research

Learning structural equation models for fMRI

A.J. Storkey, E. Simonotto, H. Whalley, S. Lawrie, L.M. Murray and D. McGonigle

Online

Structural equation models can be seen as an extension of Gaussian belief networks to cyclic graphs, and we show they can be understood generatively as the model for the joint distribution of long term average equilibrium activity of Gaussian dynamic belief networks. Most use of structural equation models in fMRI involves postulating a particular structure and comparing learnt parameters across different groups. In this paper it is argued that there are situations where priors about structure are not firm or exhaustive, and given sufficient data, it is worth investigating learning network structure as part of the approach to connectivity analysis. First we demonstrate structure learning on a toy problem. We then show that for particular fMRI data the simple models usually assumed are not supported. We show that is is possible to learn sensible structural equation models that can provide modelling benefits, but that are not necessarily going to be the same as a true causal model, and suggest the combination of prior models and learning or the use of temporal information from dynamic models may provide more benefits than learning structural equations alone.

A.J. Storkey, E. Simonotto, H. Whalley, S. Lawrie, L.M. Murray and D. McGonigle (2007). Learning structural equation models for fMRI. Advances in Neural Information Processing Systems. 19:1329--1336.

A.J. Storkey, E. Simonotto, H. Whalley, S. Lawrie, L.M. Murray and D. McGonigle (2007). <a href="https://indii.org/research/learning-structural-equation-models-for-fmri/">Learning structural equation models for fMRI</a>. <em>Advances in Neural Information Processing Systems</em>. <strong>19</strong>:1329--1336.

@Article{Storkey2007,
  title = {Learning structural equation models for f{MRI}},
  author = {Amos J. Storkey and Simonotto, E. and Whalley, H. and Lawrie, S. and Lawrence Matthew Murray and David McGonigle},
  journal = {Advances in Neural Information Processing Systems},
  year = {2007},
  volume = {19},
  pages = {1329--1336},
  url = {http://books.nips.cc/papers/files/nips19/NIPS2006_0368.pdf}
}
blog Latest
GPU Programming in the Cloud
How to develop on remote cloud instances, and a roundup of cloud service providers.

Lawrence Murray

22 Nov 22

GPU Programming in the Cloud
blog Related
Matrix Gradients of Scalar Functions
Understanding the building blocks of reverse-mode automatic differentiation.

Lawrence Murray

7 Nov 22

research Next
Continuous Time Particle Filtering for fMRI

L.M. Murray and A. Storkey

software Previous Archive
Tintii
Selective colour and saturation control.
Tintii