Jove
Visualize
Contact Us
JoVE
x logofacebook logolinkedin logoyoutube logo
ABOUT JoVE
OverviewLeadershipBlogJoVE Help Center
AUTHORS
Publishing ProcessEditorial BoardScope & PoliciesPeer ReviewFAQSubmit
LIBRARIANS
TestimonialsSubscriptionsAccessResourcesLibrary Advisory BoardFAQ
RESEARCH
JoVE JournalMethods CollectionsJoVE Encyclopedia of ExperimentsArchive
EDUCATION
JoVE CoreJoVE BusinessJoVE Science EducationJoVE Lab ManualFaculty Resource CenterFaculty Site
Terms & Conditions of Use
Privacy Policy
Policies

Related Experiment Videos

Doubting these doubts--a reply to Collet.

J Möcks1

  • 1Institut für Angewandte Mathematik, Universität Heidelberg, F.R.G.

Biological Psychology
|April 1, 1989
PubMed
Summary

Principal component analysis (PCA) for event-related potentials remains valid despite criticisms. PCA

Related Concept Videos

You might also read

Related Articles

Articles linked to this work by shared authors, journal, and citation graph.

Sort by
Same author

Effect of insulin analogs on the decline of hemoglobin in diabetic patients with nephropathy.

Experimental and clinical endocrinology & diabetes : official journal, German Society of Endocrinology [and] German Diabetes Association·2010
Same author

Potential of mid-infrared spectroscopy to aid the triage of patients with acute chest pain.

The Analyst·2009
Same author

Comparison of mid-infrared and Raman spectroscopy in the quantitative analysis of serum.

Journal of biomedical optics·2005
Same author

Cardiovascular mortality in haemodialysis patients treated with epoetin beta - a retrospective study.

Nephron·2000
Same author

The safety of treatment with recombinant human erythropoietin in clinical use: a review of controlled studies.

International journal of molecular medicine·1998
Same author

Testing and estimation for variable single evoked brain potentials.

IEEE engineering in medicine and biology magazine : the quarterly magazine of the Engineering in Medicine & Biology Society·1990

Area of Science:

  • Neuroscience
  • Cognitive Science
  • Psychophysiology

Background:

  • Critiques have questioned the validity of principal component analysis (PCA) when applied to event-related potentials (ERPs).
  • Collet's (1989) analysis, based solely on the correlation matrix, was presented as a challenge to PCA's assumptions.
  • This response addresses specific points raised in Collet's comment regarding PCA's application in ERP research.

Discussion:

  • Collet's critique is limited as it relies only on the correlation matrix, failing to address PCA assumptions independent of this matrix.
  • The model proposed by Collet incorporates PCA results, yet its comparative analysis and parameter count are consequently invalidated.
  • Collet's model lacks predictive power and fails to achieve genuine data reduction, contradicting the principle of parsimony.

Key Insights:

  • PCA's core assumptions are not disproven by analyses confined to the correlation matrix.
  • Collet's model is less general than PCA, potentially applicable only to slow brain potentials, not broader ERP components.
  • PCA offers superior data reduction and generalizability for analyzing complex electrophysiological data.

Outlook:

  • Further validation of PCA's utility in diverse electrophysiological paradigms is warranted.
  • Refining models that integrate PCA with other analytical techniques can enhance understanding of brain activity.
  • Continued methodological debate will refine best practices for analyzing event-related potentials.

Related Experiment Videos