Study On The Phase Resetting Effect Of Event-related Potential Modulated By The Frequency Of Background EEG

Event-related potentials (ERPs), closely related to sensory and cognitive activity,is comprehensive response of the neural pathways of perception and the variousfunctional areas of cerebral cortex to external sensory stimuli and internal cognitivethinking, and implys wealth information on neurophysiology, brain cognition andpsychology, so is hailed as "window of observing brain thinking". It is highly worthbeing studied both in theory and application. As main advantage, it has a good timeresolution. But the signal amplitude of single trial is too small, often submerged in thebackground electroencephalogram (EEG), to be detected directly. Thus, although ERPhas been widely used in various studies about brain function since founded, its neuralmechanisms lies in hot dispute, and is difficult to clarify.There have been two assumptions about ERP generation mechanism: evokedmodel and phase resetting model. Traditional research tried to find evidence todisassociate the two models by grouping Spontaneous EEG, however, which wouldbring about obscure "grouping artifact". Taking into consideration that ERPgeneration is inseparable from background EEG, this paper modulated backgroundEEG into a regular signal by steady-state visual evoked paradigm to improve thesignal to noise ratio. On the other hand, the artificial defects caused by groupingmethod was avoided through modulating the state of pre-stimulate background EEGby frequency. In addition, this paper extracted the temporal phase feature by Hilberttransform, which avoided artifacts introduced by narrowband filtering to make theevidence more reliable.Through optimization, this paper designed six different experimental paradigmsto modulate background EEG, and transient stimulus in three kinds of intensity toevoke ERP. Eight subjects participated in the experiment. ERP was obtained byaveraging between trials. Results show that during P1time window, the phase ofbackground EEG reset as response to transient stimulus, which varied in forms ofexperimental conditions. And the initial phase of background EEG had a significantinfluence on the latency of N1. Then, this paper extracted the temporal phase byHilbert transform, and computed the phase distribution by Rayleigh test. Results showthat during N1time window, at low stimulus intensity, significant phase concentrationappeared earlier in paradigm1010than in1001, and lasted longer in paradigm1010than in1001. This difference of phase distribution was consistent with the difference of N1latency between two paradigms caused by initial phase. Finally, this paper usedshort-time Fourier transform to extract the time-frequency characteristics of eventrelated spectral perturbation and inter trial coherence, and analyzed main effects usingtwo-way Analysis of Variance. Results show that the generation of N1component atlow stimulus intensity was more in line with the phase resetting model, but moreconsistent with evoked model at high-intensity stimulus.By analyzing the differences of ERP evoked by stimulus under differentexperimental paradigms and different stimulus intensity, this paper found the initialphase of background EEG at low stimulus intensity was strongly related to latency ofP1and N1, and had significant effects in single whole power and phase changesduring P1and N1time window. Results in this paper can contribute to a moreprofound understanding of ERP generation mechanism and integrate the findings ofERP and background EEG.