We used functional magnetic resonance imaging (fMRI) to examine whether the act of writing involves different neuro-psychological mechanisms between the two script systems of the Japanese language: kanji (ideogram) and kana (phonogram). The main experiments employed a 2 × 2 factorial design that comprised writing-to-dictation and visual mental recall for kanji and kana. For both scripts, the actual writing produced a widespread fronto-parietal activation in the left hemisphere. Especially, writing of kanji activated the left posteroinferior temporal cortex (lPITC), whereas that of kana also yielded a trend of activation in the same area. Mental recall for both scripts activated similarly the left parieto-temporal regions including the lPITC. The writing versus mental recall comparison revealed greater activations in the left sensorimotor areas and right cerebellum. The kanji versus kana comparison showed increased responses in the left prefrontal and anterior cingulate areas. Especially, the lPITC showed a significant task-by-script interaction. Two additional control tasks, repetition (REP) and semantic judgment (SJ), activated the bilateral perisylvian areas, but enhanced the lPITC response only weakly. These results suggest that writing of the ideographic and phonographic scripts, although using the largely same cortical regions, each modulates the visual word-retrieval system according to their graphic features. Furthermore, comparisons with two additional tasks indicate that the activity of the lPITC increases especially in expressive language operations regardless of sensory modalities of the input stimulus.

We recorded magnetic and electrical responses simultaneously in an auditory detection task to elucidate the brain areas involved in auditory processing. Target stimuli evoked magnetic fields peaking at approximately the same latency of around about 400 msec (M400) over the anterior temporal, superior temporal, and parietal regions on each hemisphere. Equivalent current dipoles (ECDs) were analyzed with a time-varying multidipole model and superimposed on each subject's magnetic resonance image (MRI). Multiple independent dipoles located in the superior temporal plane, inferior parietal lobe, and mesial temporal region best accounted for the recorded M400 fields. These findings suggest that distributed activity in multiple structures including the mesial temporal, superior temporal, and inferior parietal regions on both hemispheres is engaged during auditory attention and memory updating.