{"about":{"site":"https://codewithpapers.app","non_affiliation":"Code with Papers and Syntology are not affiliated with, endorsed by, or sponsored by Papers with Code, Meta, or the pwc-archive mirror.","licence":"CC BY-SA 4.0","licence_url":"https://creativecommons.org/licenses/by-sa/4.0/legalcode","attribution":"https://codewithpapers.app/attribution","modified":"archive material modified by Syntology; see the attribution page"},"url":"/paper/interpretability-of-multivariate-brain-maps","title":"Interpretability of Multivariate Brain Maps in Brain Decoding: Definition and Quantification","arxiv_id":"1603.08704","date":"2016-03-29","proceeding":null,"authors":["Seyed Mostafa Kia"],"abstract":"Brain decoding is a popular multivariate approach for hypothesis testing in\nneuroimaging. It is well known that the brain maps derived from weights of\nlinear classifiers are hard to interpret because of high correlations between\npredictors, low signal to noise ratios, and the high dimensionality of\nneuroimaging data. Therefore, improving the interpretability of brain decoding\napproaches is of primary interest in many neuroimaging studies. Despite\nextensive studies of this type, at present, there is no formal definition for\ninterpretability of multivariate brain maps. As a consequence, there is no\nquantitative measure for evaluating the interpretability of different brain\ndecoding methods. In this paper, first, we present a theoretical definition of\ninterpretability in brain decoding; we show that the interpretability of\nmultivariate brain maps can be decomposed into their reproducibility and\nrepresentativeness. Second, as an application of the proposed theoretical\ndefinition, we formalize a heuristic method for approximating the\ninterpretability of multivariate brain maps in a binary magnetoencephalography\n(MEG) decoding scenario. Third, we propose to combine the approximated\ninterpretability and the performance of the brain decoding model into a new\nmulti-objective criterion for model selection. Our results for the MEG data\nshow that optimizing the hyper-parameters of the regularized linear classifier\nbased on the proposed criterion results in more informative multivariate brain\nmaps. More importantly, the presented definition provides the theoretical\nbackground for quantitative evaluation of interpretability, and hence,\nfacilitates the development of more effective brain decoding algorithms in the\nfuture.","url_abs":"http://arxiv.org/abs/1603.08704v1","url_pdf":"http://arxiv.org/pdf/1603.08704v1.pdf","source":{"archive":"pwc-archive (Hugging Face), CC BY-SA 4.0","snapshot":"2025-07-28","licence_url":"https://creativecommons.org/licenses/by-sa/4.0/legalcode","row_kind":"abstracts"},"code_links":[{"paper_slug":"interpretability-of-multivariate-brain-maps","repo_url":"https://github.com/smkia/interpretability","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":0,"framework":"none","reach":null}],"tasks":[{"task_slug":"brain-decoding","task_name":"Brain Decoding"},{"task_slug":"model-selection","task_name":"Model Selection"},{"task_slug":"hypothesis-testing","task_name":"Two-sample testing"}],"methods":[{"method_slug":"interpretability","method_name":"Interpretability"}],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"atlas_url":null,"mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}