Regular Series


Vol. 49 (2018), No. 12, pp. 1949 – 2138

LVIII Cracow School of Theoretical Physics Neuroscience: Machine Learning Meets Fundamental Theory

Zakopane, Poland; June 15–23, 2018

Life at the Edge: Complexity and Criticality in Biological Function

abstract

Why life is complex and — most importantly — what is the origin of the over abundance of complexity in nature? This is a fundamental scientific question which, paraphrasing the late Per Bak, “is screaming to be answered but seldom is even being asked”. In this article, we review recent attempts across several scales to understand the origins of complex biological problems from the perspective of critical phenomena. To illustrate the approach, three cases are discussed, namely the large scale brain dynamics, the characterization of spontaneous fluctuations of proteins, and the physiological complexity of the cell mitochondria network.


Multi-level Explanations in Neuroscience I: From Genes to Subjective Experiences

abstract

Brains are the most complex systems in the known Universe. Understanding brain dynamics, control of behavior and mental processes is the ultimate challenge for science. It requires multi-level explanations, starting from evolutionary pressures, genes, proteins, cells, networks of neurons, psychophysics, subjective experiences at the mental level, and social interactions. Many branches of science contribute to this endeavor. Physics provides experimental and theoretical tools at the molecular and brain signal processing level, and mathematical tools at the level of neurodynamics. Inspirations from understanding brains are of great practical importance in many fields, including neuropsychiatry, neuropsychology and artificial intelligence. Neurodynamics provides the best language to link low-level molecular phenomena to high-level cognitive functions. Computational simulations help to understand molecular dynamics and analyze real brain signals. This is a very fruitful area of research that requires global, interdisciplinary effort of experts from many branches of science.


all authors

M.K. Komorowski, M. Aghabeig, J. Nikadon, T. Piotrowski, J. Dreszer, B. Bałaj, M. Lewandowska, J. Wojciechowski, N. Pawlaczyk, M. Szmytke, A. Cichocki, W. Duch

Multi-level Explanations in Neuroscience II: EEG Spectral Fingerprints and Tensor Decompositions for Understanding Brain Activity — Initial Results

abstract

Two examples of muli-level explanations of brain activity are provided. First study is aimed at extraction of information from EEG to recognize which brain regions are active using spectral fingerprinting. It is based on forward and inverse modeling of electric potentials measured by sensors placed on the scalp, and computing power spectra from different brain locations. Reliable recognition of specific brain activity using EEG may lead to better diagnostic and therapeutic methods, and various new ways of building brain–computer interfaces. In the second study, infant EEG data collected at our BabyLAB were used to derive Event-Related Potentials (ERPs), in an oddball paradigm with two types of deviant stimuli (easy and hard) and one standard stimuli. Tensor decomposition of these signals, conforming to non-negative Canonical Polyadic decomposition (NCPD) model and non-negative Tucker decomposition (NTD), is used to characterize differences in processing these stimuli. Multi-domain temporal, spectral, time-frequency representation (TFR) and spatial information features are simultaneously analyzed for more reliable representation of the underlying source of brain activity. Results show right-side asymmetry for 5-frequency (Hz) theta band and may be due to the dynamical process of expectancy and surprise, corresponding to deviant detection reflected in the mismatch response.


Across Neurons and Silicon: Some Experiments Regarding the Pervasiveness of Nonlinear Phenomena

abstract

The nonlinear dynamics of neurons can be viewed as the substrate through which the vastity of mental states and processes making up our subjective experience emerges from the brain as a physical object. While at present linear dynamical systems and devices may appear to have greater practical usefulness owing to their easier mathematical tractability, nonlinear phenomena pervade nature at all scales and harbor immense generative potential. Such phenomena have aspects of universality and, therefore, can be elicited, among other possible scenarios, also in analog electronic networks containing one or more nonlinear elements, and these are particularly convenient to realize and study experimentally. Here, a concise review of the author’s work in this area is presented, without any attempt to comprehensively survey the field. Firstly, atypical circuits based on bipolar-junction transistors, inverter gates, and neon lamps are introduced; these recapitulate, at least phenomenologically, certain aspects of neural dynamics such as the generation of irregular spike trains. Secondly, the spontaneous emergence of synchronization patterns featuring modular organization, remote entrainment or implementing viable walking gaits is illustrated in networks constructed of those and other circuits. Some reflections on the potential relevance of comparing such profoundly different physical systems experimentally and possible directions for future work are given.


Interpreting Electrical Signals from the Brain

abstract

One of the holy grails of brain research is to understand how the brain functions in a way that would allow us to alter it in positive and productive ways. In this respect, one of the most promising tools is the measurable electrical signal that is produced by brain tissue. However, as there are many ways to approach experimentation and analysis of the signal, it is of value to have a framework in which to do so. This series of lectures attempts to provide this in three distinct parts. The first part lays out a framework for asking and answering questions about what is universal versus unique across species and individuals in the context of specific motifs versus statistical features. It includes thinking on how to interpret an aggregate field signal, how to understand the signal from the perspective of field potentials at different resolutions (LFP, ECoG, EEG) and the choice of systems and preparations of study (in vivo, in vitro, species choice). The second part describes the phenomenology of avalanches and coherence potentials in LFP and ECoG and the way they fit into the framework described in part I. Accordingly, it describes the insights that arise about species universality versus uniqueness, as well as behavior on instantaneous versus integrated timescales. The third part explores the unique opportunities afforded by the noninvasive nature of EEG to combine dynamical views with behavioral inputs and outputs. I provide a schema that considers the design of studies that relate acute and integrated inputs or life experiences to dynamics, and, in turn, to cognitive and emotional outcomes and behaviors. I also suggest that insights from LFP and ECoG can drive new waveform-based analytical approaches and insights with EEG and provide an example to demonstrate this.


The Kinematics of Spike Trains

abstract

Neural cells, the main agents in the brain responsible for processing of sensory information, animating our limbs, our thoughts, desires, and actions, communicate with each other by sending electric pulses called action potentials or spikes. This communication can be described with point processes which we introduce here simply.


top

ver. 2024.03.17 • we use cookies and MathJax