Search Abstracts | Symposia | Slide Sessions | Poster Sessions
Neural substrates of phonology-to-orthography consistency on Chinese character and Japanese Kanji reading based on fMRI investigations
Poster Session C, Friday, October 25, 4:30 - 6:00 pm, Great Hall 3 and 4
bingfeng shan1, ayumi seki2; 1Hokkaido University Graduate School of Education, 2Hokkaido University Faculty of Education
Both Chinese and Japanese utilize Chinese characters (kanji in Japanese), which share many similarities in orthography and semantics. However, they differ in phonology-to-orthography consistency. In Chinese, each character typically has one pronunciation, whereas most kanji in Japanese have multiple possible readings depending on the context. This consistency difference suggests different neural mechanisms of phonological processing across the two languages. It remains unclear whether the consistency difference alone can fully explain the divergence in Chinese characters reading between Japanese and Chinese. Therefore, this study employed functional magnetic resonance imaging (fMRI) to compare the neural activations associated with Chinese characters/kanji reading under varying consistency conditions, aiming to investigate the relationship between cross-linguistic differences and phonology-to-orthography consistency. The study recruited 12 Chinese and 12 Japanese university students as participants. Chinese character/kanji stimuli were divided into four consistency conditions (low, medium, high, fully consistent) based on the proportion of the most frequent pronunciation relative to all possible pronunciations. Participants silently read the presented Chinese character/kanji stimuli and judged whether a specified syllable was included, while their neural responses were recorded using 3T Siemens MRI scanner. In the analysis, whole-brain analysis compared neural activations across all consistency conditions between the two languages. Additionally, a two-way ANOVA with language and consistency as factors identified brain regions modulated by consistency level. These consistency-related regions were defined as regions of interest (ROIs). Percent signal change was extracted from the ROIs for each consistency condition. In addition to these consistency-related ROIs from the ANOVA, the left inferior frontal gyrus (IFG) and middle frontal gyrus (MFG) were also included in the ROI analysis, as previous studies have implicated these regions in consistency effects. The whole-brain analysis showed broad bilateral brain network engagement during reading in both Chinese and Japanese, with no significant differences between the two language groups. The ANOVA revealed a main effect of consistency level in bilateral fusiform gyri (FG), left inferior parietal lobule (IPL), and left postcentral gyrus (PG). A main effect of language was found in noncritical regions like the cuneus, posterior cingulate, and left cerebellum. No interaction between language and consistency emerged. In the ROI analysis, the bilateral FG, left IPL, and left PG exhibited a similar inverted U-shaped pattern across languages, peaking at medium condition. However, in the left IFG and MFG, the modulation of activity by consistency levels differed between the Chinese and Japanese. After controlling for consistency level, no significant differences were observed between Chinese character and kanji reading, suggesting highly overlapping neural networks for phonological processing. The similar activity patterns in the bilateral FG, left IPL, and left PG further indicate that consistency effects operate equivalently across the two languages. However, distinct consistency effects in left IFG, MFG imply these regions may contribute differentially to the phonological processing of Chinese characters across the two languages. Beyond consistency, there may be other factors influenced by the inherent disparities between the writing systems could modulate the neural substrates underlying character reading.
Topic Areas: Reading, Phonology