University of Cambridge, Center for Neuroscience, Faculty of Education, 184 Hills Road, CB2 2PQ, United Kingdom.
Biol Psychol. 2010 May;84(2):354-67. doi: 10.1016/j.biopsycho.2010.04.002. Epub 2010 Apr 13.
We dissociated ERP markers of semantic (numerical distance) vs. syntactic (place value) incongruence in the domain of arithmetic. Participants verified additions with four-digit numbers. Semantic incongruencies elicited the N400 ERP effect. A centro-parietal (putative P600) effect to place value violations was not related to arithmetic syntax. Rather, this effect was an enlarged P3b reflecting different surprise values of place value vs. non-place value violations. This potential confound should be considered in numerical cognition experiments. The latency of the N400 and P3a effects were differentially affected by place value analysis. The amplitude of the P3a and that of a fronto-central positive effect (FP600) was sensitive to place value analysis and digit content. Results suggest that ERPs can index the syntactical analysis of multi-digit numbers. Both ERP and behavioral data confirmed that multi-digit numbers were decomposed into their constituent digits, rather than evaluated holistically.
我们在算术领域中分离了语义(数值距离)与句法(位值)不一致的 ERP 标记。参与者用四位数验证加法。语义不一致引起了 N400 ERP 效应。对位置值违反的中央-顶叶(假定的 P600)效应与算术语法无关。相反,这种效应是放大的 P3b,反映了位置值与非位置值违反的不同惊讶值。在数字认知实验中应考虑这种潜在的混淆因素。N400 和 P3a 效应的潜伏期受到位值分析的不同影响。P3a 的幅度和额中央正效应(FP600)的幅度对位值分析和数字内容敏感。结果表明,ERP 可以指数多位数的句法分析。ERP 和行为数据都证实,多位数被分解为其组成数字,而不是整体评估。