learn that electricity and magnetism are inextricably linked; a changing magnetic field creates an electric field, and current and changing electric fields give rise to magnetic fields.
We have Maxwell to thank for these basic insights. But they didn’t occur to him suddenly and out of nowhere. The evidence he needed arrived in bits and pieces, over the course of more than 50 years.
Otto Schmitt, not one allergic to speculating a bit, said this of Maxwell, and alluding to more bits and pieces that may be finally starting to bubble up around the mind/brain biophysics:
I suspect that there is another whole layer of biomathematics dealing with these mental processes–radiation-like phenomena, but probably not just ordinary Maxwell’s Equations things… We traditionally think of electromagnetic fields as having an H vector and an E vector and that each generates the other… so I came up with this notion that there could be a non-orthogonal electromagnetic field which of course wouldn’t really bother with the shielded wires and so on, have different properties, and that Maxwell simply hadn’t run his mathematics into these special mathematics. (Harkness, 2002)
The Dayton IEEE virology study section is continuing with the Columbia MOOC, we are now focusing on viral genome lectures (#6, #7). There will probably be more to relate to the molecular design intelligence discussion, especially as we delve into the structures and mechanisms that genomes, both host and viral, have evolved to optimize their adaptive responses.