Title: Do local field potentials and the EEG primarily reflect inhibitory processes?
Alain Destexhe, Paris-Saclay Institute of Neuroscience, CNRS, Gif sur Yvette, France.
Abstract: We will review recent data and analyses suggesting that the local field potential (LFP) is strongly related to the activity of inhibitory neurons. Computational models are used to test biophysically-plausible mechanisms to explain these observations. We find that inhibitory conductances in pyramidal cells form the dominant component of the LFP. We discuss whether more global signals, such as the electro-corticogram (EcoG) or the electro-encephalogram (EEG), would also primarily reflect inhibitory processes.
Title: Eigenmodes of the brain: a graph spectral theory of brain activity
Ashish Raj, School of Medicine, UCSF, San Francisco, US
Abstract: Although neural mass and field models are capable of replicating complex power spectra and functional connectivity across wide frequency ranges, it is not clear whether they encode the minimal essential rules of brain communication. We will discuss recent approaches that rely on graph theoretic models, especially a linear model that eschews detailed local oscillator dynamics in favor of the effect of long-range fiber connectivity captured by the graph. We will explore whether graph eigenmodes have the capacity to capture essential features of brain activity, without regard to detailed local properties.
Title: A modelling approach for describing stimulation and electrophysiological recording of unconscious and conscious brain states
Anna Cattani, Department of Biomedical and Clinical Sciences "Luigi Sacco", Milan, Italy
Abstract:
The clinical assessment of patients with disorder of consciousness critically depends on the patients’ residual ability to create a connection with the outside world by conveying their subjective experience through motor behavior. However, the severity of the brain injury may lead the patient to be unable to perform any kind of movement or adequately understand the required command. For this reason, consciousness may go undetected in brain-injured patients who are unable to communicate. To overcome this clinical problem, it has been developed a theory-driven, objective measure of the level of consciousness (Perturbational Complexity Index – PCI) calculated as the algorithmic complexity of the spatiotemporal pattern of electrophysiological recordings (EEG) obtained by perturbing the cortex with transcranial magnetic stimulation (TMS) (Casali et al. Sci Tr Med 2014). This measure well correlates with the level of consciousness in single individuals and discriminates wakefulness from NREM sleep and anesthesia in healthy subjects, and patients in minimally conscious state (MCS) from patients in unresponsive wakefulness syndrome (UWS). Intracranial recordings in human suggest that in all the conditions in which consciousness is lost, the inescapable occurrence of a down-state after an initial activation could break-off cortico-cortical causal interactions (Pigorini et al. NeuroImage 2015), thus impairing the ability of thalamocortical circuits to sustain long-range, deterministic, complex patterns of activation, a theoretical requisite for consciousness (Tononi et al. Nat Rev Neurosci, 2016).
We employ a modeling approach at both the meso and macro levels (thus paralleling respectively intracranial and TMS/EEG recordings) to investigate at what extent peculiar dynamics at the level of single brain areas and cortico-cortical connections are key ingredients to promote complex causal interactions among different cortical areas (Sarasso et al. Clin EEG Neurosci 2014), and thus consciousness.
Title: Emergence of frequency-specific long-range coherence in the neuroanatomical Connectome
Dr. Joana Cabral, Life and Health Sciences Research Institute, University of Minho, Portugal
Abstract:
Despite evermore-detailed characterizations of neuronal activity from the micro- to mesoscopic scale, the principles governing macroscopic oscillatory activity at the system-level remain unclear. I will describe a biophysical mechanism for the transient emergence of macroscopic oscillations from the neuroanatomical network, grounded on universal principles governing the formation of frequency-specific weakly-stable orbits in delay-coupled nonlinear systems. Using a phenomenological network model representing interactions between local field potentials (with intrinsic resonance at 40Hz) in the space-time structure of the human Connectome, we demonstrate the spontaneous formation of a repertoire of collective oscillations organized in space and time and peaking between 0.5-20Hz, explaining spectral, spatial and temporal features of multimodal neuroimaging data.
Title: Computational model of EEG: why we should be bothered
Benedetta Franceschiello & Katharina Glomb, organisers
Abstract: In our presentation we will review the state of the art of computational models in EEG, the major pitfalls and the main outcomes of this research field. We will use the Kuramoto model as a starting point: brain regions are modeled as simple oscillators which are coupled according to the empirically measured anatomical connections between macroscopic brain regions. We will see how the introduction of other modelling scenarios (local field potential, pyramidal cells) on the level of the brain regions could provide insights in understanding EEG dynamics, overcoming limitations that source localization techniques usually have.
Title: Thalamocortical connectivity models account for functional interplay between spectra of extracellular activity in the two areas
Prof. Alberto Mazzoni, BioRobotics Institute, Scuola Superiore Sant'Anna, Pisa, Italy
Abstract: The thalamus is a key step in the processing of sensory stimuli, and plays a relevant role in gating information about the external world, both in sleep and awake state. A deeper understanding of thalamus dynamics in healthy and pathological states could be of help to tackle a number of neurological disorders, such as Tourette's Syndrome and epilepsy. A number of studies attempted at reconstructing the thalamic activity underlying a given cortical activity recorded through EEG, but the knowledge about thalamocortical connectivity is still too limited to fully achieve this result. The talk will present recent modeling studies focusing on the relationship between the global activity of the thalamus, captured with Local Field Potential recordings, and the dominant frequency bands in the cortex. We propose a general set of transmission rules of thalamocortical connections that could help clarify the relationship between the activity observed in the cortex and the associated thalamic drive.