Brain activation in the auditory processing of Cantonese rhymes in native Cantonese speakers using functional magnetic resonance imaging : an event related study

Abstract

Previous functional studies have demonstrated that major language processing areas at the inferior frontal (Broca's area) and superior temporal gyri (Wernicke's area) were activated by auditory stimuli with the English vowels and Mandarin rhymes. Cantonese is a major southern Chinese dialect. We hypothesized that the Cantonese rhymes would activate the inferior frontal gyrus. We hypothesized that it would also activate the superior temporal gyrus. The activation pattern by Cantonese rhymes would be similar to that from previous studies with English vowels. We also hypothesized the activation pattern from Cantonese rhymes would be similar to that from previous studies with Mandarin rhymes. The purpose of this study was to investigate the feasibility of using functional magnetic resonance imaging (fMRI) with auditory stimuli in Cantonese rhymes in the determination of language processing areas in native Cantonese speakers. The experiment tested 15 native Cantonese speakers aged from 19 to 25 years. Blood oxygen level dependent (BOLD) fMRI brain images were acquired on each subject using a 1.5 Tesla scanner while they were performing a discrimination task of Cantonese rhymes and filtered rhymes. The filtered rhymes were made by applying a low-pass filter, which attenuated frequencies higher than 200 Hz, onto the Cantonese rhymes. A number of brain regions were commonly activated by both Cantonese rhymes and filtered rhymes. On the frontal and parietal cortices, they were bilateral medial frontal (FGM), inferior frontal gyri (IFG), precentral gyri and postcentral gyri. On the temporal cortex, bilateral superior temporal sulci/gyri (STS/STG) were activated by Cantonese rhymes or filtered rhymes. On the occipital cortex, bilateral lingual gyri (LG) and right calcarian sulcus were activated by Cantonese rhymes as well as filtered rhymes. In the comparison of Cantonese rhymes vs. filtered rhymes, although significantly stronger positive activations were only demonstrated at the left middle frontal gyri (MFG) and IFG with Cantonese rhymes, activity in some other brain areas specifically with Cantonese rhymes, but not with filtered rhymes, were noted in separate comparisons of Cantonese rhymes and filtered rhymes with silence. They included bilateral MFG, the left superior parietal gyrus (SPG) and the left middle temporal gyrus (MTG). Stronger and more extensive activations were found in the left brain than the right brain in the comparison of Cantonese rhymes vs. silence. Negative BOLD signals were consistently found bilaterally at angular gyri (AG), and the anterior and posterior cingulate gyri in separate comparisons of Cantonese rhymes and filtered rhymes with silence. Our result suggested the left IFG and the MFG were responsible in semantic processing of Cantonese rhymes, which was consistent with the previous findings on English vowels and Mandarin rhymes. The activation in the left hemisphere by the Cantonese rhymes in the present study was as extensive as that by English vowels and Mandarin rhymes. With the use of Cantonese rhymes as auditory stimuli, language processing areas in native Cantonese speakers can be identified. Left language lateralization can also be demonstrated with the auditory stimuli of Cantonese rhymes. The present results also showed that the left MFG plays an important role in Cantonese rhymes processing, apart from the Broca's and Wernicke's areas. The present study was the first fMRI study using Cantonese rhymes to localize the language processing areas. It provides neuroscientists evidences of using the auditory Cantonese rhymes to localize the language processing areas in Cantonese speaking community.