 |
1st ETRW on Speech Production Modeling:
From Control Strategies to Acoustics
|
 |
|
1st ETRW on Speech Production Modeling:
From Control Strategies to Acoustics
|
|
Factors affecting the reaction time (RT) in producing syllable sequences were analyzed to infer the neural process of speech command generation. This study examines four parameters in meaningless Japanese words with CV syllables, namely (i) number of syllables (1 -4), (ii) word initial consonant (/k/ vs. /t/), (iii) syllable sequence pattern (same vs. different consonant, e.g. /tatata/ vs. /takata/), and (iv) articulatory preparation for the utterance ("waiting for the onset signal with mouth closed" vs. "waiting for the signal with preparation for the word-initial consonant"). The results indicate that the RT measure varies with each factor in such a manner that the complexity of utterance plays a common determinant role in RT prolongation. The comparison in RT between two preparatory conditions for the utterance revealed that factors (i) and (iii) reflect the motor command generation stage in the neural process, while factor (ii) is attributable not only to the command generation stage but also to the articulation stage for realization of the word-initial consonant.
Bibliographic reference. Masaki, Shinobu / Honda, Kiyoshi (1996): "Control of speech command generation for Japanese words estimation from reaction time measurement", In SPM-1996, 105-108.