Gast ::
| Hauptseite > Publikationsdatenbank > Analysis and quantitative comparison of neural network dynamics on a neuron-wise and population level |
| Hauptseite > Publikationsdatenbank > Analysis and quantitative comparison of neural network dynamics on a neuron-wise and population level |
| Book/Dissertation / PhD Thesis | FZJ-2024-01955 |
2024
Forschungszentrum Jülich GmbH Zentralbibliothek, Verlag
Jülich
ISBN: 978-3-95806-738-7
This record in other databases:
Please use a persistent id in citations: urn:nbn:de:0001-20240516102957373-1285584-9 doi:10.34734/FZJ-2024-01955
Abstract: Our goal is to better understand the working mechanisms of biologicalneural systems. To this end, describing neural systems as networksprovides a powerful and widely-used analysis approach. The networksyntax of interacting nodes exhibiting joint dynamics facilitates thequantitative characterizations of neural systems across scales. More-over, this approach enables us to construct systematic comparisons ofneural network descriptions across domains.We aim to identify characterizations of neural network activity thatreflect the underlying connectivity and relate to the network’s abil-ity to process information. In this context, we explore characteristicmeasures from experimental and simulated data sources. Concretely,we look at cortical activity data from mice and monkeys, from dif-ferent recording techniques like implanted electrode arrays, laminarprobes, ECoG, and calcium imaging, and further from simulationsof stochastic processes, spiking, and mean-field network models. Weinvestigate activity measures of different complexity, including mea-sures on the level of individual neurons, higher-order measures ofcoordinated spiking activity, and population-level field potential mea-sures describing spatial wave patterns. Such activity characterizationsalways represent an abstraction, and the right level of detail dependson the data type and the question of interest. For a given context, theappropriate abstraction level allows us to integrate and compare dataand models from heterogeneous sources.Evaluating the similarity between such different network descrip-tions is a common demand in computational neuroscience. Extendingthe concept of validation, we formalize and apply cross-domain com-parisons in model vs. experiment, model vs. model, and experiment vs.experiment scenarios. In this framework, we further evaluate and ex-tend existing statistical testing approaches and look at reproducibility,sources of variability, and technical limitations.Through our exploration of network activity characterizations andtheir comparability, we evaluate the relationship between networkconnectivity, activity, and function. Concretely, over the course offive research projects, we implement and demonstrate systematic ap-proaches to validate model simulators, statistically evaluate networkorganization, infer network connectivity from activity data, combinedata sources of wave activity, and relate wave activity to external in-fluences and behavior. With a focus on open and collaborative sciencepractices, we implement our methodologies as reusable open-sourcetools while building upon existing open-source tools and standards.
; 
|
The record appears in these collections: |
PDF
Fulltext by OpenAccess repository