恐惧和悲伤情绪的调节具有不同的神经基础,通过分析这两种情绪调节过程中事件相关电位的差异,探索其电生理机制.采集21名正常受试者对恐惧和悲伤情景图片进行自然注视和认知重评任务时的头皮脑电信号,选取4个感兴趣区域(左额中央区、右额中央区、左顶中央区、右顶中央区),分析刺激后500～5 000ms时间范围内(500 ms为一个时间窗,共9个时间窗)两种情绪刺激诱发得到的事件相关电位晚成分(LPP)的幅值差异.统计分析采用重复测量方差分析方法,组内因素为时间(9个水平)、区域(4个水平)、任务条件(2个水平:自然注视和认知重评)以及情绪(2个水平:恐惧和悲伤).结果发现,在刺激后500～1 000 ms,在右顶中央区,恐惧刺激诱发的LPP显著高于悲伤刺激诱发的LPP(P＜0.05),自然注视时为(8.815±1.153)μV vs(5.834±1.317)μV,认知重评时为(7.094±1.036)μV vs(6.643±1.158)μV.在刺激后1 500～2 500 ms各时间窗,悲伤刺激的认知重评显著增强额中央区活动(P＜0.05),其中刺激后1 500～2 000ms的结果为(10.100±2.205)μV vs(2.515±1.676)μV;刺激后2 500～4 000 ms各时间窗,任务条件具有主效应,恐惧和悲伤刺激的调节都能显著提高左侧额中央的LPP幅值(P＜0.05),其中刺激后2 500～3 000 ms的结果恐惧为(2.957±1.767)μV vs(1.899±2.969)μV,悲伤为(7.957±2.305)μV vs(-0.051±2.018)μV.结果表明,恐惧和悲伤刺激的调节具有不同的时间过程,且左侧额中央区域在情绪调节过程中具有重要作用.

选取8名轻度认知障碍(mild cognitive impairment,MCI)老年人和17名正常老年人参与图片颜色的认知任务,并同步记录他们的脑电信号,然后采用相对能量、样本熵和累积残余熵分析方法,研究这两组被试在执行任务时的不同脑区脑电活动的差异.研究发现:1)两组被试alpha相对能量无差异,但MCI患者额中央区和顶枕中央区的脑电熵高于正常人;2)两组被试额区脑电的熵值都高于其他区域,中线区域的脑电熵值低于左右半球;3)刺激类型对两组被试的脑电活动有影响,在完成匹配任务时,两组被试大脑前侧区域的脑电活动更为复杂,并且正常人大脑右侧中央区域的脑电活动复杂性较高;4)与样本熵相比,累积残余熵能更细致地刻画出MCI患者脑电的非线性特征.结果表明,MCI患者认知过程中的脑电活动存在异常,信号更为复杂.与样本熵相比,累积残余熵更适合于分析脑电的非线性特征.

This paper focused on the brain functional connectivity difference between the depression and the healthy during emotion processing.A visual search task about positive and negative face was performed in 12 depressed patients and 12 control subjects.As a measure of functional connectivity,EEG coherence based on MVAR modeling in delta,theta,alpha bands was calculated from 15 electrode sites.Further,the authors calculated anterior short-range connectivity,posterior short-range connectivity,left fronto-parietal and right fronto-parietal long-range connectivity.It was found that: 1) in depressed patients,the strength of anterior short-range connectivity for delta,theta and alpha bands were significantly smaller than the controls during 1~200 ms after stimulation onset,as well as the fronto-parietal long-range connectivity for delta and theta bands;2) in normal group,the coherence value of anterior and posterior short-range connectivity for delta,theta and alpha bands were significantly larger for the positive stimulus than for the negative one.Similarly the left fronto-parietal long-range connectivity for delta and theta bands,and right fronto-parietal long-range connectivity for alpha band were larger for the positive stimulus.While for the depression,there was no emotional difference.These results indicated that the involvement of frontal and left fronto-pariertal network might be functionally insufficient in depressed patients which were reflected by decreased frontal and left fronto-pariertal coherence during emotion processing.

Studying the functional network during the interaction between emotion and cognition is an important way to reveal the underlying neural connections in the brain and nowadays, it has become a hot topic in cognitive neuroscience. Granger causality (GC), based on multivariate autoregressive (MVAR) model, and being able to be used to analyse causal characteristic of brain regions has been widely used in electroencephalography (EEG) in event-related paradigms research. In this study, we recorded the EEGs from 13 normal subjects (6 males and 7 females) during emotional face search task. We utilized Granger causality to establish a causal model of different brain areas under different rhythms at specific stages of cognition, and then convinced the brain dynamic network topological properties in the process of emotion and cognition. Therefore, we concluded that in the alpha band, (1) negative emotion face induced larger causal effects than positive ones; (2) 100-200ms emotional signal was the most prominent ones while 300-400ms and 700-800ms would take the second place; (3) The rear brain region modulated the front in the process of causal modulation; (4) The frontal and pillow area involved in the brain causal modulation as a key brain area; and (5) Negative partiality existed in the information processing, especially during 0-100ms after the negative expression stimulation.

Cognitive and emotional dysfunction in depressed patients has been broadly researched. However, the physiological mechanism of the emotional impairment is still obscure. In this study, according to the ethical principles and the procedures of informed consent, we enrolled 16 depression patients and 16 normal subjects in a "learning-judgment" cognitive task. Emotional faces were chosen and used as the visual stimuli, while the subjects' scalp electroencephalographs (EEG) were recorded. We calculated the coherences of the event-related EEG signals between different functional brain regions and used the repeated measure analysis of variance (ANOVA) to analyze the coherences. The results revealed, (1) The depressed patients had a significantly lower inter-hemisphere coherence than did the normal controls at frontal sites across all EEG basic rhythms; (2) The inter-hemispheric coherence of EEG beta activity was significantly correlated with the emotional processing during 1-200 ms after stimulus onset, and the synchronicity with the negative emotion stimulus was higher than that with neutral and positive stimuli; (3) The main effect of race was found at delta and theta bands at about 401-600 ms after stimulus onset. Moreover, the connectivity of inter-hemisphere for own-race stimulus was stronger than that for other-race stimulus.