Evaluation of the functional connectivity in the nervous system using time dependent scatter diagrams

We investigated spike trains of the feline nervous system, which are considered to be non-stationary point processes characterized by changes of the functional interneuronal connectivities. Ordinary long-term averaged correlation methods are useful as processing time series data as a quasi-stationary point process, but not as a non-stationary point process. We depicted the joint impulse configuration scatter diagram proposed by D. H. Perkel et al. for subsets of the neural spike-train to find changes of functional connectivities. The time-dependent scatter diagram was generated by arranging these scatter diagrams in chronological order. Time-dependent scatter diagrams were generated for spike trains derived from the anterior ectosylvius sulcus, the coronal sulcus, the lateral bank of the presylvius sulcus, the medial wall of the presylvius sulcus, and the medial wall of the cruciate sulcus. The patterns of time-dependent scatter diagrams for these cortices are considered to change in relation to eye movement. Changes of these patterns correspond to those of the functional connectivities related to the dynamics of eye movement in the neural system.