CoSMoMVPA philosophy

Uniform representation of data

(This is inspired by PyMVPA).

_images/cosmo_dataset_io.png

Illustration of dataset input/output.

Store data in matrix form along with additional attributes of the data

  • each row in the matrix is a sample, also referred to as pattern.
  • each column in the matrix has data for a single feature (e.g. voxel, surface node, timepoint for a channel).
  • each row and column can have additional attributes, called sample attributes (such as condition or acquisition run number) and feature attributes (such as voxel location)
  • subsets of rows and columns (and the corresponding attributes) can be split or selected (optionally after matching), and combined), which makes many operations simple:
_images/cosmo_dataset.png

Illustration of dataset structure

_images/slice_sa.png

Illustration of slicing rows (samples). In the masks, white elements are selected and black elements are not selected.

_images/slice_fa.png

Illustration of slicing columns (features). In the masks, white elements are selected and black elements are not selected.

A searchlight analysis is as easy as ROI analysis

  • a searchlight is like a repeated ROI analysis, where data in each searchlight can be described by a neighborhood of features around a center feature.
  • this approach can be applied equally to volume-based fMRI, surface-based fMRI, and MEEG data (assuming a neighborhood can be defined for each feature).
  • a searchlight map is created simply by applying an MVPA measure to data in each searchlight.