RELIABILITY OF ELECTROMAGNETIC MIDSAGITTAL ARTICULOMETRY AND ELECTROPALATOGRAPHY DATA ACQUIRED SIMULTANEOUSLY

This study evaluates the effects on electromagnetic midsagittal articu lometry (EMA) coils from electrode activation on an electropalatograph y (EPG) artificial palate, and on EPG contact patterns induced by EMA coils, In order to investigate the first issue, five transducer toils on a fixed magazine were placed equidistantly among themselves and wit h respect to the three transmitter coils of an EMA helmet. Two palate positions were used, i.e., perpendicular to the magazine (to emulate p ossible effects on coils placed on the nose, upper incisors, jaw, and lips) and parallel to it (to imitate effects on coils placed on the to ngue surface). A thin leaf of aluminum firmly pressed by a mold was us ed to create contact activation. Data for different degrees of contact activation were subtracted from data obtained with no artificial pala te (perpendicular condition) or with an artificial palate with zero '' on'' electrodes (parallel condition). Only for the most extreme condit ion, i.e., a high degree of contact activation and a short coil-to-art ificial palate distance, do the effects slightly exceed the noise leve l for the EMA equipment. The analysis of the second issue was carried out for coils attached to the tongue tip, the tongue blade, and the to ngue dorsum during the production of several consonants. Coil-to-Lingu al contact effects occur mostly near the midline of the artificial pal ate at different locations, depending on the degree of contact and the location of the consonant. (C) 1996 Acoustical Society of America.