No abstract

For Bayesian inference on the mixture of factor analyzers (MFA), natural conjugate priors on the parameters are introduced and then a Gibbs sampler that generates parameter samples following the posterior is constructed. In addition, a deterministic estimation algorithm is derived by taking modes instead of samples from the conditional posteriors used in the Gibbs sampler. This is regarded as a maximum a posteriori (MAP) estimation algorithm with hyperparameter search. The behaviors of the Gibbs sampler and the deterministic algorithm are compared on a simulation experiment.

No abstract

We carried out multi-dipole estimation and pursued spatio-temporal brain activity on a time scale of several milliseconds during an auditory discrimination task using a whole-cortex type SQUID system. Neuronal activities were estimated in the medial (hippocampus, parahippocampal gyrus, etc.) and lateral temporal cortices (superior and middle temporal gyri, etc.), the dorsolateral prefrontal cortex (middle and inferior frontal gyri, etc.) and the parietal cortex (supramarginal gyrus, etc.) in the 280-400 ms latency range. The activity in the posterior hippocampal region was the most prominent and long-lasting in parallel with the activities in the other regions. Therefore, the posterior hippocampal region is a central structure engaged in auditory discrimination. The whole-cortex neuromagnetic measurements provided the possibility of imaging the time-varying activities of the human cortico-hippocampal neural networks.

No abstract