How measurement noise limits the accuracy of brain-behaviour predictions

Abstract Major efforts in human neuroimaging strive to understand individual differences and find biomarkers for clinical applications by predicting behavioural phenotypes from brain imaging data. To identify generalisable and replicable brain-behaviour prediction models, sufficient measurement reli...

Full description

Saved in:
Bibliographic Details
Main Authors: Martin Gell, Simon B. Eickhoff, Amir Omidvarnia, Vincent Küppers, Kaustubh R. Patil, Theodore D. Satterthwaite, Veronika I. Müller, Robert Langner
Format: Article
Language:English
Published: Nature Portfolio 2024-12-01
Series:Nature Communications
Online Access:https://doi.org/10.1038/s41467-024-54022-6
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1850119002715586560
author Martin Gell
Simon B. Eickhoff
Amir Omidvarnia
Vincent Küppers
Kaustubh R. Patil
Theodore D. Satterthwaite
Veronika I. Müller
Robert Langner
author_facet Martin Gell
Simon B. Eickhoff
Amir Omidvarnia
Vincent Küppers
Kaustubh R. Patil
Theodore D. Satterthwaite
Veronika I. Müller
Robert Langner
author_sort Martin Gell
collection DOAJ
description Abstract Major efforts in human neuroimaging strive to understand individual differences and find biomarkers for clinical applications by predicting behavioural phenotypes from brain imaging data. To identify generalisable and replicable brain-behaviour prediction models, sufficient measurement reliability is essential. However, the selection of prediction targets is predominantly guided by scientific interest or data availability rather than psychometric considerations. Here, we demonstrate the impact of low reliability in behavioural phenotypes on out-of-sample prediction performance. Using simulated and empirical data from four large-scale datasets, we find that reliability levels common across many phenotypes can markedly limit the ability to link brain and behaviour. Next, using 5000 participants from the UK Biobank, we show that only highly reliable data can fully benefit from increasing sample sizes from hundreds to thousands of participants. Our findings highlight the importance of measurement reliability for identifying meaningful brain–behaviour associations from individual differences and underscore the need for greater emphasis on psychometrics in future research.
format Article
id doaj-art-5a2f9505de814dae81bea403199aac78
institution OA Journals
issn 2041-1723
language English
publishDate 2024-12-01
publisher Nature Portfolio
record_format Article
series Nature Communications
spelling doaj-art-5a2f9505de814dae81bea403199aac782025-08-20T02:35:44ZengNature PortfolioNature Communications2041-17232024-12-0115111210.1038/s41467-024-54022-6How measurement noise limits the accuracy of brain-behaviour predictionsMartin Gell0Simon B. Eickhoff1Amir Omidvarnia2Vincent Küppers3Kaustubh R. Patil4Theodore D. Satterthwaite5Veronika I. Müller6Robert Langner7Department of Psychiatry, Psychotherapy and Psychosomatics, Medical Faculty, RWTH Aachen UniversityInstitute of Neuroscience and Medicine (INM-7: Brain & Behaviour), Research Centre JülichInstitute of Neuroscience and Medicine (INM-7: Brain & Behaviour), Research Centre JülichInstitute of Neuroscience and Medicine (INM-7: Brain & Behaviour), Research Centre JülichInstitute of Neuroscience and Medicine (INM-7: Brain & Behaviour), Research Centre JülichDepartment of Psychiatry, Perelman School of Medicine, Penn Lifespan Informatics and Neuroimaging Center, University of PennsylvaniaInstitute of Neuroscience and Medicine (INM-7: Brain & Behaviour), Research Centre JülichInstitute of Neuroscience and Medicine (INM-7: Brain & Behaviour), Research Centre JülichAbstract Major efforts in human neuroimaging strive to understand individual differences and find biomarkers for clinical applications by predicting behavioural phenotypes from brain imaging data. To identify generalisable and replicable brain-behaviour prediction models, sufficient measurement reliability is essential. However, the selection of prediction targets is predominantly guided by scientific interest or data availability rather than psychometric considerations. Here, we demonstrate the impact of low reliability in behavioural phenotypes on out-of-sample prediction performance. Using simulated and empirical data from four large-scale datasets, we find that reliability levels common across many phenotypes can markedly limit the ability to link brain and behaviour. Next, using 5000 participants from the UK Biobank, we show that only highly reliable data can fully benefit from increasing sample sizes from hundreds to thousands of participants. Our findings highlight the importance of measurement reliability for identifying meaningful brain–behaviour associations from individual differences and underscore the need for greater emphasis on psychometrics in future research.https://doi.org/10.1038/s41467-024-54022-6
spellingShingle Martin Gell
Simon B. Eickhoff
Amir Omidvarnia
Vincent Küppers
Kaustubh R. Patil
Theodore D. Satterthwaite
Veronika I. Müller
Robert Langner
How measurement noise limits the accuracy of brain-behaviour predictions
Nature Communications
title How measurement noise limits the accuracy of brain-behaviour predictions
title_full How measurement noise limits the accuracy of brain-behaviour predictions
title_fullStr How measurement noise limits the accuracy of brain-behaviour predictions
title_full_unstemmed How measurement noise limits the accuracy of brain-behaviour predictions
title_short How measurement noise limits the accuracy of brain-behaviour predictions
title_sort how measurement noise limits the accuracy of brain behaviour predictions
url https://doi.org/10.1038/s41467-024-54022-6
work_keys_str_mv AT martingell howmeasurementnoiselimitstheaccuracyofbrainbehaviourpredictions
AT simonbeickhoff howmeasurementnoiselimitstheaccuracyofbrainbehaviourpredictions
AT amiromidvarnia howmeasurementnoiselimitstheaccuracyofbrainbehaviourpredictions
AT vincentkuppers howmeasurementnoiselimitstheaccuracyofbrainbehaviourpredictions
AT kaustubhrpatil howmeasurementnoiselimitstheaccuracyofbrainbehaviourpredictions
AT theodoredsatterthwaite howmeasurementnoiselimitstheaccuracyofbrainbehaviourpredictions
AT veronikaimuller howmeasurementnoiselimitstheaccuracyofbrainbehaviourpredictions
AT robertlangner howmeasurementnoiselimitstheaccuracyofbrainbehaviourpredictions