EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

Statistical inference on representational geometries

Heiko H Schütt iD, Alexander D Kipnis, Jörn Diedrichsen iD, Nikolaus Kriegeskorte iD

eLife · 23.08.2023 · 12

DOI: 10.7554/elife.82566 ↗

Kurz gefasst

Fachartikel in der Zeitschrift eLife, online erschienen am 23.08.2023. Verfasst von Heiko H Schütt und 3 weiteren Autor:innen. Beteiligte Einrichtung: Zuckerman Institute, Columbia University, New York, United States. Hauptthemen laut MEDLINE: Big Data, Neurosciences. Die Arbeit wurde bisher 48-mal zitiert (Crossref) und nennt 119 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: Deutsche Forschungsgemeinschaft.

AbstractSchlagwörterBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Neuroscience has recently made much progress, expanding the complexity of both neural activity measurements and brain-computational models. However, we lack robust methods for connecting theory and experiment by evaluating our new big models with our new big data. Here, we introduce new inference methods enabling researchers to evaluate and compare models based on the accuracy of their predictions of representational geometries: A good model should accurately predict the distances among the neural population representations (e.g. of a set of stimuli). Our inference methods combine novel 2-factor extensions of crossvalidation (to prevent overfitting to either subjects or conditions from inflating our estimates of model accuracy) and bootstrapping (to enable inferential model comparison with simultaneous generalization to both new subjects and new conditions). We validate the inference methods on data where the ground-truth model is known, by simulating data with deep neural networks and by resampling of calcium-imaging and functional MRI data. Results demonstrate that the methods are valid and conclusions generalize correctly. These data analysis methods are available in an open-source Python toolbox ( rsatoolbox.readthedocs.io ).

Schlagwörter

MeSH-Schlagwörter (MEDLINE)

Big DataNeurosciencesHumansCalcium, DietaryGeneralization, PsychologicalNeural Networks, Computer

Stichwörter der Autor:innen

humanmouseneurosciencerepresentational similarity analysisstatistical inferencetoolbox

Substanzen

Calcium, Dietary

Autor:innen und Institutionen

  • Zuckerman Institute, Columbia University, New York, United States.
  • Western University, London, Canada.

ORCID-Profile: Heiko H Schütt, Jörn Diedrichsen, Nikolaus Kriegeskorte

Förderung

  • Deutsche Forschungsgemeinschaft

Zitieren

APA

Schütt, H. H., Kipnis, A. D., Diedrichsen, J., & Kriegeskorte, N. (2023). Statistical inference on representational geometries. eLife, 12. https://doi.org/10.7554/elife.82566

Vancouver

Schütt HH, Kipnis AD, Diedrichsen J, Kriegeskorte N. Statistical inference on representational geometries. eLife. 2023;12. doi:10.7554/elife.82566

BibTeX
@article{Schtt2023,
  author = {Schütt, Heiko H and Kipnis, Alexander D and Diedrichsen, Jörn and Kriegeskorte, Nikolaus},
  title = {Statistical inference on representational geometries},
  journal = {eLife},
  year = {2023},
  volume = {12},
  doi = {10.7554/elife.82566}
}
Kostenlos starten →