I am reading a relatively new paper about Treatment for Establishing Motor Program Organization (TEMPOSM) in childhood apraxia of speech (CAS) by Miller et al. It is a good paper showing that this treatment is effective for reducing segmentation between syllables and improving stress contrast when producing multisyllable nonsense words. There were 11 children and a novel design so the quality of the paper is good. I think the work this group is doing on the treatment of CAS is important.
That said, there is something that is bothering me about the CAS literature (beyond the bizarre adherence to schema theory by EVERYONE, see Speech Therapy and Theories of Speech Motor Control: Part I | Developmental Phonological Disorders.)
What is bothering me that is that we (including me) keep writing that prosody is a core feature of CAS but when you go looking for evidence for that it is really hard to find. Miller et al say that “acoustic measures of the speech of children with CAS show evidence of a disruption in temporal control of speech, marked by increased duration and reduced variability in duration of speech segments.” Seven papers are cited to support disruptions in the duration of speech segments by children with CAS. I have read all those papers many times but I decided to go back to them again. I summarize the evidence here.
In the first paper, Ballard et al describe a prototype of the ReST treatment that targeted syllable stress in 3 children with good effects for at least 2 of the children. But is the problem with stress marking exhibited by these three children a core feature of CAS?
The second paper used a systematic procedure to show a deficit in planning syllables during speech production, recruiting children with apraxia and children with normal speech development. Niland et al were interested in intra- versus inter-syllabic coarticulation strength and duration structure across the two groups of children. The measures related to coarticulation did not reveal differences in this study. However, differences emerged for the durations of specific segments as a function of prosodic structure of the phrases. That is, in structures like CVs#CVC the duration of the consonants and vowels will differ compared to CV#sCVC because total syllable duration is normally taken into account. The expected metrical adjustments were less systematic in the speech produced by children with CAS. The difficulty of the CAS group is attributed to a problem with ‘phonetic encoding’. This study has the strongest evidence of all the papers cited. It would be interesting to see the result for other children with a speech sound disorder.
The third paper proposes a coefficient of variation as a marker for CAS. This idea is very popular and I have cited this paper fairly often. It is easy to forget what this coefficient is however. One part of the equation refers to duration of speech events (parts of the acoustic waveform that are speech) while the other part refers to duration of pause events (parts of the acoustic waveform that are not speech, or pauses). The ratio of these two durations is the CVR and this metric is different because the CAS group had less variation in the duration of speech events and more variation in the duration of pause events, when looking at effect sizes, although the differences were not statistically different, comparing speakers with normal speech, speech delay, and CAS. So maybe not a difference at all, and if so, a difference in what?
A related paper, also by Shriberg and colleagues, introduces the lexical stress ratio which seems self-explanatory. The abstract to this paper says that the upper and lower extremes of this ratio were associated with CAS (compared to speech delay) and that the results reflect “the prosodic consequences of a praxis deficit in speech motor control.” I defy any reader to find the evidence for that conclusion in the tables and figures. Seriously. Even after you get through the problems with reliable classification of the children and the stress patterns. I should stop citing this paper.
Munson et al also looked at lexical stress, comparing instrumental measures and perceptual judgments. Their conclusions are interesting. Instrumental measures such as vowel duration and fundamental frequency revealed no differences between the speech production of children with CAS versus phonological disorder. There were some differences detected on the basis perceptual judgment. The authors speculate that the greater number of articulation errors in the speech produced by the CAS group accounts for this finding. This I think is a very important observation. It is difficult to equalize these groups of children on all the variables that may account for listener judgments.
Finally, Velleman and Shriberg conducted a very careful study of lexical stress differences, comparing the production of various metrical patterns by children with CAS or speech delay. Lexical stress errors were coded by ear, specifically identifying syllable omissions and vowel augmentations. The frequency of these errors was roughly similar across the groups in this study although the errors persisted to a greater age in the groups of children with apraxia of speech.
So, my point is that these papers provide weak evidence of a deficit in prosody as a core feature of CAS —that is very weak to no evidence. People hear something that sounds like a problem with prosody when they listen to these children. Acoustic measures are more likely to reveal a difference than perceptual evidence. Comparison to children with normal speech development is more likely to reveal a difference than comparison to children with speech delay/phonological disorder. It has been suggested that the impression of prosodic disruptions may stem from the greater number of articulation errors in the children’s speech.
I have a question about this issue of prosodic disruptions. Let’s say that there are prosodic disruptions (we know that at least some children have difficulty producing normal sounding lexical stress). How do they arise in the child’s speech? Is the cause the same in every child’s speech? Several of the studies suggest a difficulty with a high level stage of speech planning and lexical retrieval (see Kircher et al 2003). Here it is important to find the right word and plan the syllables in the right order with the right stress pattern. Young children and even adults produce errors due to the selection of wrong words or wrong syllable templates during phonological planning. Subsequently, planning and execution of the speech gestures to produce the intended syllables should result in the utterance. However, mistiming of these elements – loudness, duration, pitch – may result in anomalies: syllables that are longer than they are supposed to be or syllable breaks that are misplaced or syllables produced with roughly equal pitch. Two children who produce speech with disrupted prosody might do so for the same reason or for different reasons. A clear standard for measuring and comparing prosody in children might help to sort that out.
References in order of appearance:
Miller, H. E., Ballard, K. J., Campbell, J., Smith, M., Plante, A. S., Aytur, S. A., & Robin, D. A. (2021). Improvements in Speech of Children with Apraxia: The Efficacy of Treatment for Establishing Motor Program Organization (TEMPOSM). Developmental Neurorehabilitation, 24(7), 494-509. doi:10.1080/17518423.2021.1916113
Ballard, K. J., Robin, D. A., & McCabe, P. (2010). A treatment for dysprosody in childhood apraxia of speech. Journal of Speech, Language & Hearing Research, 53, 1227-1245.
Nijland, L., Maassen, B., van der Meulen, S., Gabreëls, F., Kraaimaat, F. W., & Schreuder, R. (2003). Planning of syllables in children with developmental apraxia of speech. Clinical Linguistics & Phonetics, 17(1), 1-24. doi:10.1080/0269920021000050662
Shriberg, L. D., Green, J. R., Campbell, T. F., McSweeny, J. L., & Scheer, A. R. (2003). A diagnostic marker for childhood apraxia of speech: the coefficient of variation ratio. Clinical Linguistics & Phonetics, 17, 575–595.
Shriberg, L. D., Campbell, T. F., Karlsson, H. B., Brown, R. L., Mcsweeny, J. L., & Nadler, C. J. (2003). A diagnostic marker for childhood apraxia of speech: the lexical stress ratio. Clinical Linguistics & Phonetics, 17(7), 549 – 574.
Munson, B., Bjorum, E. M., & Windsor, J. (2003). Acoustic and perceptual correlates of stress in nonwrods produced by children with suspected developmental apraxia of speech and children with phonological disorder. Journal of Speech, Language, and Hearing Research, 46, 189-202.
Velleman, S. L., & Shriberg, L. D. (1999). Metrical analysis of the speech of children with suspected developmental apraxia of speech. In Journal of Speech, Language, and Hearing Research (Vol. 42, pp. 1444-1460).
Kircher, T. T. J., Brammer, M. J., Levelt, W., Bartels, M., & McGuire, P. K. (2004). Pausing for thought: engagement of left temporal cortex during pauses in speech. NeuroImage, 21(1), 84-90. doi:https://doi.org/10.1016/j.neuroimage.2003.09.041
Developmental Phonological Disorders 