Interpreting neural decoding models using grouped model reliance

Cite This

Files in this item

Checksum: MD5:c8267542384d4efa0f2fa7d56f8dcd6e

VALENTIN, Simon, Maximilian HARKOTTE, Tzvetan POPOV, 2020. Interpreting neural decoding models using grouped model reliance. In: PLoS Computational Biology. Public Library of Science (PLoS). 16(1), e1007148. eISSN 1553-7358. Available under: doi: 10.1371/journal.pcbi.1007148

@article{Valentin2020-01Inter-49265, title={Interpreting neural decoding models using grouped model reliance}, year={2020}, doi={10.1371/journal.pcbi.1007148}, number={1}, volume={16}, journal={PLoS Computational Biology}, author={Valentin, Simon and Harkotte, Maximilian and Popov, Tzvetan}, note={Article Number: e1007148} }

<rdf:RDF xmlns:dcterms="" xmlns:dc="" xmlns:rdf="" xmlns:bibo="" xmlns:dspace="" xmlns:foaf="" xmlns:void="" xmlns:xsd="" > <rdf:Description rdf:about=""> <dcterms:issued>2020-01</dcterms:issued> <dcterms:available rdf:datatype="">2020-04-23T08:55:19Z</dcterms:available> <bibo:uri rdf:resource=""/> <void:sparqlEndpoint rdf:resource="http://localhost/fuseki/dspace/sparql"/> <dc:rights>Attribution 4.0 International</dc:rights> <dc:language>eng</dc:language> <dspace:isPartOfCollection rdf:resource=""/> <dcterms:title>Interpreting neural decoding models using grouped model reliance</dcterms:title> <dcterms:rights rdf:resource=""/> <foaf:homepage rdf:resource="http://localhost:8080/jspui"/> <dcterms:isPartOf rdf:resource=""/> <dc:contributor>Valentin, Simon</dc:contributor> <dcterms:hasPart rdf:resource=""/> <dc:creator>Harkotte, Maximilian</dc:creator> <dspace:hasBitstream rdf:resource=""/> <dc:contributor>Popov, Tzvetan</dc:contributor> <dc:contributor>Harkotte, Maximilian</dc:contributor> <dcterms:abstract xml:lang="eng">Machine learning algorithms are becoming increasingly popular for decoding psychological constructs based on neural data. However, as a step towards bridging the gap between theory-driven cognitive neuroscience and data-driven decoding approaches, there is a need for methods that allow to interpret trained decoding models. The present study demonstrates grouped model reliance as a model-agnostic permutation-based approach to this problem. Grouped model reliance indicates the extent to which a trained model relies on conceptually related groups of variables, such as frequency bands or regions of interest in electroencephalographic (EEG) data. As a case study to demonstrate the method, random forest and support vector machine models were trained on within-participant single-trial EEG data from a Sternberg working memory task. Participants were asked to memorize a sequence of digits (0–9), varying randomly in length between one, four and seven digits, where EEG recordings for working memory load estimation were taken from a 3-second retention interval. The present results confirm previous findings insofar as both random forest and support vector machine models relied on alpha-band activity in most subjects. However, as revealed by further analyses, patterns in frequency and topography varied considerably between individuals, pointing to more pronounced inter-individual differences than previously reported.</dcterms:abstract> <dc:date rdf:datatype="">2020-04-23T08:55:19Z</dc:date> <dc:creator>Valentin, Simon</dc:creator> <dc:creator>Popov, Tzvetan</dc:creator> </rdf:Description> </rdf:RDF>

Downloads since Apr 23, 2020 (Information about access statistics)

Valentin_2-1b2p0oaw030ow3.pdf 87

This item appears in the following Collection(s)

Attribution 4.0 International Except where otherwise noted, this item's license is described as Attribution 4.0 International

Search KOPS


My Account