面孔识别中脑电成分P1，N170和N250的民族特征反应

The Ethnic Characteristic Reaction of EEG Components P1, N170 and N250 in Face Recognition

摘要: 目的：为了调查面孔识别中的脑电成分P1，N170，N250是否能够准确的反映出面孔的民族特征，本研究设计了专门检测面孔民族特征信息的实验范式进行探究。方法：选取31名大学生作为被试，采用“平均面孔”以凸显民族信息，改变“平均面孔”的变形程度操纵被试对该特征信息的判断，同时记录被试的脑电与行为数据。结果：对被试反应时间数据的方差分析显示出变形水平的主效应十分显著F(6,174) = 25，P < 0.001， η_p²=0.46。被试民族与目标面孔民族交互作用显著F(1,29) = 8.7，P = 0.006， 。脑电结果显示同族与异族面孔所引起的P1，N170波幅差异并不显著t(30) = 1.70，p = 0.09；t(30) = 1.87，P = 0.07，而N250波幅差异十分显著t(30) = 2.84，P = 0.01。结论：面孔识别中早期的脑电成分P1，N170成分对面孔的民族特征并不敏感，N250成分对不同民族面孔的特征信息具有良好的检出性。

Abstract: Objective: To investigate whether the eeg components P1, N170 and N250 in face recognition can accurately reflect the ethnic characteristics of faces. In this study, an experimental paradigm spe-cifically designed to detect the ethnic characteristics of faces was designed to explore. Methods: Thirty-one college students were selected as subjects, and “average face” was used to highlight na-tional information, and the deformation degree of “average face” was changed to manipulate the judgment of the characteristic information. Meanwhile, eeg and behavioral data of the subjects were recorded. Results: The variance analysis of the reaction time data showed that the main effect of deformation level was very significant: F(6,174) = 25, P < 0.001, η_p²=0.46 . The participants’ ethnicity had significant interaction with the target face ethnicity (F(1,29) = 8.7, p = 0.006, and  ). The eeg results showed that the amplitude of P1, N170 caused by homologous and alien faces was not significantly different t(30) = 1.70, P = 0.09. t(30) = 1.87, P = 0.07, and N250 amplitude difference is very significant t(30) = 2.84, P = 0.01, d = 0.47. Conclusion: In the early stage of face recognition, components P1 and N170 are not sensitive to the ethnic characteristics of faces, while N250 has a good ability to detect the characteristic information of faces of different ethnic groups.