Measurement of Neural Activity Related to Processing of Katakana Characters, Words and Inner Speech

Abstract

This review summarizes a recent experiment conducted using fMRI and MEG to measure neural activation related to orthographic (more generally visual form), phonological, and lexico-semantic processes for visually presented Japanese katakana (phonogram) characters, pseudocharacters, and strings of these characters. Activity related to visual form processing was observed at latencies earlier than about 200 ms in the lateral extrastriate visual cortex, fusiform gyrus, and posterior inferior temporal areas. (1) The activation was the same between katakana characters and pseudocharacters, and between katakana strings or words and pseudocharacter strings. (2) The activation occurred in both hemispheres for single characters, but was left-lateralized for strings of characters. The activity related to phonological processing was observed at latencies later than about 200 ms in Wernicke's area, the supramarginal gyrus, and Broca's area or insula. To investigate differences in functional roles between the phonology-related areas, we conducted another experiment that required inner speech but no other phonological processes, and found that activity at all three phonological areas depended significantly on task demands. This result suggests that these areas worked together for inner speech, although they may each play different roles. Although activation for semantic processing could not be dissociated from that for the other processes in this study, there is a possibility that the semantic processing shares activation foci, e.g., near Wernicke's area, with the phonological processing.