The Processing Characteristic Of Mismatch Negativity

Whether the MMN is the index of the automatic process of the brain is always a debate for the MMN research, at the same time it is also an important arguments to explain the mechanism of the MMN formation. Many paradigms has put forward their own explanations for the characteristic of the automatic and non-automatic for the MMN, however, they each has a certain limitation which they can’t make sure the attention source of the health subject is pure and in their ignore path will be contaminated by some kind of attention but still can’t quantify this source. The reason is that there is no index to reflect the subtle fluctuate of attention resource of the normal subjects. As regards to the experiments results from the attention deficit subjects, for participants under the situation of brain function defect is not clear so it can’t be used to explain the phenomenon of the health subject.To solve the problems above, this paper probes into across the channel delay reaction to improve the experimental paradigm to better control subjects’ attention resources in order to clear the relationship between MMN and attention. And collect the behavior data during the experiment, fitting the data with diffusion model the parameters can be used as the index of the attention resource allocation. Due to the low signal noise ratio (SNR) during the MMN extraction which influences the MMN component extraction, here we use traditional averaging of EEG integrating with wICA algorithm to get the result. Finally on the combination of peak latency and peak amplitude of MMN we can make a conclusion of the relationship between MMN and attention.After experiment and data analysis, the parameters fitting by diffusion model using behavior data can explain the underlying decision processes under different conditions on the subjects’ attention which modulate by the design of the experimental. Quantitatively determine the subjects’ attention. The variation of the peak latency is significant for different conditions and has a significant correlation between parameters of diffusion model:a correlation coefficient r between the boundary separation is0.63, r=0.63between non-decision time and a correlation coefficient of0.58between drift rate. This proves that the parameters of diffusion model can be used as index of the subjects attention resource allocation, and MMN peak latency was positively correlated with intensity of attention, so we make a conclusion that the MMN is a semi-automatic processing.