| EEG Synchronization is considered the conformity or binding of the brainfunctional areas. Advanced brain function requires many relevant brain regions with aspecific function between the nervous system at different levels to achieve integrationand coordination. Years of animal experiments proved that the major potentialmechanisms of information integrating and information processing for the brain are thesynchronization oscillation of the neural network. Synchronization phenomenon of EEG,especially in the high frequency band, EEG synchronization oscillation, is considered akey part of the establishment of communicating information between different brainregionsEEG recorded on the scalp is the synchronous reflection of the synchronousactivities of many neurons or neural networks in the brain. Therefore, at the macro level,EEG is available to study such a mechanism. That is, at a certain temporal resolution, bystudying the local scope and global scope of the distribution of EEG synchronization,estimate the general picture of biological activity profile of electricity of the localcortical cortex and the overall cortical internal.EEG synchronization problem are involved in a wide range of research problems,in this article, starting from the research methods, more systematic analysis, comparison,conclusions and improvement the relevance of existing research methods and theirapplication to visual information processing mechanisms study including the two visualpathways, the spatial attention and the integration between visual and audio perception.The main works done in this thesis are as follows:1. Based on the 128-channel high temporal-spatial resolution EEG and Fouriertransform, event-related coherence technique is adopted to study the cognitiveactivation patterns in the two visual pathways. In this study, two different ERPs, oneresponding to the task of form perception (task 1) and another evoked by both formdiscrimination and spatial location (task 2), are involved into calculation ofevent-related coherence. The coherence is calculated between occipital area and othersix areas in different EEG frequency bands. Those sub-areas located between occipital and frontal occipital are determined by the similarity of neighbored electrodes beforecoherence calculation. The results indicated that inγl band (28-39Hz) , both thecoherence and the electrode-pair density responding to simultaneously form perceptionand spatial location are statistically stronger than those in only form perception task,which was even more obvious in occipital and frontal area. These findings furthersuggest that the increasing of synchronization found inγl band is due to theparticipation of dorsal stream, which can strengthen the information exchangingbetween occipital and frontal occipital.2. Information processing mechanism in the upper and lower visual fields plays animportant role to study brain functions. Synchronous activity of brain areas is a latentmechanism for the integration or binding of brain functional areas, and can be used tostudy the difference between the upper and lower eye fields. The difference of thesynchronization is revealed by calculating a nonlinear interdependence of ERP fromleft-right temporal electrode in the visual fields for different task. These results supportthat the lower visual field is superior to the upper visual field in the ability of processingvisual information.3. A method based piecewise Prony method (PPM) was presented to research thephase synchronization. The piecewise Prony method (PPM) decomposes EEG signalinto a summation of sinusoidal components with decaying or growing envelope in orderto obtain the magnitude of the frequency components, frequency and phase information,and use this phase information to calculate difference of the phase of two-channel signal.And then, we can measure the degree of phase synchronization of two-channel signalsaccording to Shannon entropy. The method is firstly evaluated and compared with othermethods using simulated transient non-stationary data, and the simulated results showedthat PPM has better anti-noise performance, with higher frequency and phase resolution.The phase synchronization entropy based PPM is consistent with the physiologicalstates when this method is applied to a real rat EEG data. These results confirmed thatthis phase synchronization analytic method based PPM is robust to noise andnon-stationary, and it is a promising tool for EEG signal processing. 4. To study synchronization between visual area and audio area during cognitiveprocessing, an experiment of visual-audio simultaneous stimulus is designed to acquirecontinuous 128-channel EEG recordings. ERP evoked by stimulus is processed bymeans of Hilbert phase synchronization method. Under such trial condition, thesynchronous index between occipital area and temporal area is larger than the indexcalculated between occipital area and parietal area. Moreover, synchronous activity willoccur among other relevant brain areas automatically during the cognitive procedure.
|
PDF Full Text Request