Multivariate pattern analysis of fMRI: the early beginnings

Neuroimage. 2012 Aug 15;62(2):852-5. doi: 10.1016/j.neuroimage.2012.03.016. Epub 2012 Mar 9.


In 2001, we published a paper on the representation of faces and objects in ventral temporal cortex that introduced a new method for fMRI analysis, which subsequently came to be called multivariate pattern analysis (MVPA). MVPA now refers to a diverse set of methods that analyze neural responses as patterns of activity that reflect the varying brain states that a cortical field or system can produce. This paper recounts the circumstances and events that led to the original study and later developments and innovations that have greatly expanded this approach to fMRI data analysis, leading to its widespread application.

Publication types

  • Historical Article
  • Review

MeSH terms

  • Artificial Intelligence*
  • Brain / physiology
  • Brain Mapping / history
  • Brain Mapping / methods
  • History, 20th Century
  • History, 21st Century
  • Image Processing, Computer-Assisted / history*
  • Image Processing, Computer-Assisted / methods*
  • Magnetic Resonance Imaging / history*
  • Magnetic Resonance Imaging / methods*