Looking outside the searchlight

Joset A. Etzel; Michael W. Cole; Todd S. Braver

doi:10.1007/978-3-642-34713-9_4

Looking outside the searchlight

Joset A. Etzel, Michael W. Cole, Todd S. Braver

Research output: Contribution to journal › Conference article › peer-review

4 Scopus citations

Abstract

Searchlight analysis (information mapping) with pattern classifiers is a popular method of multivariate fMRI analysis often interpreted as localizing informative voxel clusters. Applicability and utility of this method is limited, however, by its dependency on searchlight radius, the assumption that information is present at all spatial scales, and its susceptibility to overfitting. These problems are demonstrated in a dataset in which, contrary to common expectation, voxels identified as informative do not clearly contain more information than those not so identified.

Original language	English
Pages (from-to)	26-33
Number of pages	8
Journal	Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
Volume	7263 LNAI
DOIs	https://doi.org/10.1007/978-3-642-34713-9_4
State	Published - 2012
Event	International Workshop on Machine Learning and Interpretation in Neuroimaging, MLINI 2011, Held at Neural Information Processing, NIPS 2011 - Sierra Nevada, Spain Duration: Dec 16 2011 → Dec 17 2011

Keywords

fMRI
information mapping
MVPA
searchlight analysis

Access to Document

10.1007/978-3-642-34713-9_4

Cite this

@article{394811cc87874e509ddfb4dcb143a294,

title = "Looking outside the searchlight",

abstract = "Searchlight analysis (information mapping) with pattern classifiers is a popular method of multivariate fMRI analysis often interpreted as localizing informative voxel clusters. Applicability and utility of this method is limited, however, by its dependency on searchlight radius, the assumption that information is present at all spatial scales, and its susceptibility to overfitting. These problems are demonstrated in a dataset in which, contrary to common expectation, voxels identified as informative do not clearly contain more information than those not so identified.",

keywords = "fMRI, information mapping, MVPA, searchlight analysis",

author = "Etzel, {Joset A.} and Cole, {Michael W.} and Braver, {Todd S.}",

year = "2012",

doi = "10.1007/978-3-642-34713-9_4",

language = "English",

volume = "7263 LNAI",

pages = "26--33",

journal = "Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)",

issn = "0302-9743",

note = "International Workshop on Machine Learning and Interpretation in Neuroimaging, MLINI 2011, Held at Neural Information Processing, NIPS 2011 ; Conference date: 16-12-2011 Through 17-12-2011",

}

TY - JOUR

T1 - Looking outside the searchlight

AU - Etzel, Joset A.

AU - Cole, Michael W.

AU - Braver, Todd S.

PY - 2012

Y1 - 2012

N2 - Searchlight analysis (information mapping) with pattern classifiers is a popular method of multivariate fMRI analysis often interpreted as localizing informative voxel clusters. Applicability and utility of this method is limited, however, by its dependency on searchlight radius, the assumption that information is present at all spatial scales, and its susceptibility to overfitting. These problems are demonstrated in a dataset in which, contrary to common expectation, voxels identified as informative do not clearly contain more information than those not so identified.

AB - Searchlight analysis (information mapping) with pattern classifiers is a popular method of multivariate fMRI analysis often interpreted as localizing informative voxel clusters. Applicability and utility of this method is limited, however, by its dependency on searchlight radius, the assumption that information is present at all spatial scales, and its susceptibility to overfitting. These problems are demonstrated in a dataset in which, contrary to common expectation, voxels identified as informative do not clearly contain more information than those not so identified.

KW - fMRI

KW - information mapping

KW - MVPA

KW - searchlight analysis

UR - http://www.scopus.com/inward/record.url?scp=84870046011&partnerID=8YFLogxK

U2 - 10.1007/978-3-642-34713-9_4

DO - 10.1007/978-3-642-34713-9_4

M3 - Conference article

AN - SCOPUS:84870046011

SN - 0302-9743

VL - 7263 LNAI

SP - 26

EP - 33

JO - Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)

JF - Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)

T2 - International Workshop on Machine Learning and Interpretation in Neuroimaging, MLINI 2011, Held at Neural Information Processing, NIPS 2011

Y2 - 16 December 2011 through 17 December 2011

ER -

Looking outside the searchlight

Abstract

Keywords

Access to Document

Other files and links

Fingerprint

Cite this