Hierarchical coding of characters in the ventral and dorsal visual streams of Chinese language processing.

Functional and spatial hierarchical organization of increasingly language-like word forms has been proposed for alphabetic languages at the occipitotemporal cortex for visual word recognition. In the logographic Chinese language system, similar functional and spatial hierarchical presentations of brain responses to sublexical orthographic structure are beginning to be explored. In this study, we used whole-brain fMRI to show that a hierarchical coding of increasingly language-like character type is present in multiple Chinese language processing areas. Fluent Chinese readers were presented with Chinese synonyms/non-synonym pairs, identical/non-identical non-pronounceable pseudo-character pairs constructed with Chinese radicals, and identical/non-identical Korean character pairs. We observed the presence of a spatial gradient for increasing language-like character types in the ventral and dorsal visual streams of the cortex. At the left occipitotemporal cortex of the ventral visual stream, we observed a posterior-to-anterior gradient of character type selectivity with the anterior fusiform region being more selective for real Chinese characters and the posterior fusiform region being more selective for Korean characters. At the left and right intraparietal sulci of the dorsal visual stream, a medial-to-lateral gradient of character type selectivity was observed, with the lateral edge being more selective for real Chinese characters, the medial edge being more selective for pseudo-characters, and with less activation attributable to Korean characters. Spatial gradients of selecting character type were also identified in prefrontal cortex, dorsal striatum and lateral temporal cortex. The results suggest that the left occipitotemporal cortex and both left and right intraparietal sulci are tuned with a functional and spatial hierarchical sensitivity to the presence of semantic elements as well as different orthographic structures.

Can Cantonese rhymes be used in the assessment of hemispheric dominance for language?

English vowels had been proposed in previous studies to be used as a simple tool for the brain mapping of language. A proper fMRI study of Cantonese rhymes, each of which being a required and fundamental unit of a Cantonese syllable, remains to be carried out. Using an auditory task with Cantonese rhymes which carry no semantic meaning, we observed a minimal amount of positive BOLD signal at the caudate nucleus when Cantonese rhymes were contrasted with their corresponding filtered sounds. Typical language activating regions of the prefrontal cortex, the medial prefrontal cortex and the lateral temporal cortex on both left and right sides were not activated by Cantonese rhymes. Based on the absence of brain activation at the typical language areas in the contrast of Cantonese rhymes relative to filtered sounds, the auditory task with Cantonese rhymes may not be a robust tool for the individual clinical assessment of hemispheric dominance for language.