A new statistical method for testing hypotheses of neuropsychological/MRI relationships in schizophrenia: Partial least squares analysis

Paul G. Nestor, Brian O'Donnell, Robert W. McCarley, Margaret Niznikiewicz, John Barnard, Zi Jen Shen, Fred L. Bookstein, Martha E. Shenton

Research output: Contribution to journalArticle

38 Citations (Scopus)

Abstract

We applied partial least squares (PLS) as a novel multivariate statistical technique to examine neuropsychological correlates of magnetic resonance imaging (MRI) measures of brain volumes in a well studied sample of 15 male patients with chronic schizophrenia. In the current study, because the total number of measures far surpassed the total number of subjects, extant multivariate techniques such as canonical correlation could not be used to examine relationships among simultaneous measures of MRI and neuropsychology. Moreover, because MRI measures were expected to be highly inter-correlated, as would be neuropsychological test scores, extant multivariate statistical techniques would be substantially limited because they typically assume statistical independence among sets of measures. PLS, on the other hand, proved to be especially well suited to examining the relationships among function and anatomy measures in this sample, where statistically significant relationships were demonstrated that were entirely consistent with prior studies using univariate correlation techniques. In particular, statistically significant relationships emerged among sets of MRI temporal lobe measures and neuropsychological tests of verbal memory and categorization as well as among MRI frontal measures and neuropsychological tests of working memory.

Original languageEnglish
Pages (from-to)57-66
Number of pages10
JournalSchizophrenia Research
Volume53
Issue number1-2
DOIs
StatePublished - Jan 1 2002

Fingerprint

Least-Squares Analysis
Schizophrenia
Magnetic Resonance Imaging
Neuropsychological Tests
Neuropsychology
Temporal Lobe
Short-Term Memory
Anatomy
Brain

Keywords

  • Magnetic resonance imaging
  • Neuropsychology
  • Partial least squares
  • Schizophrenia

ASJC Scopus subject areas

  • Psychiatry and Mental health
  • Behavioral Neuroscience
  • Biological Psychiatry
  • Neurology
  • Psychology(all)

Cite this

A new statistical method for testing hypotheses of neuropsychological/MRI relationships in schizophrenia : Partial least squares analysis. / Nestor, Paul G.; O'Donnell, Brian; McCarley, Robert W.; Niznikiewicz, Margaret; Barnard, John; Shen, Zi Jen; Bookstein, Fred L.; Shenton, Martha E.

In: Schizophrenia Research, Vol. 53, No. 1-2, 01.01.2002, p. 57-66.

Research output: Contribution to journalArticle

Nestor, Paul G. ; O'Donnell, Brian ; McCarley, Robert W. ; Niznikiewicz, Margaret ; Barnard, John ; Shen, Zi Jen ; Bookstein, Fred L. ; Shenton, Martha E. / A new statistical method for testing hypotheses of neuropsychological/MRI relationships in schizophrenia : Partial least squares analysis. In: Schizophrenia Research. 2002 ; Vol. 53, No. 1-2. pp. 57-66.
@article{852cadd5f54d48ecb0ed17a73253be4d,
title = "A new statistical method for testing hypotheses of neuropsychological/MRI relationships in schizophrenia: Partial least squares analysis",
abstract = "We applied partial least squares (PLS) as a novel multivariate statistical technique to examine neuropsychological correlates of magnetic resonance imaging (MRI) measures of brain volumes in a well studied sample of 15 male patients with chronic schizophrenia. In the current study, because the total number of measures far surpassed the total number of subjects, extant multivariate techniques such as canonical correlation could not be used to examine relationships among simultaneous measures of MRI and neuropsychology. Moreover, because MRI measures were expected to be highly inter-correlated, as would be neuropsychological test scores, extant multivariate statistical techniques would be substantially limited because they typically assume statistical independence among sets of measures. PLS, on the other hand, proved to be especially well suited to examining the relationships among function and anatomy measures in this sample, where statistically significant relationships were demonstrated that were entirely consistent with prior studies using univariate correlation techniques. In particular, statistically significant relationships emerged among sets of MRI temporal lobe measures and neuropsychological tests of verbal memory and categorization as well as among MRI frontal measures and neuropsychological tests of working memory.",
keywords = "Magnetic resonance imaging, Neuropsychology, Partial least squares, Schizophrenia",
author = "Nestor, {Paul G.} and Brian O'Donnell and McCarley, {Robert W.} and Margaret Niznikiewicz and John Barnard and Shen, {Zi Jen} and Bookstein, {Fred L.} and Shenton, {Martha E.}",
year = "2002",
month = "1",
day = "1",
doi = "10.1016/S0920-9964(00)00171-7",
language = "English",
volume = "53",
pages = "57--66",
journal = "Schizophrenia Research",
issn = "0920-9964",
publisher = "Elsevier",
number = "1-2",

}

TY - JOUR

T1 - A new statistical method for testing hypotheses of neuropsychological/MRI relationships in schizophrenia

T2 - Partial least squares analysis

AU - Nestor, Paul G.

AU - O'Donnell, Brian

AU - McCarley, Robert W.

AU - Niznikiewicz, Margaret

AU - Barnard, John

AU - Shen, Zi Jen

AU - Bookstein, Fred L.

AU - Shenton, Martha E.

PY - 2002/1/1

Y1 - 2002/1/1

N2 - We applied partial least squares (PLS) as a novel multivariate statistical technique to examine neuropsychological correlates of magnetic resonance imaging (MRI) measures of brain volumes in a well studied sample of 15 male patients with chronic schizophrenia. In the current study, because the total number of measures far surpassed the total number of subjects, extant multivariate techniques such as canonical correlation could not be used to examine relationships among simultaneous measures of MRI and neuropsychology. Moreover, because MRI measures were expected to be highly inter-correlated, as would be neuropsychological test scores, extant multivariate statistical techniques would be substantially limited because they typically assume statistical independence among sets of measures. PLS, on the other hand, proved to be especially well suited to examining the relationships among function and anatomy measures in this sample, where statistically significant relationships were demonstrated that were entirely consistent with prior studies using univariate correlation techniques. In particular, statistically significant relationships emerged among sets of MRI temporal lobe measures and neuropsychological tests of verbal memory and categorization as well as among MRI frontal measures and neuropsychological tests of working memory.

AB - We applied partial least squares (PLS) as a novel multivariate statistical technique to examine neuropsychological correlates of magnetic resonance imaging (MRI) measures of brain volumes in a well studied sample of 15 male patients with chronic schizophrenia. In the current study, because the total number of measures far surpassed the total number of subjects, extant multivariate techniques such as canonical correlation could not be used to examine relationships among simultaneous measures of MRI and neuropsychology. Moreover, because MRI measures were expected to be highly inter-correlated, as would be neuropsychological test scores, extant multivariate statistical techniques would be substantially limited because they typically assume statistical independence among sets of measures. PLS, on the other hand, proved to be especially well suited to examining the relationships among function and anatomy measures in this sample, where statistically significant relationships were demonstrated that were entirely consistent with prior studies using univariate correlation techniques. In particular, statistically significant relationships emerged among sets of MRI temporal lobe measures and neuropsychological tests of verbal memory and categorization as well as among MRI frontal measures and neuropsychological tests of working memory.

KW - Magnetic resonance imaging

KW - Neuropsychology

KW - Partial least squares

KW - Schizophrenia

UR - http://www.scopus.com/inward/record.url?scp=0036133149&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=0036133149&partnerID=8YFLogxK

U2 - 10.1016/S0920-9964(00)00171-7

DO - 10.1016/S0920-9964(00)00171-7

M3 - Article

C2 - 11728838

AN - SCOPUS:0036133149

VL - 53

SP - 57

EP - 66

JO - Schizophrenia Research

JF - Schizophrenia Research

SN - 0920-9964

IS - 1-2

ER -