Speech task and timing considerations in MRI research

• Published in: The Journal of the Acoustical Society of America 2003-04, Vol.113 (4_Supplement), p.2258-2259
• Main Authors: Epstein, Melissa A., Stone, Maureen
• Format: Article
• Language: English
• ISSN: 0001-4966; 1520-8524
• DOI: 10.1121/1.4780465

In order to create dynamic magnetic resonance images, subjects must repeat tokens as much as 30 times in a row, therefore, their precision is critical for image quality. Word repetition rate can be 1 or 1.5 s, and is matched to the rate of image recording. Unlike reciting tokens to the beat of a metronome, subjects must say each token during an extended period of noise that lasts about 75% of the repetition period. Furthermore, some tokens may be more difficult to say repetitively in these conditions. Therefore, this study simulated the acoustics of a MRI recording session to examine the effects of subject, token, and repetition rate on temporal precision. Subjects repeat up to ten mono- to trisyllabic words 30 times each at two repetition rates. Measurements are made of onsets and offsets of one or more phonemes within each word. Preliminary results (five subjects) indicate that there is an effect of subject and word. Further subjects are being measured to corroborate these results and to determine if one repetition rate is better than another and to test subject reliability.