Department of Special Education and Communication Disorders


Document Type


Date of this Version



Published in Journal of Speech, Language, and Hearing Research 46 (August 2003), pp. 1016-1021; doi: 10.1044/1092-4388(2003/079) & doi: 10.1044/1092-4388(2003/080) Copyright © 2003 American Speech-Language-Hearing Association. Used by permission.


Dworkin et al. comment: We would like to comment on Green, Moore, and Reilly’s article, which appeared in the February 2002 issue of this journal [Journal of Speech, Language, and Hearing Research]. In that investigation, these clinical researchers examined upper lip, lower lip, and mandibular movements during repetitive bisyllable word productions by infants, toddlers, young children, and adults with normal developmental and neurologic histories. Kinematic traces from these articulators were analyzed using a computer-based movement tracking system. Results revealed that these oral structures may have sequential neuromotor developmental schedules, characterized by more mature movement patterns for speech emerging earlier in the mandible than in either the upper or lower lip. That is, that normal speech development involves the integration of lip and tongue activities into a more well-established, biomechanically dominant jaw operating sensorimotor system. To facilitate our response to this investigation, we have chosen first to extend the results by elaborating on the causally related role of the mandible in certain speech disordered populations, and second, to highlight how adjunctive methods of data collection may have strengthened the validity of the overall findings.

Green et al. respond: We appreciate the opportunity to clarify our previous findings in response to questions raised by Dworkin, Meleca, and Stachler (2003), who have cited our work with respect to its implications for the study and treatment of speech motor impairment. These researchers raise the possibility that differences in articulatory performance across the age groups may have been related to group differences in speech sampling methods rather than to development. They imply that because our younger (i.e., 1- or 2-years-old) participants’ utterances were obtained with reduced experimental control relative to the older participants’ (play vs. reading), their articulatory performance should have been less consistent than the older participants’ because of naturally occurring variations in loudness and rate, and potential “upstream” effects related to “torso and limb adjustments.” Several aspects of our experimental design mitigate these concerns and, most importantly, our findings are the opposite of predicted effects arising from speech sampling differences across age groups: (a)Adult-like stability was observed in the infants’ jaw movement patterns despite the fact that infant vocalizations were elicited under less controlled conditions. (b) Each participant served as his or her own control, which permitted the evaluation of differences across articulators during development. (c) Postprocessing techniques minimized linear-scaling differences in articulatory movement across repetitions. (d) Finally, trunk mobility was restricted during data collection.