Reply to Friston and David: After comments on: The identification of interacting networks in the brain using fMRI: Model selection, causality and deconvolution
暂无分享,去创建一个
l rights reserved. conditional dependencies between the parameters, i.e., the degree of redundancy in the parameterization of the differential equations and ii) the sensitivity of the model output to changes in the parameters, which is itself a function of the complexity of the input, i.e., the experimental design (Deneux and Faugeras, 2006). In this context, the sufficient degrees of freedom in a hemodynamic model in order for neuronal parameters to remain unaffected can also be interpreted as themaximum identifiable complexity (and inevitably: veridicality) that still allows the parameters of the model to be uniquely estimated in practice. An important challenge lies in the usage of different imaging modalities (possibly simultaneously) to increase the complexity and realism of connectivity models that can be identified and compared (Valdes-Sosa et al., 2009).
[1] Jose M. Sanchez-Bornot,et al. Model driven EEG/fMRI fusion of brain oscillations , 2009, Human brain mapping.
[2] Olivier Faugeras,et al. Using nonlinear models in fMRI data analysis: Model selection and activation detection , 2006, NeuroImage.