Abstract: Event related potential(ERP) literature on recognition memory has found that recollection is associated with the LPC old/new effect, which occurs around 500- 800ms at posterior electrodes, and familiarity is associated with the FN400 old/new effect, which occurs around 300-500ms at anterior electrodes. The view that the FN400 reflects familiarity has been challenged by some recent researches. One of the evidences against the view that the FN400 reflects familiarity is that the FN400 is not different from the N400, which reflects semantic processing. However, this result was challenged by some later studies that found the FN400 was different from the N400. The present study investigated the relationship between the FN400 and the N400 using picture of visual objects as stimuli and ERP technology. The present study used the masked conceptual priming paradigm in the study phase and the remember/know (R/K) paradigm in the test phase. 200 pictures were presented in the study phase. Half were preceded by a masked semantically related picture (related condition), and half were preceded by a masked semantically unrelated picture (unrelated condition). Subjects performed an R/K recognition memory test in the test phase. 100 studied pictures and 50 unstudied pictures were presented in each test block (two blocks in total). Continuous EEG was recorded using the NeuroScan system, with 64 electrodes embedded in a hat, in both the study and test phase (band pass: 0.05-100Hz, 30Hz low pass filtered in offline processing). Epochs were created beginning 100 ms prior (serve for baseline correction) to and ending 900 ms after the onset of stimulus presentation in both the study and test phase. The FN400 familiarity effect was defined as the ERP difference between K hits and correct rejections at 300-500ms in the test phase and the N400 conceptual priming effect was defined as the ERP difference between related and unrelated condition at 300-500ms in the study phase. The behavioral results revealed that masked conceptual priming facilitated the response time in related condition compared to unrelated condition, [685(±51）ms vs. 704(±57)ms，t(17) = 5.741，p < 0.001，d = 1.35]and that both R response and K response in the test phase were above chance level. The ERP results revealed that masked conceptual priming was associated with the N400 occurred at 300-500ms and that familiarity was related to the FN400 occurred at 300-500ms and recollection was related to the LPC occurred at 500-800ms. The comparison of the topographic distribution of the FN400 familiarity effect and the N400 conceptual priming effect revealed that there was no difference between the topographic distribution of the FN400 and the N400 effect, [F(61，1037) = 0.675，p = 0.65，η_p^2 = 0.04). These results suggested that the FN400 and the N400 for picture stimuli are similar (i.e., they both reflect semantic processing), which supported the idea that FN400 effect reflects conceptual priming in the test phase but not familiarity.

摘要： 本研究通过比较记忆编码阶段语义启动引起的N400效应和提取阶段熟悉性引起的FN400效应的地形图考察图形材料诱发的N400和FN400是否是相同的ERP成分。实验结果发现，语义启动引起的N400效应和熟悉性引起的FN400效应的地形图没有显著差异。这说明图形材料诱发的FN400和N400是相同的ERP成分，两者都反映了语义加工。本研究为FN400效应反映了语义启动而不是熟悉性的观点提供了新的实验证据。