Stability and composition of functional synergies for speech movements in children with developmental speech disorders

Available online at www.sciencedirect.com

Journal of Communication Disorders 44 (2011) 59–74

Stability and composition of functional synergies for speech movements in children with developmental speech disorders H. Terband a,b,c,d,e,*, B. Maassen a,b,c,d,e, P. van Lieshout f,g,h,i, L. Nijland a,c a Medical Psychology, Radboud University Nijmegen Medical Centre, Nijmegen, The Netherlands Pediatric Neurology Centre, Radboud University Nijmegen Medical Centre, Nijmegen, The Netherlands c ENT, Radboud University Nijmegen Medical Centre, Nijmegen, The Netherlands d Centre for Language and Cognition, University of Groningen, Groningen, The Netherlands e School of Behavioural and Cognitive Neurosciences, University Medical Centre, University of Groningen, Groningen, The Netherlands f Department of Speech-Language Pathology, Oral Dynamics Lab, University of Toronto, Toronto, Canada g Department of Psychology, University of Toronto, Toronto, Canada h Institute of Biomaterials and Biomedical Engineering, University of Toronto, Toronto, Canada i Toronto Rehabilitation Institute, Toronto, Canada b

Received 24 March 2010; received in revised form 25 June 2010; accepted 5 July 2010

Abstract The aim of this study was to investigate the consistency and composition of functional synergies for speech movements in children with developmental speech disorders. Kinematic data were collected on the reiterated productions of syllables spa (/spa+/) and paas (/pa+s/) by 10 6- to 9-year-olds with developmental speech disorders (five with speech sound disorder [SSD] and five with subtype childhood apraxia of speech [CAS]) and six normally speaking children using electro-magnetic midsagittal articulography (EMMA). Results showed a higher variability of tongue tip movement trajectories and a larger contribution of the lower lip relative to the jaw in oral closures for the five children with CAS compared to normally developing controls, indicating that functional synergies for speech movements in children with CAS may be both delayed and less stable. Furthermore, the SSD group showed a composition of tongue tip movements that is different from both CAS and controls. These results suggest that the differences in speech motor characteristics between SSD and subtype CAS are qualitative rather than quantitative. At the same time, the results suggest that both SSD and subtype CAS increase movement amplitude as an adaptive strategy to increase articulatory stability. Although in direct comparison no exclusive characteristics were found to differentiate subtype CAS from the group of children with SSD and from normally developing children, these preliminary results are promising for quantifying the role of speech motor processes in childhood speech sound disorders. Learning outcomes: The reader will be able to: (1) describe the development of speech motor control and explain the role of functional synergies/coordinative structures; (2) explain the measurement of the stability and composition of speech movements; (3) identify the difficulties in studying disordered speech motor development; (4) describe the differences in speech motor characteristics between SSD and subtype CAS; (5) describe the potential role of motor control strategies in developmental speech disorders. # 2010 Elsevier Inc. All rights reserved. Keywords: Speech motor control; Speech motor development; Developmental speech disorders; Childhood apraxia of speech; Functional synergies; Coordinative structures; Speech movement patterns

* Corresponding author at: Department of Neurolinguistics & University Medical Centre, University of Groningen, Oude Kijk in’t Jatstraat 26, Groningen 9712 EB, The Netherlands, Tel.: +31 50 3635800. E-mail address: [email protected] (H. Terband). 0021-9924/$ – see front matter # 2010 Elsevier Inc. All rights reserved. doi:10.1016/j.jcomdis.2010.07.003

60

H. Terband et al. / Journal of Communication Disorders 44 (2011) 59–74

1. Introduction The diagnosis of childhood apraxia of speech (CAS) constitutes one of the main challenges in the field of pediatric speech pathology. Despite the substantial disagreement about the diagnostic differentiation between CAS and other developmental speech sound disorders (SSD), most authors do agree that the core impairments for each disorder should be localized at a different level of speech production. Whereas SSD in general is thought to involve primarily phonological processes, the subtype CAS is usually seen as a speech motor disorder with a core deficit in planning and/or programming the spatiotemporal parameters of movement sequences (ASHA, 2007; Hall, Jordan, & Robin, 2007; Maassen, Nijland, & Terband, 2010; Shriberg, 2010). The distinction between phonological and speech motor production processes thus is a central issue in the diagnostic differentiation between CAS and other developmental speech disorders. Isolating motor processing levels from phonological (and other higher cognitive) levels proves rather complicated in general, but especially in the developmental stage. Both in normal as in disordered development, cognitive modules are rather the outcome of development than the starting point. The progression to the adult system is a gradual and continuous process comprising interactions between emerging modules (Karmiloff-Smith, 2006; Karmiloff-Smith, Scerif, & Ansari, 2003; Karmiloff-Smith & Thomas, 2003). A specific underlying impairment in speech motor planning and/or programming may interfere with the development on other levels of processing, such as phonological and linguistic processing levels or motor execution (Maassen, 2002; Maassen et al., 2010). As a result, the symptoms of CAS do not directly reflect a deficit on one particular stage or level of sensory-motor processing and the differential diagnosis of CAS requires detailed information on the developmental history of the child, combined with performance indicators on specific speech motor tasks and fine-grained acoustic and/or kinematic measurements (Maassen et al., 2010; Terband & Maassen, 2010). 1.1. Normal speech motor development One approach to the study of speech motor development is to focus on the consistency and stability of movement patterns (Smith, 2006; Smith, Johnson, McGillem, & Goffman, 2000). At the lowest level of speech motor control, the development of speech motor coordination entails the development of functional synergies of muscle activations (or coordinative structures). In this way, the degrees of freedom are reduced, which makes the control task simpler. As a result, the dynamic coordination among orofacial structures becomes more consistent as the speech motor system matures (Bernstein, 1967; Kelso, Tuller, Vatikiotis-Bateson, & Fowler, 1984; Thelen, 1991; Thelen, Kelso, & Fogel, 1987). Measuring coordination and movement variability in speech production thus provides a way of assessing the progression of speech motor development. Movement variability can be expressed by a variety of variability measures for kinematic movement parameters such as amplitude, duration and peak velocity. These measures, however, index the movement trajectories for a given speech task (e.g., bilabial closure) at specific, single points in time, reducing the amount of information that is potentially available in the entire trajectory. These variables may also be less robust in dealing with more extreme variations (outliers) and are not consistent in direct comparisons for a given speaker and task (Alfonso & Van Lieshout, 1997; Van Lieshout & Namasivayam, 2010). To examine the entire movement trajectory over time, Smith and colleagues developed the spatiotemporal variability index (STI, Smith, Goffman, Zelaznik, Ying, & McGillem, 1995; Smith et al., 2000). The STI captures the variability in movement patterns of selected motion trajectories of repeated utterances. To this end, the movement trajectories are time and amplitude normalized. A target movement template is estimated as the mean displacement amplitude value on the normalized time axis and each movement track is matched against the template along the time axis at a certain number of intervals. The STI is then calculated by summing the standard deviations across the individual tracks. A lower STI value indicates a smaller deviation from the target movement template, and thus less variability (Lucero, 2005; Smith et al., 1995, 2000). A number of studies have investigated speech motor development utilizing the STI. Smith and Goffman (1998) compared the stability of movement trajectories in two groups of eight children (aged 4 and 7 years) with a same sized group of young adults. The STI was measured over the movement trajectories of the lower lip in 15 repetitions of the sentence Buy Bobby a puppy.1 Results showed a higher STI in the lower lip movement trajectories of the 4-year-olds as 1

More specifically ‘‘the interval from the peak velocity of the first opening movement from/b/to/aI/in buy to the peak velocity of the last opening movement for the/p/to/i/in puppy’’ (Smith & Goffman, 1998, p. 21).

H. Terband et al. / Journal of Communication Disorders 44 (2011) 59–74

61

compared to the adults. Regarding 7-year-olds, the results were not that clear as the differences between the 7-yearolds and the adults did not reach statistical significance. In a second study that used exactly the same subjects, Goffman and Smith (1999) investigated the stability of lower lip movements during the production of the words van, fan, ban, pan, and man. With these stimuli, results showed a significant difference in STI of lower lip movements among all three groups (4-year-olds > 7-year-olds > adults). This second study also investigated movement variability by means of the coefficients of variation of amplitude, peak velocity, and duration of lower lip movements. The results on these measures showed a similar pattern of movement variability decreasing with age. The notion of infants’ articulatory movement trajectories and movement coordination becoming more stable during linguistic/phonemic development has recently been confirmed in a longitudinal study (Grigos, 2009; Grigos, Saxman, & Gordon, 2005). Jaw, lower lip, and upper lip movement variability (STI) and voice onset time (VOT) were investigated in the productions of the phoneme [p] in the word papa in six children from the age of 19 months during the acquisition of the voicing contrast. Results showed that movement variability decreased with time, whereas VOT increased as children acquired the voiceless [p]. Although the coupling between lip opening/closing and glottal opening/closing was not measured, these findings suggest that the coordinative organization between the oral and laryngeal articulators is refined as children acquire the voicing contrast (and their phonemic system develops). Instead of focusing on single articulators as in the previous studies Smith and Zelaznik (2004) investigated the consistency of the inter-articulator relationships upper lip–lower lip (lip aperture) and lower lip–jaw. STI was measured over whole repeated sentences produced by six groups of 30 children and adults ranging from 4 to 22 years of age. Results showed that also at the phrase-level variability decreases with age. Furthermore, speech motor development was found to be nonlinear. Results showed that the time course of the development of functional synergies is protracted. Variability was found to be higher in the 4- and 7-year-olds as compared to adults. Then, the 7to 12-year period was characterized by a plateau in development and the stability of the functional synergies for speech movements only started to be adult-like after age 14. A second approach to investigate the development of functional synergies focuses on the composition of speech movements. In other words, this approach looks at the large developmental changes in the relation of different components that constitute articulatory movements. For example, mandibular oscillations are prominently present from babbling onset and the jaw has been found to be predominant in early speech production. More specifically, the mandible is thought to provide the foundation for acquiring control over the other articulators, which are more specialized (MacNeilage & Davis, 1990; Moore, 2004). Dissociation of lip and tongue movement from the mandible develops as the child’s phonetic system expands (Green & Nip, 2010; MacNeilage & Davis, 1990; Moore, 2004). This notion has been confirmed in various behavioral experiments. An acoustic study by Nittrouer (1993) investigated schwa-stop-vowel utterances in three groups of ten children (3.5, 5, and 7 years of age) and ten adults. Detailed analyses of formant frequencies suggested that jaw movements achieve adult-like qualities by the age of 3.5 years, while tongue gestures were found to be more constrained by phonetic context in all three groups of children as compared to adults. These results show that jaw gestures mature earlier than tongue gestures, which is consistent with basic principles of motor control since the task of controlling the tongue for speech is more complex (more degrees of freedom). Green, Moore, Higashikawa, and Steeve (2000) investigated the development of jaw and lip coordination using kinematic measurements with a video-based movement tracking system. Productions of bilabial consonants in the words baba, papa, and mama were recorded in three groups of children (six 1-year-olds, ten 2-year-olds, and ten 6year-olds) and ten adults. Results showed that the relative contribution to oral closures differed significantly across age groups. More specifically, they found that in 1- and 2-year-olds, the jaw contributed most to the oral closure, whereas in the 6-year-old children and the adults, lower lip and jaw contributed equally. A follow-up study further examined the developmental course of upper lip, lower lip, and jaw control using an index of similarity between the articulator movements of children in three age groups (five 1-year-olds, nine 2-year-olds, and ten 6-year-olds) and those produced by a group of ten adults (Green, Moore, & Reilly, 2002). The results on crosscorrelations of the children’s movement trajectories with the average adult movement trajectory showed that the jaw movements of the 1- and 2-year-olds were significantly more adult-like than their upper and lower lip movements. Six-year-olds produced articulatory movements similar to those of the adults. Together, these studies indicate that during development the functional synergies for speech movements change dramatically in their composition. Infants (1- and 2-year-olds) rely largely on the jaw in the realization of oral closures, where 6–7-year-olds show a more adult-like differentiation between lips and jaw, a process that still undergoes

62

H. Terband et al. / Journal of Communication Disorders 44 (2011) 59–74

refinement from there on (Green et al., 2000, 2002; Green & Nip, 2010). Moreover, the results suggest that the skills of controlling the different articulators do not develop uniformly, but have distinct developmental schedules in which the development of the jaw precedes the development of upper- and lower lip contributions in speech tasks. Recently, both consistency and composition of the tongue tip–jaw and lower lip–jaw functional synergies were investigated in a single study comparing a group of six 7-year-old children and a group of eight adults (Terband, van Brenk, van Lieshout, Nijland, & Maassen, 2009). Two target utterances (spa and paas) were recorded with electromagnetic midsagittal articulography (EMMA) in a reiterated speech task. Results showed higher variability in children compared to adults for tongue tip and jaw, but not for lower lip movement trajectories. Furthermore, the relative contribution to the oral closure of lower lip was smaller in children compared to adults, whereas in this respect no difference was found for tongue tip. These results support and extend the findings that speech motor development follows different trajectories for different articulators. For the lower lip–jaw synergy, children rely to a large extent on closing the jaw in the realization of oral closures, where adults show a more distinct contribution of the lower lip. A different pattern was found for the tongue tip–jaw synergy. Results showed an adult-like contribution of the tongue tip in the realization of oral constrictions in 7-year-olds compared to adults, but with larger variability. Without data on the contribution of the tongue tip in the realization of oral constrictions in younger children, it is not possible make inferences about the development of the composition of the tongue tip–jaw functional synergy. Obtaining data of tongue tip movements in younger children constitutes a challenge for further research. However, the results on the variability of movement trajectories suggest that the tongue tip–jaw synergy stabilizes later in development than the lower lip–jaw synergy. 1.2. Speech motor control in developmental speech disorders The stability and composition of functional synergies for speech movements has not yet been studied in children with developmental speech disorders, and studies that utilized kinematic measurements with this population are relatively scarce in general. For children with CAS, acoustic studies have found a reduction in vowel distinction and deviant coarticulation patterns as compared to normally developing controls. In a study by Nijland et al. (2002), the most prevalent result was a larger within-subject variability in the children with CAS, but the speech motor behavior was not found to be typically immature. Previous studies have also found that children with CAS make more substitution errors in producing a CCV-sequence in a cluster than when the consonants are separated by a syllable boundary (Maassen, Nijland, & Van der Meulen, 2001; Nijland et al., 2003). Furthermore, the adjustment of the durational structure as a function of syllabic organization was found to be less systematic (more variable) in children with CAS than in normally developing children. However, detailed acoustic analysis revealed no effects of syllabic structure on coarticulation or the variability of coarticulation, neither for the normally developing children nor for the group of children with CAS. In a follow-up study exploring the kinematics and dynamics of closing movements of tongue tip and lower lip in children with developmental speech disorders, the group of children with CAS showed larger kinematic stiffness2 values than controls for CVC, but not for CCV sequences for both articulators (Terband, van Brenk, Maassen, & Nijland, 2008). All these results were interpreted as indicating that children with CAS have a deficit in planning syllables at the motor level. 1.3. Aim of the present study The current study sets out to explore the stability and composition of functional synergies for speech movements in children with developmental speech disorders. A group of ten 6- to 9-year-olds with speech sound disorders (SSD) is compared with six age-matched normally developing children. Among the group of children with SSD, five children exhibited predominantly apraxic characteristics and were labeled subtype CAS. The differential diagnosis was based on currently used clinical criteria (see Section 2 for more details). The aim of the present study is to investigate the speech motor components that lie underneath the clinical symptoms of CAS and SSD. 2 The stiffness of a movement is defined as the peak velocity value divided by the corresponding amplitude value and as such provides information on the potential use of a certain motor control strategy in the context of a mass-spring model (Munhall, Ostry, & Parush, 1985). From the massspring model perspective, the stiffness coefficient can be seen to represent two things: (1) a control parameter for movement cycle duration (Munhall et al., 1985); (2) actuator ‘‘stiffness’’, a relative index of the level of muscle activity (in terms of co-contraction) that is used to produce a movement (Nelson, Perkell, & Westbury, 1984; Ostry, Keller, & Parush, 1983; Perkell, Zandipour, Matthies, & Lane, 2002).

H. Terband et al. / Journal of Communication Disorders 44 (2011) 59–74

63

1.4. Research questions and hypotheses The current literature presented above shows that variability decreases with maturation and during linguistic/ phonemic development (Goffman & Smith, 1999; Grigos, 2009; Smith & Goffman, 1998; Smith & Zelaznik, 2004). For children with CAS, acoustic studies have found a larger within-subject variability as compared to normally developing children (Maassen et al., 2001; Nijland et al., 2002, 2003). Based on these previous findings, we predict the variability of movement trajectories to be higher as apraxic characteristics are more predominant (i.e. CAS > SSD > controls). Since children with CAS have been found to have more difficulty in producing consonants in a cluster than consonants that are separated by a syllable boundary (Maassen et al., 2001; Nijland et al., 2003), this effect is expected to be stronger for CCV- than for CVC-sequences. Furthermore, previous findings suggest that the stabilization of the lower lip–jaw synergy precedes the stabilization of the tongue tip–jaw synergy (Terband, van Brenk, et al., 2009). Therefore, we expect the group effect to be larger for tongue tip as compared to lower lip movement trajectories. With respect to the composition of articulatory movements, we do not have clear predictions for the pathological groups. For normal development it has been shown that infants rely largely on the jaw in the realization of oral closures and dissociation of lip and tongue movement from the mandible comes with development (Green & Nip, 2010; MacNeilage & Davis, 1990; Moore, 2004). Based on the notion that a disorder disrupts development, we expect to find a stronger reliance on the jaw as apraxic characteristics are more predominant (i.e. CAS > SSD > controls). Since for children of around 7 years-old compared to adults the relative contribution to the oral closure was found to be smaller for lower lip, but not for tongue tip (Terband, van Brenk, et al., 2009), we expect to find this effect (smaller relative contributions) especially for lower lip. 2. Methods and materials 2.1. Participants Ten 6- to 9-year-olds with developmental speech disorder (6;2–8;9 years;months, mean 7.51, SD 0.99; 6 males and 4 females) participated in the study. Six aged-matched normally developing children (6;4–9;8 years;months, mean 7.47, SD 1.24; 3 males and 3 females) functioned as controls. The children with developmental speech disorder were recruited from a primary special school for children with severe speech, language and hearing impairment. Inclusioncriteria were conform Nijland (2009). No participants suffered from hearing problems (pure-tone thresholds not exceeding 25 dBHL), language comprehension problems (a score less than 1 SD below population average), subnormal intelligence (IQ less than 1 SD below population average), organic disorders in the orofacial area, gross motor disturbances, or dysarthria. 2.2. Diagnosis of SSD and subtype CAS Among the children with SSD, a subgroup of five children exhibiting predominantly apraxic characteristics was distinguished (subtype CAS; 7;2–8;9 years;months, mean 7.83, SD 0.82; 2 males and 3 females). The remaining five children with SSD exhibited predominantly phonological deficits (SSD; 6;2–8;9 years;months, mean 7.32, SD 1.11; 4 males and 1 female). The diagnosis of the children with speech sound disorders was established using standardized language perception and production tests. Furthermore, the children’s speech therapists were asked to provide information about speech characteristics experienced by the child in the present and in the past. A detailed description of the diagnostic tasks can be found in Appendix B. The selection of the children with subtype CAS was based on the procedure described by Nijland (2009). Three variables were used to differentiate between CAS and SSD:  Clinical judgment: based on semi-spontaneous utterances, picture naming, word repetition and the oral-motor movement assessment, each child was clinically diagnosed as having CAS or SSD.  The percentage of CAS-characteristics and SSD-characteristics (in the present or past) as reported by the speech therapists. Based on existing literature (e.g., ASHA, 2007; Dodd, 2005; Hall et al., 2007; Stackhouse, 1992), we defined two sets of speech characteristics that are typical for CAS and SSD. The CAS-characteristics were:

64

H. Terband et al. / Journal of Communication Disorders 44 (2011) 59–74

unintelligible speech for parents and others; inconsistency in articulation errors; groping or searching articulation (both prevocalic groping or silent posturing, and searching articulatory behavior during sound production, Hall et al., 2007; Terband, Maassen, Guenther, & Brumberg, 2009), slow progress in therapy, and articulation errors comprising simplifications but also more complex patterns. The SSD-characteristics were defined as: unintelligible speech for others, but not for parents; consistency in articulation errors; and consonantal errors mostly comprising simplifications. Percentages from 67% and upwards were qualified as ‘‘high’’.  The abilities in a maximum performance task (diadochokinesis). Inability was concluded when children could not produce [pataka] in isolation or in a sequence of two repetitions. Classification was implemented as follows:  (clinical judgment = CAS or SSD/CAS) + ([pataka] could not be produced) ! CAS classification  (clinical judgment = CAS or SSD/CAS) + (high % CAS-characteristics) + ([pataka] in sequence, but no acceleration) ! CAS classification  (clinical judgment = SSD or SSD/CAS) + (no evident [pataka] problems) ! SSD classification  (clinical judgment = SSD or SSD/CAS) + (high % SSD-characteristics) + ([pataka] in sequence incorrect, but speeding up on two different consonants) ! SSD classification A detailed overview of the diagnostic data can be found in Appendix C. 2.3. Speech task and data collection The research paradigm followed a previous study (Terband, van Brenk, et al., 2009) and consisted of a reiterated speech task comprising the Dutch words spa (/spa+/: 1. mineral water; 2. spade; 3. a luxurious health and beauty resort) and paas (/pa+s/: (1) the adjective for Easter; (2) in Dutch, phonetically equivalent to the soccer term pass). The use of minimal word pairs that consist of the same gestures, but in a different order or phase relationship (Browman & Goldstein, 1992), provides a window on the role of syllabic structure in the coordination of speech movements. Differences in word production based on possible differences in semantic categories (/spa+/ is a noun, whereas /pa+s/ can be an adjective and a noun) are not to be expected. We focused on movement data of fluent and perceptually correct utterances (the selection procedure is elaborated in detail in Section 2.6). The stability of speech motor execution was assessed by the cyclic spatiotemporal variability index (cSTI) of the movement trajectories of jaw, lower lip and tongue tip, which captures the variability of direction specific, cyclic movement patterns (Van Lieshout, Bose, Square, & Steele, 2007; Van Lieshout, Rutjens, & Spauwen, 2002). The composition of the tongue tip–jaw and the lower lip–jaw synergies was assessed by isolating the amplitude components of tongue tip and lower lip from the total articulatory movement in the realization of the oral constrictions for respectively the [s] and the [p]. Additionally, we investigated the amplitude of jaw opening during the [a+]. 2.4. Procedures Articulatory data were collected using an AG100 Carstens electro-magnetic midsaggittal articulograph (EMMA) with time-aligned audio signal (Carstens Medizinelektronic GmbH). With this setup, it is possible to obtain 2dimensional articulatory data in a biologically non-hazardous manner (Hasegawa-Johnson, 1998; Van Lieshout & Moussa, 2000). Position data were sampled at 400 Hz, acoustic data at 16 kHz. Experimental procedures followed the same protocol as used in previous studies from our labs (e.g. Nijland, Maassen, Hulstijn, & Peters, 2004; Terband, van Brenk, et al., 2009; Van Lieshout et al., 2002, 2007). A helmet with three transmitter coils was placed on the participants’ head. Transducer coils were attached in midline position to the upper and lower lip, jaw (lower incisors), tongue tip (1 cm behind the actual tongue tip), and tongue body (2 cm behind the tongue tip coil). Two additional coils were placed on the nose bridge and on the gums of the upper incisors for reference purposes. The coils were attached using surgical tissue glue (tongue and gums; Henkel Indermil) or micropore sticky tape (lips and nose). After coil attachment (but before data collection), the participants were asked to answer a few questions about their daily pastime to enable them to familiarize themselves with the presence of the coils on their articulators. During data collection, stimuli were presented on a computer screen. Repetitions of the two one-syllable words spa (/spa+/) and

H. Terband et al. / Journal of Communication Disorders 44 (2011) 59–74

65

paas (/pa+s/) were recorded in either 5 s or 12 s trials. The participants’ task was to repeat the syllable at a self-chosen normal, comfortable pace. Each trial was preceded by a preparation interval in which the participants were told what syllable they had to repeat. If necessary, the experimenter modeled the syllable once or twice. The participants were instructed to take a deep breath to allow him/her to repeat the stimulus for the length of the trial in a single breath (if possible). After each trial, the acoustic speech sample was played back over a connected speaker system and the validity of the trial was judged by the experimenter. In the case of clear phonemic production errors, pauses, interruptions, or rate accelerations and decelerations, the trial was repeated at the end of the series. Each session took about 45–60 min per child on average. 2.5. Data processing Data processing followed a standardized protocol (Van Lieshout & Moussa, 2000) also used in previous studies completed in our labs (e.g. Nijland et al., 2004; Terband, van Brenk, et al., 2009; Van Lieshout et al., 2002, 2007). Prior to analysis, raw articulatory data were corrected to compensate for variations in helmet positions and rotational misalignments, i.e. ‘twist’ and ‘tilt’ movements of the head using custom made software provided by the manufacturer (Hoole & Nguyen, 1997). Corrected position data over time for all individual articulators were stored on a computer hard drive, separate for vertical (Y) and horizontal dimensions (X). Subsequently, peaks and valleys in the position and velocity signals were marked by an algorithm using relative amplitude (10% of maximum amplitude) and time (a minimum interval of 0.5 s between successive events) criteria. In some cases where peaks and valleys were incorrectly identified by the algorithm, peak/valley assignment was corrected manually. The peak-to-peak or valley-to-valley segment is referred to as a movement cycle (each cycle corresponds to a single repetition of the syllable). To calculate cSTI, individual direction-specific movement cycles (defined by the peaks and valleys in the signal) were amplitude- and time-normalized and aligned. At 2% intervals in relative time, separate standard deviations were then computed for the overlapping segments. The sum of these standard deviations constitutes the cSTI, thus providing a measure for the uniformity of cyclic movement patterns (Van Lieshout, 2004; Van Lieshout et al., 2002, 2007). Kinematic parameters were derived from the position data for every movement cycle. In the current study we focused on movement amplitudes of tongue tip and lower lip in the realization of the constrictions for respectively the [s] and the [p] and the amplitude of jaw opening during the [a+]. The tongue tip and lower lip signals were corrected for jaw movement to isolate the amplitude component of lower lip and tongue tip from the combined articulatory movement. The correction was made using an estimate of jaw rotation based on the principal component of the mandible coil trajectory (Westbury, Lindstrom, & McClean, 2002). Thus, the tongue tip and lower lip kinematics reflect predominantly the unique contributions of these articulators for the production of [p] and [s], respectively. 2.6. Data analysis All data were screened for visible movement artifacts. Furthermore, all trials were screened for the presence of dysfluencies and only portions of the data that were spoken in a perceptually correct and fluent manner were selected. The selection was made by consensus between two trained phoneticians that are experienced with pathological speech, one of whom is the first author. Based on this screening, the tongue tip data in /spa+/ of participant 120 (SSD) were discarded. A very large amplitude component (54.05 mm) and erratic movement trajectory indicated a loose transmitter coil. Furthermore, all data from participant 112 (CAS) for the target /pa+s/ were discarded because of the presence of dysfluencies. The number of fluent utterances or cycles per trial differed per child and per stimulus and was not equally distributed over groups. The children in the control group on average produced 6.0 (SD 2.66) fluent cycles per trial, as compared to respectively 3.9 (SD 1.62) and 4.4 (SD 1.56) for the SSD and subtype CAS groups. Based on these numbers, the analyses were limited to a maximum number of five cycles for all subjects if available. Separate analyses were performed for individual movement effectors (lower lip, tongue tip, and jaw). First, we used a repeated measures analysis of variance, with group (control, SSD, and CAS) as between-subject factor and syllable structure (CVC [/pa+s/vs.] CCV [/spa+/]) as within-subjects factor. However, the assumption of homogeneity of variance failed in all cases but one, as the Levene’s tests turned out to be significant. Although it is statistically possible to perform corrections (e.g. Nijland, Maassen, & Van der Meulen, 2003; Nijland et al., 2003), this always causes a loss of power (Max & Onghena, 1999; Quene´ & van den Bergh, 2004). A solution for this problem can be found in a Linear

66

H. Terband et al. / Journal of Communication Disorders 44 (2011) 59–74

Table 1 Means and standard deviations for the variability of movement trajectories (cSTI) and the contribution of articulators to the oral closure (amplitude), differentiated per group, articulator, and syllable structure. cSTI

Amplitude (mm)

paas

Control

SSD

CAS

Tongue tip Lower lip Jaw Tongue tip Lower lip Jaw Tongue tip Lower lip Jaw

spa

paas

spa

Mean

SD

Mean

SD

Mean

SD

Mean

SD

8.11 8.56 8.12 17.32 8.64 12.45 17.11 9.68 11.61

2.31 4.43 2.07 5.94 1.34 4.834 13.09 5.96 7.50

9.63 6.70 7.51 8.13 11.88 9.57 18.82 8.41 11.28

4.38 2.18 2.67 1.21 6.90 3.61 8.79 3.50 6.89

5.57 3.49 11.48 9.88 6.47 14.78 5.62 11.04 11.39

2.92 1.83 3.17 5.09 3.72 9.44 1.69 4.88 6.91

5.74 4.20 11.32 12.12 6.63 18.38 5.95 9.45 14.25

2.25 1.71 4.78 7.92 7.19 8.15 1.69 5.27 7.70

Mixed Model analysis, which is a form of general linear model that does not assume homogeneity of variance, sphericity, or compound symmetry, and, moreover, allows for missing data (Quene´ & van den Bergh, 2004). Therefore, statistical testing was done by means of a Linear Mixed Model, with subject and syllable structure as correlated terms, and group and syllable structure as fixed factors, with the level of significance set at .05. 3. Results Means and standard deviations for the variability of movement trajectories (cSTI) and the contribution of articulators to the oral closure (amplitude) are given in Table 1. For the variability of movement trajectories, statistical analysis revealed a main effect of group for the cSTI of tongue tip movement trajectories [F(2, 22.549) = 4.573, p < .05]. Pairwise comparisons showed a significantly higher cSTI for CAS children compared to the control group (Fig. 1). All other comparisons did not reach statistical significance. Furthermore, the results revealed no main or interaction effects involving syllable structure and no effects with respect to the movement trajectories of lower lip and jaw.

Fig. 1. Mean variability of movement trajectories of tongue tip, lower lip, and jaw in the words /pa+s/ and /spa+/ in children with SSD and subtype CAS compared to normally developing controls. Statistically significant differences are marked by an asterisk.

H. Terband et al. / Journal of Communication Disorders 44 (2011) 59–74

67

Fig. 2. Contribution of articulators to functional synergies for speech movements in children with SSD and subtype CAS in comparison with normally developing controls. Bars represent mean amplitudes of the net movement components of tongue tip and lower lip in the realization of the constrictions for respectively the [s] and the [p], and the amplitude of jaw opening during the [a+] in the words /pa+s/ and /spa+/. Statistical significance is marked by an asterisk.

With respect to the composition of functional synergies for speech movements, statistical analysis showed main effects of group for the amplitude component of tongue tip in the realization of the constriction for the [s] [F(2, 23.148) = 5.450, p < .05] and lower lip in the realization of the closure for the [p] [F(2, 23.087) = 5.703, p = .01]. Pairwise comparisons revealed the tongue tip amplitude component to be significantly larger for the SSD group than for both the CAS and the control groups (Fig. 2). Furthermore, the children with CAS were found to have a significantly larger amplitude component of the lower lip as compared to the control group. No other significant differences between groups were found. No effects were observed for jaw opening during the [a+] and the results revealed no main or interaction effects involving syllable structure.

4. Discussion 4.1. Summary of findings The purpose of this study was to investigate the consistency and composition of functional synergies for speech movements in children with developmental speech disorder as compared to age-matched normally developing controls. In the group of the children with SSD, we identified a group with subtype CAS, exhibiting predominantly apraxic characteristics. Based on previous acoustic studies (Maassen et al., 2001; Nijland et al., 2002, 2003), the variability of movement trajectories was predicted to be higher as apraxic characteristics are more predominant (i.e. CAS > SSD > controls). Furthermore, the effect was expected to be stronger for CCV- than for CVC-sequences and for tongue tip as compared to lower lip movement trajectories. With respect to the composition of functional synergies, we expected to find smaller relative contributions to oral closures again for children who show predominantly apraxic characteristics, especially for lower lip movements. In direct comparison, statistical tests revealed no exclusive significant differences between CAS and the other children with SSD. However, we did find a larger amplitude component of tongue tip for SSD compared to both CAS and controls. Furthermore, the results yielded differences in the characteristics of functional synergies for CAS compared to controls that were not found for SSD when compared to controls. First, there was a larger variability of tongue tip movement trajectories in children with CAS compared to normally developing controls. The SSD group showed an intermediate level of variability, as predicted, but this was not significantly different from either CAS or

68

H. Terband et al. / Journal of Communication Disorders 44 (2011) 59–74

controls. Previous studies have shown variability to decrease with maturation and during linguistic/phonemic development (Goffman & Smith, 1999; Grigos, 2009; Smith & Goffman, 1998; Smith & Zelaznik, 2004). The increased variability in tongue tip movement trajectories that we found for CAS in the present study, therefore, does not necessarily reflect a core impairment, but could also reflect developmental delay. In addition, the lack of variability between groups in the lower lip and jaw movement trajectories suggests that the lower lip and jaw movement trajectories stabilize earlier in development than the tongue tip. Further research on tongue tip movement trajectories in younger children is needed to clarify these issues, even though the use of EMMA with younger children is a major challenge. Second, contrary to our predictions, the children with CAS exhibited a larger contribution of the lower lip to the oral closure of the [p] compared to controls. Again, the SSD group showed intermediate results, but not significantly different from either CAS or controls. This larger contribution of the lower lip for CAS compared to controls does not seem to correspond to developmental delay. Behavioral experiments have shown that infants rely mainly on the jaw in the realization of oral closures (Green et al., 2000, 2002). The contribution of lower lip in the realization of oral closures increases as the speech production system matures. Children exhibit a more adult-like differentiation between lip and jaw at around 6–7 years of age, but this process still undergoes refinement from there on (Green et al., 2000, 2002). In a previous study, we found the contribution of the lower lip to the oral closure for the [p] to be still smaller for normally developing children at the age of 7 as compared to adults (Terband, van Brenk, et al., 2009). 4.2. Motor control strategies In addition to the larger contribution of the lower lip in the production of /p/ for CAS compared to controls, the current results did not show any differences in the stability of movement trajectories of the lower lip-jaw functional synergy. These findings suggest that the increase in amplitude of lower lip movements might reflect an adaptive strategy. Previous studies with children and adults have shown a direct relationship between movement amplitude and control stability, indicating that larger movements are potentially stabilizing (Van Lieshout et al., 2002, 2007). Therefore, the present findings may indicate that the larger lower lip amplitudes produced by the children with CAS may have helped them to achieve relatively stable movement coordination.3 In this respect, the results on the tongue tip–jaw functional synergy provide an interesting contrast. Movement trajectories of the tongue tip–jaw synergy were found to be less stable for CAS compared to controls, and results revealed no differences in the composition of tongue tip movements. The question that arises is why children with CAS do not use the same adaptive strategy (increase movement amplitude) for tongue–tip movements. This question becomes even more relevant since this appears to be exactly what the present results indicate for children in the SSD group (which is discussed in more detail below). A possible explanation can be found in the different characteristics of the speech sounds that were investigated. The sound [p] is a bilabial stop whereas [s] is a fricative, and apart from the obvious difference in vocal tract location, these sounds differ in constriction degree. This latter aspect is important as it sets completely different requirements for the gestural control task. In the case of a fricative, the closing movement requires accurate control of the articulator’s velocity in order for the articulator to achieve the appropriate narrow degree of constriction. A stop, however, does not necessarily require accurate control of the closing speed. In this case, the endpoint of the closing movement is formed by a solid body (i.e. the upper lip or the hard palate) and the moving articulator can be stopped at the right degree of constriction (full closure) by the solid body. As a result, closing movements can be more ballistic, i.e. the movement can be carried out with a signal initiating the movement but without a subsequent signal to control the degree of deceleration of the articulator (see also Lo¨fqvist & Gracco, 1997, 2002). Especially in such cases, larger movement amplitudes are a profitable strategy, containing the advantages of a stronger neural signal4 without the possibility of increasing the difficulty of the control task. The present findings also indicate a positive relationship between movement amplitude and control stability for tongue tip in the group of children with SSD. The results showed a larger amplitude component of tongue tip for SSD 3

On a side note, the strategy of increasing control stability by larger movements might provide an explanation for the slowing down of speech rate that is generally observed in disordered speech. Movement duration (and also peak velocity) necessarily increases when the amplitude is increased while maintaining the same acceleration. 4 A stronger neural signal is less susceptible to interference of noise and makes it easier to maintain a stable coordination between efferent and afferent signals in rhythmic movements, i.e., the neural oscillator and the involved effector system (e.g., Beek, Peper, & Daffertshofer, 2002; Kelso et al., 1998; Van Lieshout, 2004; Williamson, 1998).

H. Terband et al. / Journal of Communication Disorders 44 (2011) 59–74

69

compared to both CAS and controls, in combination with cSTI values of tongue tip movement trajectories that are equal to controls and smaller than children with CAS. The picture that thus emerges is consistent with the common conception in speech motor development that the control over the jaw stabilizes first, followed by the lips, and then the tongue. For the children with CAS, the results indicate that they have difficulties with the control of both lower lip and tongue tip. In the case of lower lip, control is stabilized by increasing movement amplitude, but this strategy is not applied to tongue tip movement control for the production of fricatives (as shown by the larger cSTI values). The children with SSD have more advanced motor control of their articulatory movements. This group demonstrates stable lower lip movements, but still has difficulties controlling tongue tip. In contrast to the CAS group, the children with SSD are able to stabilize tongue tip control for the production of fricatives by increasing movement amplitude, balancing this strategy with the requirements for a precise narrow constriction. Where they have problems and when it is in their reach, the children with CAS and SSD appear to use the same strategy. Further research is needed to acquire a more precise image of the relationship between movement amplitude and control stability during (disordered) development. 4.3. Syllable structure Since previous studies have found that children with CAS have more difficulties in producing the CCV-sequence than the CVC-sequence, variability was predicted to be higher for the CCV-sequence compared to the CVC-sequence as apraxic characteristics are more predominant. However, no effect of syllabic structure was found. This might be due to data selection criteria. In the present study, only perceptually correct utterances were selected for analysis. Although previous studies have found children with CAS to make more substitution errors in producing a CCV-sequence compared to a CVC-sequence, detailed acoustic analysis of the utterances that were perceptually fluent revealed no effects of syllabic structure for coarticulation, nor for the variability of coarticulation (Nijland et al., 2003). Furthermore, the children with CAS showed no systematic adaptation of durational structure dependent on syllabic organization. In the light of these results, the current lack of an effect of syllabic structure makes sense. This is emphasized by the fact that we also did not find any effects of syllable structure for the normally developing children in the present study. Although Nijland and colleagues did not find any differences with respect to coarticulation, they did find a systematic adjustment of durational structure depending on syllabic organization for the normally developing children (Nijland et al., 2003). However, there is an important methodological difference between this study and the present study with respect to speech task. In the case of Nijland and colleagues, the target utterances were embedded in a sentence, whereas in the current study, we used a reiterated speech task. The fact that we did not find any differences for normally developing children might suggest that the reiterative speech task did not tap into syllabic programming. A previous study from our lab that utilized the same reiterated speech task (/pa+s/ vs /spa+/) also did not reveal any effect of syllable structure for the control group for closing movements of tongue tip and lower lip on a number of kinematic (amplitude, peak velocity, duration), dynamic (stiffness, cSTI) and gestural coordination (mean relative phase and phase variability) parameters (Terband et al., 2008). It is possible that the task was not difficult enough for the control group to reveal any differences. However, it should be noted that in the above-mentioned previous study, we did find an effect of syllabic structure on the kinematic stiffness of tongue tip and lower lip closing movements for the group of children with CAS. In other words, although reiterated speech tasks might not tap into phonological preparation stages, they do potentially reflect the impact of changes in linguistic requirements on motor execution. The conclusion is that there are limitations that need to be considered when using reiterative speech tasks to study motor control in relation to higher phonological and linguistic processing levels. 4.4. Differential diagnosis What do the present findings mean for the diagnosis of developmental speech sound disorders and, more specifically, the differential diagnosis of subtype CAS? At the level of functional synergies for speech movements, clear differences between groups were demonstrated. Although far from a single diagnostic criterion that can be used clinically, these findings are promising for the issue of differential diagnostics. The results showed differences in the stability of tongue tip movements and differences in the composition of lower lip movements for CAS compared to controls that were not found for SSD compared to controls. Furthermore, the results showed differences between SSD and both CAS and controls in the composition of tongue tip movements, but no differences were found between CAS

70

H. Terband et al. / Journal of Communication Disorders 44 (2011) 59–74

and controls. The results thus constitute a double dissociation at a particular stage in development and in that sense provide a welcomed contribution to the profiling of CAS. An unavoidable problem inherent to the search for diagnostic criteria to differentiate pathological groups is definitional circularity (ASHA, 2007; Shriberg, 2010). The study participants were diagnosed in part based on their speech motor characteristics and the results show differences in the same realm of speech motor behavior. The fact that we find a double dissociation between CAS and SSD in the current study indicates that we may be on the right track, but it is clear that the findings reported in this paper should be considered tentative. Another limitation that needs to be considered is formed by the small group sizes. First, we only acquired data from ten children with developmental speech disorders, five in each subgroup (SSD and subtype CAS), and six normally developing controls. Although consistent within each group, the results need to be verified with a larger number of children. Second, the participants in the current study comprehend a wide but single age-range (6–9 years). Our results suggest a similarity between SSD and subtype CAS in the adaptive strategy that is used to increase articulatory stability (increase movement amplitude). Furthermore, the specific double dissociation of speech motor characteristics that we found for these groups of children at this particular stage in development suggests that the differences between SSD and subtype CAS are qualitative rather than quantitative. With the change in speech motor characteristics, the differences between groups may shift during development. In this context, it is essential to study the relationship between movement amplitude and control stability in these clinical groups both earlier and later in development. Finally, it should be noted that the reiterated speech tasks in the present study revealed only speech motor control aspects. To get to a more complete profile of CAS and SSD, relations with higher level planning processing must be studied. In conclusion, the present study shows that examining the consistency and composition of functional synergies for speech movements may provide valuable insight into the mechanisms involved in developmental speech disorders. The paradigm is promising as a tool for studying the differential characteristics of children with developmental speech disorders. Acknowledgements This study was funded by the Netherlands Organization for Scientific Research (NWO). We gratefully thank all children, their parents, speech therapists, and schools for participating in this study. Appendix A. Continuing education 1. The development of functional synergies is a way of the speech motor system to: (a) reduce the redundancy of the control task; (b) reduce the specificity of the control task; (c) reduce the degrees of freedom of the control task; (d) all of the above. 2. The progression of speech motor development can be assessed by: (a) measuring the kinematic characteristics of individual articulators; (b) measuring the stability of functional synergies for speech movements; (c) measuring the composition of functional synergies for speech movements; (d) both (b) and (c). 3. Which of the following statements is true (more than one may apply)? (a) A specific underlying impairment in speech motor planning and/or programming may interfere with the development on other levels of processing. (b) The evolving nature of clinical symptoms constitutes one of the main difficulties in the diagnosis of developmental speech disorders. (c) The differentiation between involvement of motor planning and phonological processes requires detailed information on the developmental history of the child. (d) None of the above statements is true. 4. Which of the following statements is true (more than one may apply)? (a) Control stability increases with decreasing movement amplitude.

H. Terband et al. / Journal of Communication Disorders 44 (2011) 59–74

71

(b) As they do not tap into phonological preparation stages, reiterated speech tasks do not reflect the impact of changes in linguistic requirements on motor execution. (c) Developmental speech disorders differ from normal speech motor development in that the control over the jaw stabilizes first, followed by the lips, and then the tongue. (d) None of the above statements is true. 5. The results of this study indicate that: (a) the children with SSD use larger movement amplitudes as an adaptive strategy to stabilize movement coordination; (b) the children with subtype CAS use larger movement amplitudes are used as an adaptive strategy to stabilize movement coordination; (c) both (a) and (b); (d) neither (a) nor (b).

Appendix B. Description of the diagnostic tasks (Nijland, 2009). Task/assessment

Description

Apraxic characteristics

Unintelligible speech for parents and others; inconsistency in articulation errors; groping or searching articulation; slow progress in therapy; articulation errors comprise simplifications but also more complex patterns. Score is the percentage of the characteristics present. Unintelligible speech for others, but not for parents; consistency in articulation errors; consonantal errors mostly comprising simplifications. Score is the percentage of the characteristics present. Auditory discrimination task from the Dutch translation of the PALPA (Bastiaanse, Bosje, & Visch-Brink, 1995). Score is proportion correct. 36 pairs of CVC—words that were either the same (18 pairs), differed on one consonant (initial or final; 12 pairs), or were metatheses of each other (6 pairs, e.g. ‘lor’ vs. ‘rol’). 36 pairs of CVC-non-words that were either the same (18 pairs), differed on one consonant (initial or final; 12 pairs), or were metatheses of each other (6 pairs, e.g. ‘tus’ vs. ‘sut’). Percentage correctly executed oral-motor tasks. Positioning of the lips in rest; lip protrusion; lip spreading; stick out the tongue; move the tongue to corners of the mouth; move tongue up; move tongue down; click with the tongue. Score is proportion correct. Lips protrude and spread; move tongue left and right; move tongue up and down; open and close the jaw; check whether velum closes when blowing; check whether velum closes when sucking. Score is proportion correct. Same tasks as sequential, but in higher tempo. Maximum performance task using utterances of [pataka]. The children were first asked to produce ‘pataka’ once and when they succeeded they were asked to produce ‘pataka’ in a sequence of several repetitions of ‘pataka’. After that the children were asked to speed up while producing a sequence of ‘pataka’. 0 = [pataka] could not be produced; 1 = [pataka] could be produced. 0 = perfect; 1 = [pataka] in sequence in normal rate, but no acceleration; 2 = [pataka] in sequence incorrect ([t] or [k] could not be pronounced), but speeding up on two different consonants ([pata], [taka]) was possible; 3 = no fluent [pataka], not in sequence; 4 = no [pataka] production either in isolation or in a sequence of two. This task consisted of 50 images depicting words with different consonants, consonant clusters and vowels at various positions (initial, medial, final). Most words are derived from the Dutch word list based on the research by Hodson and Paden (1981). Same task as Picture naming 1, but with ten words with complex consonant patterns. Repetition task using the same ten words as in Picture naming 2. The words were presented through headphones and the children were asked to repeat them. Same task as Word repetition, using ten non-word stimuli with complex consonant patterns. The non-words had similar syllable structures as the words and consisted of syllables that did not exist as words in Dutch. Backing; abnormal stopping; denasalization; h-zation. Score is the proportion of atypical consonantal substitutions relative to the total number of consonantal substitutions. CV; VC; CVC; CCV; VCC; CVCC; CCVC; CCVCC. Score is proportion correct.

Phonological characteristics Palpa Words Non-words Oral-motor movement assessment Isolation Sequential

Seq. fast Diadochokinesis

PTK-score PTK-judgment

Picture naming 1 (50 words logo-art)

Picture naming 2 (10 words) Word repetition (10 words) Non-word repetition (10 non-words) Atypical substitution processes Syllable structures correct

Subject ID

Subject ID

122 123 111 112 114 110 115 119 120 121 203 209 210 214 222 200

CAS CAS CAS CAS CAS SSD SSD SSD SSD SSD Control Control Control Control Control Control

Age (year;month)

7;3 7;3 8;9 7;7 7;2 8;2 8;9 6;5 7;1 6;2 7;11 7;4 7;3 9;8 6;4 6;4

Classification

CAS CAS CAS CAS CAS SSD SSD SSD SSD SSD Control Control Control Control Control Control

Sex

Clinical judgment

Male Female Female Female Male Male Male Female Male Male Female Male Female Male Female Male

CAS CAS CAS CAS CAS SSD SSD SSD SSD/CAS SSD – – – – – –

Palpa

Apraxic characteristics (%)

Phonological characteristics (%)

Diadochokinesis (pataka)

Oral-motor movement assessment

Score

Judgment

Isolation (proportion correct)

Sequential (proportion correct)

Seq. fast (proportion correct)

100 100 80 100 80 80 100 80 100 20 0 0 0 0 0 0

67 67 0 100 100 67 100 100 67 0 0 0 0 0 0 0

0 1 1 0 0 1 0 1 1 1 1 1 1 1 1 –

3 3 2 4 4 2 2 3 0 0 0 0 2 0 0 –

1 0.96 0.67 0.86 1 1 0.86 0.93 1 1 0.86 1 1 0.79 1 –

1 0.83 1 0.5 0.92 0.92 1 0.83 1 0.83 0.83 1 1 1 0.92 –

0.8 0.4 – – – – – – – – – – – – – –

Picture naming 1 (50 words logo-art)

Picture naming 2 (PN2; 10 words)

Word repetition (WR; 10 words same as PN2)

Non-word repetition (10 non-words similar to WR)

Words (proportion correct)

Non-words (proportion correct)

Atypical substitution processes (proportion)

Syllable structures (proportion correct)

Atypical substitution processes (proportion)

Syllable structures (proportion correct)

Atypical substitution processes (proportion)

Syllable structures (proportion correct)

Atypical substitution processes (proportion)

Syllable structures (proportion correct)

0.89 0.83 – – – – – – – – – – – – – –

0.97 0.89 0.89 0.58 0.47 0.83 0.89 0.83 0.92 0.97 0.83 0.97 0.81 1 0.92 –

0.03 0 0 0 0.01 0.02 0.03 0.02 0 0 0 0 0 0 0 –

0.92 0.95 0.94 0.77 0.77 0.79 0.88 0.62 0.79 0.85 0.83 0.85 0.81 0.80 0.84 –

0 0 0 0 0 0 0.06 0 0 0 0 0 0 0 0 –

0.86 0.73 0.81 0.67 0.62 0.71 0.79 0.63 0.73 0.7 0.92 0.97 0.95 0.96 0.89 –

0 0.04 0 0 0.03 0 0.09 0 0 0 0 0 0 0 0 –

0.78 0.87 0.96 0.56 0.79 0.92 0.8 0.71 0.83 0.88 0.92 1 0.83 1 0.87 –

0.04 0 0.1 0.04 0.23 0.08 0 0.04 0.05 0 0.03 0.04 0 0 0 –

0.65 0.68 0.92 0.75 0.65 0.74 0.93 0.52 0.68 0.78 0.88 0.92 0.85 1 0.64 –

H. Terband et al. / Journal of Communication Disorders 44 (2011) 59–74

122 123 111 112 114 110 115 119 120 121 203 209 210 214 222 200

Classification

72

Appendix C. Overview of the children who participated in the study. (A description of the diagnostic tasks is presented in Appendix B)

H. Terband et al. / Journal of Communication Disorders 44 (2011) 59–74

73

References Alfonso, P. J., & Van Lieshout, P. H. H. M. (1997). Spatial and temporal variability in obstruent gestural specification by stutterers and controls: Comparisons across sessions. In W. Hulstijn, H. F. M. Peters, & Speech production: Motor control, brain research, and fluency disorders (pp. 151–160). Amsterdam: Elsevier Publishers. ASHA (2007). Childhood apraxia of speech [technical report]. American Speech-Language-Hearing Association. Available from www.asha.org/ policy. Bastiaanse, R., Bosje, M., & Visch-Brink, E. G. (1995). PALPA: 1995. Dutch adaptation of Kay J, Lesser R, Coltheart M. Psycholinguistic assessment of language processing in aphasia. Hove, UK: Lawrence Erlbaum Associates Ltd. Beek, P. J., Peper, C. E., & Daffertshofer, A. (2002). Modeling rhythmic interlimb coordination: Beyond the Haken–Kelso–Bunz model. Brain and Cognition, 48(1), 149–165. Bernstein, N. (1967). The coordination and regulation of movements. Oxford: Pergamon. Browman, C. P., & Goldstein, L. M. (1992). Articulatory phonology: an overview. Phonetica, 49(3–4), 155–180. Carstens Medizinelektronic GmbH. AG100. Dodd, B. (Ed.). (2005). Differental diagnosis and treatment of children with speech disorder (2nd ed.). London: Whurr. Goffman, L., & Smith, A. (1999). Development and phonetic differentiation of speech movement patterns. Journal of Experimental Psychology: Human Perception and Performance, 25(3), 649–660. Green, J. R., Moore, C. A., Higashikawa, M., & Steeve, R. W. (2000). The physiologic development of speech motor control: Lip and jaw coordination. Journal of Speech Language and Hearing Research, 43(1), 239–255. Green, J. R., Moore, C. A., & Reilly, K. J. (2002). The sequential development of jaw and lip control for speech. Journal of Speech, Language and Hearing Research, 45(1), 66–79. Green, J. R., & Nip, I. S. B. (2010). Some organization principles in early speech development. In B. Maassen & P. Van Lieshout (Eds.), Speech motor control: New developments in basic and applied research (pp. 171–188). Oxford: Oxford University Press. Grigos, M. I. (2009). Changes in articulator movement variability during phonemic development: A longitudinal study. Journal of Speech, Language and Hearing Research, 52(1), 164–177. Grigos, M. I., Saxman, J. H., & Gordon, A. M. (2005). Speech motor development during acquisition of the voicing contrast. Journal of Speech, Language and Hearing Research, 48(4), 739–752. Hall, P., Jordan, L., & Robin, D. (Eds.). (2007). Developmental apraxia of speech: Theory and clinical practice (2nd ed.). Austin, TX: Pro-ed. Hasegawa-Johnson, M. (1998). Electromagnetic exposure safety of the Carstens articulograph AG100. The Journal of the Acoustical Society of America, 104(4), 2529–2532. Hodson, B. W., & Paden, E. P. (1981). Phonological processes which characterize unintelligible and intelligible speech in early childhood. Journal of Speech and Hearing Disorders, 46(4), 369–373. Hoole, P., & Nguyen, N. (1997). Electromagnetic articulography in coarticulation research. Forschungsberichte des Instituts fu¨r Phonetik und Sprachliche Kommunikation der Universita¨t Mu¨nchen, 35, 177–184. Karmiloff-Smith, A. (2006). The tortuous route from genes to behavior: A neuroconstructivist approach. Cognitive, Affective, and Behavioral Neuroscience, 6(1), 9–17. Karmiloff-Smith, A., Scerif, G., & Ansari, D. (2003). Double dissociations in developmental disorders? Theoretically misconceived, empirically dubious. Cortex, 39(1), 161–163. Karmiloff-Smith, A., & Thomas, M. (2003). What can developmental disorders tell us about the neurocomputational constraints that shape development? The case of Williams syndrome. Development and Psychopathology, 15(4), 969–990. Kelso, J. A. S., Tuller, B., Vatikiotis-Bateson, E., & Fowler, C. A. (1984). Functionally specific articulatory cooperation following jaw perturbations during speech: Evidence for coordinative structures. Journal of Experimental Psychology: Human Perception and Performance, 10(6), 812– 832. Kelso, J. A. S., Fuchs, A., Lancaster, R., Holroyd, T., Cheyne, D., & Weinberg, H. (1998). Dynamic cortical activity in the human brain reveals motor equivalence. Nature, 392, 814–818. Lo¨fqvist, A., & Gracco, V. L. (1997). Lip and jaw kinematics in bilabial stop consonant production. Journal of Speech, Language, and Hearing Research, 40(4), 877–893. Lo¨fqvist, A., & Gracco, V. L. (2002). Control of oral closure in lingual stop consonant production. The Journal of the Acoustical Society of America, 111(6), 2811–2827. Lucero, J. C. (2005). Comparison of measures of variability of speech movement trajectories using synthetic records. Journal of Speech, Language and Hearing Research, 48(2), 336–344. Maassen, B. (2002). Issues contrasting adult acquired versus developmental apraxia of speech. Seminars in Speech and Language, 23(4), 257–266. Maassen, B., Nijland, L., & Van der Meulen, S. (2001). Coarticulation within and between syllables by children with developmental apraxia of speech. Clinical Linguistics and Phonetics, 15(1–2), 145–150. Maassen, B., Nijland, L., & Terband, H. (2010). Developmental models of childhood apraxia of speech. In B. Maassen & P. Van Lieshout (Eds.), Speech motor control: New developments in basic and applied research (pp. 243–258). Oxford: Oxford University Press. MacNeilage, P. F., & Davis, B. F. (1990). Acquisition of speech production: Frames, then content. In M. Jeannerod (Ed.), Attention and performance XIII: Motor representation and control (pp. 453–476). Hillsdale, NJ: Lawrence Erlbaum. Max, L., & Onghena, P. (1999). Some issues in the statistical analysis of completely randomized and repeated measures designs for speech, language, and hearing research. Journal of Speech, Language and Hearing Research, 42(2), 261–270. Moore, C. A. (2004). Physiologic development of speech production. In B. Maassen, R. Kent, H. F. M. Peters, P.H.H.M. van Lieshout, & W. Hulstijn (Eds.), Speech motor control in normal and disordered speech (pp. 191–211). Oxford, UK: Oxford University Press.

74

H. Terband et al. / Journal of Communication Disorders 44 (2011) 59–74

Munhall, K. G., Ostry, D. J., & Parush, A. (1985). Characteristics of velocity profiles of speech movements. Journal of Experimental Psychology: Human Perception and Performance, 11(4), 457–474. Nelson, W. L., Perkell, J. S., & Westbury, J. R. (1984). Mandible movements during increasingly rapid articulations of single syllables: Preliminary observations. The Journal of the Acoustical Society of America, 75(3), 945–951. Nijland, L., Maassen, B., Van der Meulen, S., Gabree¨ls, F., Kraaimaat, F. W., & Schreuder, R. (2002). Coarticulation patterns in children with developmental apraxia of speech. Clinical Linguistics and Phonetics, 16(6), 461–483. Nijland, L., Maassen, B., & Van der Meulen, S. (2003a). Evidence of motor programming deficits in children diagnosed with DAS. Journal of Speech, Language and Hearing Research, 46(2), 437–450. Nijland, L., Maassen, B., Van der Meulen, S., Gabree¨ls, F., Kraaimaat, F. W., & Schreuder, R. (2003b). Planning of syllables in children with developmental apraxia of speech. Clinical Linguistics and Phonetics, 17(1), 1–24. Nijland, L., Maassen, B., Hulstijn, W., & Peters, H. F. M. (2004). Speech motor coordination in Dutch-speaking children with DAS studied with EMMA. Journal of Multilingual Communication Disorders, 2(1), 50–60. Nijland, L. (2009). Speech perception in children with speech output disorders. Clinical Linguistics and Phonetics, 23(3), 222–239. Nittrouer, S. (1993). The emergence of mature gestural patterns is not uniform: Evidence from an acoustic study. Journal of Speech and Hearing Research, 36(5), 959–972. Ostry, D. J., Keller, E., & Parush, A. (1983). Similarities in the control of the speech articulators and the limbs: Kinematics of tongue dorsum movement in speech. Journal of Experimental Psychology: Human Perception and Performance, 9(4), 622–636. Ostry, D. J., & Munhall, K. G. (1985). Control of rate and duration of speech movements. The Journal of the Acoustical Society of America, 77(2), 640–648. Perkell, J. S., Zandipour, M., Matthies, M. L., & Lane, H. (2002). Economy of effort in different speaking conditions. I. A preliminary study of intersubject differences and modeling issues. The Journal of the Acoustical Society of America, 112(4), 1627–1641. Quene´, H., & van den Bergh, H. (2004). On multi-level modeling of data from repeated measures designs: A tutorial. Speech Communication, 43(1– 2), 103–121. Shriberg, L. D. (2010). A Neurodevelopmental framework for research in childhood apraxia of speech. In B. Maassen & P. Van Lieshout (Eds.), Speech motor control: New developments in basic and applied research (pp. 259–270). Oxford: Oxford University Press. Smith, A. (2006). Speech motor development: Integrating muscles, movements, and linguistic units. Journal of Communication Disorders, 39(5), 331–349. Smith, A., & Goffman, L. (1998). Stability and patterning of speech movement sequences in children and adults. Journal of Speech, Language and Hearing Research, 41(1), 18–30. Smith, A., Goffman, L., Zelaznik, H. N., Ying, G., & McGillem, C. (1995). Spatiotemporal stability and patterning of speech movement sequences. Experimental Brain Research, 104(3), 493–501. Smith, A., Johnson, M., McGillem, C., & Goffman, L. (2000). On the assessment of stability and patterning of speech movements. Journal of Speech, Language and Hearing Research, 43(1), 277–286. Smith, A., & Zelaznik, H. N. (2004). Development of functional synergies for speech motor coordination in childhood and adolescence. Developmental Psychobiology, 45(1), 22–33. Stackhouse, J. (1992). Developmental verbal dyspraxia. I. A review and critique. European Journal of Disorders of Communication, 27(1), 19–34. Terband, H., & Maassen, B. (2010). Speech motor development in childhood apraxia of speech (CAS): Generating testable hypotheses by neurocomputational modeling. Folia Phoniatrica et Logopaedica, 62, 134–142. Terband, H., van Brenk, F., Maassen, B., & Nijland, L. (2008). Articulation in children with developmental speech disorders. Poster presentation at the 14th biennial Conference on Motor Speech, Monterey, CA, March 6–9 2008. Terband, H., Maassen, B., Guenther, F. H., & Brumberg, J. (2009a). Computational neural modeling of childhood apraxia of speech (CAS). Journal of Speech, Language and Hearing Research, 52(6), 1595–1609. Terband, H., van Brenk, F., van Lieshout, P. H. H. M. , Nijland, L., & Maassen, B. (2009b). Stability and composition of functional synergies for speech movements in children and adults. Proceedings of Interspeech, 2009, 788–791. Thelen, E. (1991). Motor aspects of emergent speech: A dynamic perspective. In N. A. Krasnegor, D. M. Rumbaugh, R. L. Schiefelbusch, & M. Studdert-Kennedy (Eds.), Biological and behavioral determinants of language development (pp. 339–362). Hillsdale, NJ: Lawrence Erlbaum Associates. Thelen, E., Kelso, J. S., & Fogel, A. (1987). Self-organizing systems and infant motor development. Developmental Review, 7(1), 39–65. Van Lieshout, P.H.H.M.. (2004). Dynamical systems theory and its application in speech. In B. Maassen, R. Kent, H. F. M. Peters, P.H.H.M. van Lieshout, & W. Hulstijn (Eds.), Speech motor control in normal and disordered speech (pp. 51–82). Oxford, UK: Oxford University Press. Van Lieshout, P. H. H. M. , Rutjens, C. A. W., & Spauwen, P. H. M. (2002). The dynamics of interlip coupling in speakers with a repaired unilateral cleft-lip history. Journal of Speech, Language, and Hearing Research, 45(1), 5. Van Lieshout, P. H. H. M. , Bose, A., Square, P. A., & Steele, C. M. (2007). Speech motor control in fluent and dysfluent speech production of an individual with apraxia of speech and Broca’s aphasia. Clinical Linguistics and Phonetics, 21(3), 159–188. Van Lieshout, P. H. H. M. , & Moussa, W. (2000). The assessment of speech motor behavior using electromagnetic articulography. The Phonetician, 1, 9–22. Van Lieshout, P. H. H. M. , & Namasivayam, A. K. (2010). Speech motor variability in people who stutter. In B. Maassen & P. Van Lieshout (Eds.), Speech motor control: New developments in basic and applied research (pp. 191–214). Oxford, UK: Oxford University Press. Westbury, J. R., Lindstrom, M. J., & McClean, M. D. (2002). Tongues and lips without jaws: A comparison of methods for decoupling speech movements. Journal of Speech, Language and Hearing Research, 45(4), 651–662. Williamson, M. M. (1998). Neural control of rhythmic arm movements. Neural Networks, 11, 1379–1394.