PLS and Functional Neuroimaging: Bias and Detection Power Across Different Resampling Schemes

  • Nathan ChurchillEmail author
  • Babak Afshin-Pour
  • Stephen Strother
Conference paper
Part of the Springer Proceedings in Mathematics & Statistics book series (PROMS, volume 173)


Correlation Partial-Least Squares (PLSC) provides a robust model for analyzing functional neuroimaging data, which is used to identify functional brain networks that show the largest covariance with task stimuli. However, neuroimaging data tend to be high-dimensional (i.e., there are far more variables P than samples N), with significant noise confounds and variability in brain response. It is therefore challenging to identify the significant, stable components of PLSC analysis. Empirical significance estimators are widely used, as they make minimal assumptions about data structure. The most common estimator in neuroimaging PLS is Bootstrapped Variance (BV), which tests whether bootstrap-stabilized mean component eigenvalues (i.e., covariance) are significantly different from a permuted null distribution; however, recent studies have highlighted issues with this model. Two alternatives were proposed that instead focus on reliability of the PLSC saliences (i.e., singular vectors): a Split-half Stability (SS) model that measures the consistency of reconstructed components for split-half data, and Split-half Reproducibility (SR) which measures the reliability across independent split-half analyses. We compare BV, SS, and SR estimators on functional Magnetic Resonance Imaging (f MRI) data, for both simulated and experimental datasets. The SS and SR methods have comparable sensitivity in detecting “brain” components for most simulated and experimental conditions. However, SR shows consistently greater sensitivity for “task” components. We demonstrate that this is due to relative bias in the SS model: both “brain” and “task” components have biased null distributions, but for the low-dimensional “task” vectors, this bias becomes sufficiently high that it is often impossible to distinguish a significant effect from the null distribution.


fmri Behavioral pls Bootstrap Split-half resampling Prediction Reproducibility pca 


  1. Abdi, H.: Partial least squares regression and projection on latent structure regression (PLS Regression). Wiley Interdiscip. Rev.: Comput. Stat. 2, 97–106 (2010)Google Scholar
  2. Abdi, H., Williams, L.: Partial least square methods: partial least square correlation and partial least square regression. In: Reisfeld, B., Mayeno, A. (eds.) Methods in Molecular Biology: Computational Toxicology, pp. 549–579. Springer, New York (2012)Google Scholar
  3. Abdi, H., Chin, W., Esposito Vinzi, V., Russolilo, G., Trinchera, L.: New Perspectives in PLS and Related Methods. Springer, New York (2013)zbMATHGoogle Scholar
  4. Churchill, N.W., Spring, R., Kovacevic, N., McIntosh, A.R., Strother, S.C.: The stability of behavioral PLS results in ill-posed neuroimaging problems. In: Abdi, H., Chin, W., Esposito Vinzi, V., Russolilo, G., Trinchera, L. (eds.) New Perspectives in PLS and Related Methods, pp. 171–183. Springer, New York (2013)Google Scholar
  5. Kovacevic, N., et al.: Revisiting PLS resampling: comparing significance vs. reliability across range of simulations. In: Abdi, H., Chin, W., Esposito Vinzi, V., Russolilo, G., Trinchera, L. (eds.) New Perspectives in PLS and Related Methods, pp. 159–170. Springer, New York (2013)Google Scholar
  6. Krishnan, A., Williams, L.J., McIntosh, A.R., Abdi, H.: Partial least squares (PLS) methods for neuroimaging: a tutorial and review. NeuroImage 56, 455–475 (2011)CrossRefGoogle Scholar
  7. Lukic, A.S., Wernick, M.N., Strother, S.C.: An evaluation of methods for detecting brain activations from functional neuroimages. Artif Intell. Med. 25, 69–88 (2002)CrossRefGoogle Scholar
  8. McIntosh, A.R., et al.: Spatial pattern analysis of functional brain images using partial least squares. NeuroImage 3, 143–157 (1996)CrossRefGoogle Scholar
  9. Rosipal, R., Krämer, N.: Overview and recent advances in partial least squares. Subspace, latent structure and feature selection. In: Saunders, C. (ed.) Proceedings SLSFS, LNCS. Springer, Berlin (2006)Google Scholar
  10. Strother, S.C., Anderson, J., Hansen, L.K., Kjems, U., Kustra, R., Sidtis, J., Frutiger, S., Muley, S., LaConte, S., Rottenberg, D.: The quantitative evaluation of functional neuroimaging experiments: the npairs data analysis framework. Neuroimage 15, 747–771 (2002)CrossRefGoogle Scholar
  11. Tam, F., Churchill, N.W., Strother, S.C., Graham, S.J.: A new tablet for writing and drawing during functional MRI. Hum. Brain Mapp. 32, 240–248 (2011)CrossRefGoogle Scholar
  12. Wold, H.: Partial Least Squares. Encyclopedia of Statistical Sciences, vol. 6, pp. 581–591. Wiley, New York (1985)Google Scholar
  13. Yourganov, G., et al.: Dimensionality estimation for optimal detection of functional networks in BOLD fMRI data. NeuroImage 56, 531–543 (2011)CrossRefGoogle Scholar

Copyright information

© Springer International Publishing Switzerland 2016

Authors and Affiliations

  • Nathan Churchill
    • 1
    Email author
  • Babak Afshin-Pour
    • 2
  • Stephen Strother
    • 2
  1. 1.Li Ka Shing Knowledge InstituteTorontoCanada
  2. 2.Rotman Research InstituteBaycrest HospitalTorontoCanada

Personalised recommendations