Book review: Electrical fields and the experienced life
Christof Koch
DOI: https://doi.org/10.3389/fncom.2024.1462913
IF: 3.387
2024-10-09
Frontiers in Computational Neuroscience
Abstract:Humanity only discovered that animals and their brains run on electricity, rather than on fluids, in the late 18 th century (by the Italian physician Luigi Galvani experimenting with twitching frog muscles). The action potential, the brief (< 1 msec) and small (< 0.1 Volt) excursion of the voltage across the neuronal membrane separating the inside of cells from the outside, wasn't recorded until the middle of the 19 th century (the 'nerve current' of Emil du Bois-Reymond). Action potentials, or 'spikes' in the lingo, are the primary means by which information is relayed between nerves and from there to muscles. The underlying brief changes in the conductance of the membrane to specific ions were not analyzed until the first half of the 20 th century. This understanding found its first efflorescence in Alan Hodgkin and Andrew Huxley's elucidation of the biophysical mechanisms underlying the initiation and propagation of the action potential in the squid giant axon. They formulated a quantitative model that reproduced their observations in terms of timeand voltage-dependent membrane-bound conductances (one each for sodium and potassium ions, in parallel to a constant 'leak' conductance). Hodgkin and Huxley took three weeks, on a handcranked calculator, to solve for the propagation speed of the action potential, which was within 10% of their observed value. This astounding feat earned them a Nobel Prize a decade later.Electric Brain Signals demonstrates how far modeling of nervous tissue has matured over the ensuing seventy-plus years. Jointly composed, in a single melodious voice, by Gaute Einevoll, a well-known and genial physicist cum neuroscientist, and five of his former students at the Norwegian University of Life Sciences and the University of Oslo, the book allows the reader to develop physics-informed intuitions of how the microscopic currents from about sixteen billion nerve cells of the human neocortex, the highly convoluted outermost portion of the brain, sing the body electric, the macroscopic electrical potential measured inside the head with implanted microelectrodes and outside the scalp with EEG electrodes.Modeling these fields is based on two assumptions: firstly, neuronal activity is characterized by currents, flowing through Hodgkin-Huxley like-membrane conductances, and emanating from all points along the dendrites, cell bodies and axons of nerve cells. Secondly, at the scale of tens of micrometer and larger, extracellular space -filled with salty water squeezed into the tiny, twisting spaces outside of cells -behaves like a resistive cytoplasm. That is, membrane currents superimpose linearly, and the resultant electrical potential is given by volume conduction as dictated by Ohm's law. Endogenous magnetic effects are negligible (due to the low velocity of the charged ions and the tiny magnetic permeability of brain tissue, about the same as the surrounding air); so that's it. Everything else is a consequence of volume conduction. Neither fancy quantum effects between neurons (leaving open the possibility of intra-and inter-molecular quantum effects) nor other exotic physics appears to be relevant (although what is not disallowed by physics may, of course, be exploited by natural selection). Of course, the devil is in the details. Whether current sinks and sources superimpose cooperatively or destructively, how they contribute to the various EEG bands (delta, alpha, gamma) and so on depends on highly orchestrated synaptic volleys exciting and inhibiting densely interwoven neural lace of staggering intricacy. Taming this complexity, in tissue made from hundred sixty billion cells of five thousand highly heterogenous types, a challenge upon which the Human Brain Project foundered, is going to take all human ingenuity and then some.The book facilitates this process using a series of highly instructive toy models (and Jupyter notebooks to accompany them) to guide the reader in interpreting spatial-temporal EEG signals as reflecting underlying neural activity patterns filtered through the cortical gray matter, the highly conductive cerebral spinal fluid, the skull and the scalp. Other topics covered are electrodes that are implanted in patients and rest on top of cortex, but underneath the skull, and modern magnetoencephalography (MEG) that tracks the very weak magnetic fields produced by brain activity at higher spatial resolution than EEG.Why, in the age of machine learning, should anyone care about the messy physics of biological tissue? Indeed, does the brain even care about its own biophysics? Hemodynamic activity in the heart generates sounds that can be picked up by a cardiologist to diagnose a heart murmur or an irregular rhythm. But there is no evidence that these weak auditory signals are of any relevance to the body. What about the electrical fields generated by the brain? Do they affect its processing outside epileptic seizures -Abstract Truncated-
neurosciences,mathematical & computational biology