Slide Slam K5
The representations of Chinese characters: evidence from sub-lexical components
Xiaodong Liu1, David Wisniewski1, Luc Vermeylen1, Ana F. Palenciano2, Wenjie Liu3, Marc Brysbaert1; 1Department of Experimental Psychology, Ghent University, Belgium, 2Department of Experimental Clinical and Health Psychology, Ghent University, Belgium, 3Department of Electronics and Information Systems, Ghent University, Belgium
Little research has been done about the neural substrate of the sub-lexical level of Chinese word recognition. In particular, it is unclear how radicals participate in Chinese word processing. We compared two measures of radical combinability: position-general radical combinability (GRC) and position-specific radical combinability (SRC) depending on whether the left or right side of the radical is taken into account. We selected characters with embedded target radicals that had different GRC and SRC measures. These measures were used as predictors in a parametric modulation analysis and a multivariate representational similarity analysis (RSA). 41 native Mandarin speakers were asked to read words in search of animal words. Results showed that SRC is a better predictor than GRC in decoding the neural patterns. Whole-brain analysis indicated that SRC is encoded bilaterally in the inferior frontal gyrus (IFG, pars opercularis and pars triangularis), the middle frontal gyrus (MFG), and a region on the border of the superior parietal lobule and the inferior parietal lobule (SPL/IPL). ROI-based RSA further confirmed the results of the whole-brain analysis. Furthermore, we observed a correlation of another sub-lexical variable, logographeme composition, with bilateral activity in SPL. Logographemes refer to the basic stroke combinations that form radicals and characters. Finally, we observed consistent involvement of bilateral cerebellum activity in Chinese word recognition. Our findings confirm the importance of sub-lexical components (SRC and logographeme composition) in Chinese word recognition and also confirm that Chinese word recognition involves more bilateral processing than word recognition in alphabetical languages.