Citation Context

...ed directly using fMRI. The outputs are the regional hemodynamic responses, which are connected to the neuronal state variables using a biophysically validated forward model of hemodynamic responses (=-=Stephan, Weiskopf, Drysdale, Robinson, & Friston, 2007-=-). The central idea of DCM is to model changes in the states such that the predicted BOLD signal corresponds as closely as possible to the observed BOLD time series (Stephan, Penny, Marshall, Fink, & ...

Citation Context

...te, bilinear neurodynamical model as described above. Model predictions In DCM, neuronal activity gives rise to fMRI signals via an extended Balloon model (Buxton et al., 2004) and BOLD signal model (=-=Stephan et al., 2007-=-) for each region. This specifies how changes in neuronal activity give rise to changes in blood oxygenation that are measured with fMRI. The equations for these hemodynamics are provided in the Appen...

Citation Context

...äläinen et al., 1993). Some research does not exploit the connection provided by the HFM, treating the input to the model as stimulus time course, rather than neuronal activity (Friston et al., 2003; =-=Stephan et al., 2007-=-; Johnston et al., 2008). When, however, the connection is exploited it is possible to estimate the unobserved neural activity, such as local field potential (Mukamel et al., 2005), according to the H...

Citation Context

..., where each DCM corresponds to a specific hypothesis about functional brain architectures (e.g., Acs and Greenlee, 2008; Allen et al., 2008; Grol et al., 2007; Heim et al., 2009; Smith et al., 2006; =-=Stephan et al., 2007-=-; Summerfield and Koechlin, 2008). Although DCM is probably best known through its application to fMRI, most recent advances have focused on neurobiologically plausible models of electrophysiological ...

Citation Context

...inputs (for examples, see Kumar et al., 2007; Leff et al., 2008; Stephan et al., 2007b). A final option is to structure model space in a factorial fashion (Chen et al., 2009; Daunizeau et al., 2009b; =-=Stephan et al., 2007c-=-). The advantage of this approach is that it suggests a natural partitioning of model space into equally large subsets or families of models, which, as described below, can be compared collectively. I...

Citation Context

...dels in which the prior variances of the coupling parameters in DCM were independent of the anatomical likelihood of the corresponding connections. Using Bayesian model selection (Penny et al., 2004; =-=Stephan et al., 2007b-=-), we tested which of these models best explained experimentally measured fMRI data in a DCM of four visual areas (Stephan et al., 2007a), using data from a group of twelve healthy volunteers. Our com...

Citation Context

... significant difference in favour of the FNBN model (pb0.047 and pb0.008, respectively). The lower log-evidence for the cFNBN model provides another interesting demonstration (c.f., Grol et al. 2007; =-=Stephan et al. 2007b-=-), that increasing the complexity of a model does not necessarily improve it. In conclusion, we found that the model with nonlinear forward and backward connections was the best model and that the mod...

Citation Context

...y model space partitioning Finally, we revisited a comparison of DCMs, which were identical in network architecture (the same as m 1 in Fig. 8) but differed in the hemodynamic forward model employed (=-=Stephan et al., 2007c-=-). A three-factor design was used to construct 8 different models: (i) nonlinear vs. linear BOLD equations, (ii) classical vs. revised coefficients of the BOLD equation, and (iii) free vs. fixed param...

Citation Context

...in the usual way. Comparing DCMs in a way that is independent of their parameters is useful when trying to identify the most plausible architectures subtending observed responses (Penny et al., 2004; =-=Stephan et al., 2007-=-) and is used extensively in subsequent sections. The focus of this paper is on the approximate log-evidence 1n p(gc |m) and conditional densities p(ϑ|g c,m) and, in particular, whether they can suppo...

Citation Context

...lternative DCMs (Penny et al., 2004a), precise sampling from predicted responses (Kiebel et al., 2007), additional states at the neuronal level (Marreiros et al., 2008) andarefined hemodynamic model (=-=Stephan et al., 2007a-=-). In this paper, we describe a novel DCM for fMRI that enables one to model a class of nonlinear neuronal processes, which are important for a variety of cognitive processes, including learning and a...