Behavioral and neurobiological correlates of childhood apraxia of speech in Italian children

Brain & Language 150 (2015) 177–185

Contents lists available at ScienceDirect

Brain & Language journal homepage: www.elsevier.com/locate/b&l

Short Communication

Behavioral and neurobiological correlates of childhood apraxia of speech in Italian children q Anna Maria Chilosi a,⇑, Irene Lorenzini b, Simona Fiori a, Valentina Graziosi a, Giuseppe Rossi c, Rosa Pasquariello a, Paola Cipriani a, Giovanni Cioni a,d a

Dipartimento di Neuroscienze dell’Età Evolutiva, IRCCS Fondazione Stella Maris, Viale del Tirreno 331, 56128 Pisa, Italy Laboratorio di Linguistica ‘Giovanni Nencioni’, Scuola Normale Superiore, Piazza dei Cavalieri 7, 56126 Pisa, Italy Unità di Epidemiologia e Biostatistica, Istituto di Fisiologia Clinica Consiglio Nazionale delle Richerche, Via G.Moruzzi 1, 56124 Pisa, Italy d Dipartimento di Medicina Clinica e Sperimentale, Università di Pisa, Via Roma 67, 56126 Pisa, Italy b c

a r t i c l e

i n f o

Article history: Received 12 November 2014 Revised 27 August 2015 Accepted 24 October 2015

Keywords: Childhood apraxia of speech Speech Sound Disorders Maximum performance task Language disorders Magnetic resonance imaging Array-Comparative Genomic Hybridization

a b s t r a c t Childhood apraxia of speech (CAS) is a neurogenic Speech Sound Disorder whose etiology and neurobiological correlates are still unclear. In the present study, 32 Italian children with idiopathic CAS underwent a comprehensive speech and language, genetic and neuroradiological investigation aimed to gather information on the possible behavioral and neurobiological markers of the disorder. The results revealed four main aggregations of behavioral symptoms that indicate a multi-deficit disorder involving both motor-speech and language competence. Six children presented with chromosomal alterations. The familial aggregation rate for speech and language difficulties and the male to female ratio were both very high in the whole sample, supporting the hypothesis that genetic factors make substantial contribution to the risk of CAS. As expected in accordance with the diagnosis of idiopathic CAS, conventional MRI did not reveal macrostructural pathogenic neuroanatomical abnormalities, suggesting that CAS may be due to brain microstructural alterations. Ó 2015 Published by Elsevier Inc.

1. Introduction Childhood apraxia of speech (CAS) is a subtype of pediatric speech-language disorder, defined by the American-Speech-Lan guage-Hearing-Association (2007) as ‘‘a neurological childhood disorder in which the precision and consistency of movements underlying speech are impaired in the absence of neuromuscular deficits”. CAS is usually interpreted as a speech motor disorder, whose core deficit involves the planning and/or programming of the spatiotemporal parameters of movement sequences (ASHA, 2007). Children with CAS display reduced speech timing and sequencing skills and show particular difficulties in dynamic transitions between articulatory postures and in combining smaller units of movement into larger ones. Early oromotor and speech acquisition q Statement of significance to neurobiology of language: This manuscript investigates behavioral and neurobiological characteristics of idiopathic childhood apraxia of speech. The goals were to explore the pattern of relationships between speech and language symptoms of CAS and their etiological correlates, in order to gather information on possible clinical markers of the disorder. ⇑ Corresponding author. E-mail address: [email protected] (A.M. Chilosi).

http://dx.doi.org/10.1016/j.bandl.2015.10.002 0093-934X/Ó 2015 Published by Elsevier Inc.

difficulties in CAS may stem from the lack of a probable innate ability to form systematic mappings between articulatory gestures and their auditory effects (Maassen, Nijland, & Terband, 2010). In a recent study Shriberg, Lohmeier, Strand, and Jakielski (2012), suggested that CAS is a multi-level disorder in which both planning/ programming (transcoding) and auditory- perceptual (encoding) deficits are involved, together with memory processes. According to the ASHA consensus criteria (ASHA, 2007), three features are characteristic of CAS (a) inconsistent errors on consonants and vowels during repeated productions of syllables or words, (b) lengthened and disrupted co-articulatory transitions between sounds and syllables, and (c) inappropriate prosody, especially in the realization of lexical or phrasal stress. Other features include reduced phonetic inventory, multiple speech sound errors, disfluency and unintelligibility (ASHA, 2007). In spite of increasing interest in the study of this disorder, as recently indicated by Murray, McCabe, Heard, and Ballard (2015), differential diagnosis of CAS from other Speech Sound Disorders remains problematic because of the lack of validated assessment protocols. Moreover, after extensive behavioral, neuroimaging and genetic studies of the KE family (whose affected members had severe verbal dyspraxia and a mutation in the FOXP2 gene,

178

A.M. Chilosi et al. / Brain & Language 150 (2015) 177–185

Vargha-Khadem, Gadian, Copp, & Mishkin, 2005), very few researches on CAS simultaneously addressed these three aspects. As for etiology, apraxia of speech may be symptomatic, cryptogenic or idiopathic. Symptomatic and cryptogenic CAS can be secondary to known neurological pathologies of a metabolic (as in galactosemia, cfr. Shriberg, Potter, & Strand, 2011 and in creatine transport deficiency, cfr. Battini et al., 2007), epileptic or genetic nature (as in children with Down syndrome). Idiopathic CAS can be the only symptom present in otherwise healthy children, occurring as a neurogenic Speech Sound Disorder whose etiology and neural correlates remain poorly understood. From a genetic point of view, landmark studies on a single large family (the KE family, cfr. Lai, Fisher, Hurst, Vargha-Khadem, & Monaco, 2001), in which about half of the members were affected by orofacial and speech apraxia, identified a mutation in the FOXP2 gene located in chromosome 7q3 (Hurst, Baraitser, Auger, Graham, & Norell, 1990). Subsequently, other cases with de novo and familial FOXP2 mutations have been described. However, the results of studies by MacDermot et al. (2005) and Laffin et al. (2012) on larger populations showed a low prevalence of FOXP2 alterations (2% and 4% respectively). Moreover, using Array-Comparative Genomic Hybridization analysis (a-CGH), Laffin et al. (2012) found several copy-number variations (on chromosomes 2, 13 and 14) in 12 of their 24 patients. Other chromosomal abnormalities have been identified with a-CGH analysis (12p13.33, Thevenon et al., 2013; 16p11.2 microdeletion, Fedorenko et al., 2015; Raca et al., 2013; 2p15p16.1 microdeletion, Peter, Matsushita, Oda, & Raskind, 2014,) and with Whole Exome Sequencing analysis (gene alterations in chromosomes 3, 6, 7, 9, 17, Worthey et al., 2013). Regarding neural correlates, CAS rarely occurs in children suffering from congenital focal brain lesions of the left hemisphere (Chilosi et al., 2008). This is different from acquired adult apraxia of speech, that is generally due to left hemisphere lesions. According to adult models of speech production, planning and execution of speech movements may rely on distinct, though interlinked, neural circuitries (Liégeois & Morgan, 2012; Ogar et al., 2006). Speech planning could involve the supplementary motor area, anterior insula, dorsolateral frontal cortex and superior cerebellum; while speech execution could be subserved by the primary motor cortex, extrapiramidal system, thalamus and inferior cerebellum. Neuroimaging investigation of affected KE family members, revealed structural and functional abnormalities in a vast cortical-subcortical network including perisylvian and rolandic cortices, caudate nucleus, inferior frontal gyrus and supplementary motor area (Liégeois, Morgan, Connelly, & Vargha-Khadem, 2011; Vargha-Khadem et al., 2005). However, these findings cannot be generalized to every case of CAS. As reported in a recent review by Liégeois and Morgan (2012), in the past 13 years only twelve articles describing 45 children with CAS were accompanied by MRI investigations, which did not reveal any significant abnormality in around 60% of cases. Research on neurobiological markers of idiopathic CAS is still an emerging field for which no conclusive information is available. Moreover, very few studies have analysed the clinical, genetic and neuroanatomical correlates of this disorder in relative large populations of patients and, in particular, no data on Italian-speaking children are available. In this study we investigated the clinical, genetic and neuroradiological characteristics of 32 Italian children affected by idiopathic CAS. Our goals were to explore the patterns of relationship between speech and language symptoms of CAS and their etiological correlates, in order to gather information on possible clinical markers of this disorder. All children underwent a comprehensive assessment that included neurological and speech and language evaluation, genetic testing (a-CGH) and structural MRI.

2. Results 2.1. Clinical and neurobiological investigation No child presented neurometabolic abnormalities at biochemical testing. At neurological assessment, gross and fine motor organization was below age expectation in 65.5% of children. Evaluation of cognitive abilities (WPPSI and WISC-III or IV) showed a mean Performance IQ of 94.5 (SD 16.4; range 57–107). PIQ was in the normal range in 23 out of 32 children and in the borderline range (70–85) in 7; only 2 subjects had a mild cognitive deficit. 2.1.1. MRI findings Structural brain MRIs revealed only minor brain abnormalities: aracnoid cysts (4 cases), lower position of cerebellar amygdales (4 cases), retrovermian space enlargement (2 cases), partial thinning of corpus callosum (2 cases), frontal venous dysplasia (1 case). Proton MR spectroscopy was normal for the entire sample. 2.1.2. Genetic findings An intra- and inter-familial heterogeneity was observed in 70% of cases, in which a CAS proband had parents or relatives with a positive history of language disorder or dyslexia, or both. Six subjects presented an alteration at a-CGH with one patient (TF) showing a de novo complex genetic abnormality. In this patient, a duplication in chromsome 9 (9p22.1p24.3) involving a very large portion (18 Mb) of the terminal part was associated with a 6 Mb deletion affecting the long arm of chromosome 4 (4q35.1.q35.2). In three cases, genetic abnormalities were inherited from one parent: one female had an interstitial duplication in chromosome 8 (8p23.1), also present in her father; in a male and in his father a microdeletion in chromosome 1 (1q21.1) was found; another male child and his mother had an interstitial duplication in chromosome 5 (5p13.3). Two heterozygous twins presented an interstitial duplication in chromosome X (Xq21.1); a-CGH analysis of their parents is still under investigation. None of the parents showed signs of CAS (they received a normal education and all were currently employed). 2.2. Case history and behavioral assessment 2.2.1. Parents’ report Pre-perinatal history was normal in 23 children; some transient problems during pregnancy or delivery were reported in the remaining 9, but none suffered from severe fetal or neonatal complications. Fourteen children (43%) suffered from recurrent otitis media during the first years of life, but none had persistent auditory deficits. Based on parental reports, all children presented severe speech and language delay characterized by: – Absent (10 cases) or quantitatively reduced and qualitatively abnormal babbling (22 cases), with delayed onset (mean age 14.3 months, SD 8 months; range 9–30 months) and sporadic production of a very restricted number of speech sounds. – Delayed emergence of first words (mean age 25.9 months, SD 11.6 months; range 12–48 months). – Very slow increase of vocabulary size with delay in acquisition of the first 50 words (mean age 54.6 months, SD 14.1; range 30–83 months) and late emergence of combinatory speech (mean age 54.5, SD 13.7; range 30–83 months). – High percentage of unintelligible speech during preschool years. The difference between the proportion of unintelligible speech for family members (mean 60%) and unfamiliar adults (mean

A.M. Chilosi et al. / Brain & Language 150 (2015) 177–185

78%) was statistically significant at the Wilcoxon test for paired data (Z = 4.152, p < 0.0001). 2.2.2. Non-Verbal and verbal oro-motor skills More than 60% of children presented a deficit in isolated nonverbal (mean z-score – 5.7, SD 6) and verbal (mean z-score – 3.4, SD 4.7) oro-motor skills, and more than 80% were impaired in the execution of sequential oral movements (mean z-score – 4.6, SD 3.3). 2.2.3. Speech measures Phonetic inventories were markedly reduced (mean number of consonants 12.6 out of 21 phonemes tested, SD 3.8) and significantly differed from normal controls (F(1.60) = 80.684, p < 0.0001). Also percentages of phonetically inaccurate productions of words (mean 76%, SD 25) and non-words (67%, SD 25) were significantly higher compared to controls (F(1.60) = 316.924, p < 0.0001 and F(1.58) = 72.284, p < 0.0001 respectively). More than one third of naming errors (38%, SD 30.04) were characterized by syllable omissions, that altered word structural length and complexity. Vowels distortions and voicing errors were also present. In addition, the percentage of phonetically inconsistent productions was greater than 64%, (SD 21.5), a value significantly higher with respect to normal controls (F(1.58) = 243.897, p < 0.0001). Evaluation of Diadochokinesis by maximum performance task (DDK) was possible in 20 patients only, since data from two children were missing and ten were not able to perform the task (these cases were assigned a score of 0). The mean DDK rate (10.1 items repeated in 20 s, SD 6.6) was significantly slower in comparison to 60 normal controls (mean 25, SD 4.7; F(1.88) = 173.924, p < 0.0001). In the 20 patients who were able to execute the maximum performance task, the number of repeated syllables per second (2 syllables/s, SD 0.4) was significantly below the means obtained by the control sample (mean 3.7, SD 0.7; F(1.78) = 112.643, p = 0.001). At post-hoc analysis, the difference was statistically significant, even in comparison to the younger control subgroup with a mean age of 4.5 yrs (mean number of repeated syllables per second 3.4, SD 0.7; post hoc comparison p = 0.001). All children, except one, exhibited clear signs of inappropriate prosody (severe in 53% of cases, mild in 43%). 2.2.4. Language measures Evaluation of language abilities revealed that 75% of children performed below age expectation in more than one language area. For receptive vocabulary this corresponded to a standard score below 75, for expressive vocabulary to more than 1.5 SD below the mean and for expressive grammar to an ungrammatical speech level. In about 60% of participants expressive grammar was the most impaired domain, being characterized by a primitive combinatorial or telegraphic speech with omission of more than 70% of free morphemes in obligatory contexts. Only three children attained an age-appropriate language level characterized by the production of syntactically well-formed simple and complex sentences with an almost full control of free and inflectional morphology. Both receptive (mean standard score 77, SD 15; range 60–110) and expressive vocabulary (mean z-score 1.8, SD 1.5; range 6 to 0.5) were better preserved, with about 32% of children scoring within normal range and 34% showing a mild delay in both lexical domains. No significant correlations were found between the number of years of speech therapy and language outcome. 2.3. Principal component analysis There were seven eigenvalues greater than 1 and the proportion of variance explained was 78%. The proportion of variance explained by the first two principal components was 46.5%.

179

2.3.1. Intra-clouds correlations On the basis of the strength of correlation among the variables included in the analysis, four main aggregations were found in the subspace of the first two principal components (see Fig. 1). Most variables within each aggregation correlated positively with each other (see Table 1) and were grouped as follows: – I Aggregation: Speech programming level. This included inaccuracy for words and non-words, inconsistency of errors, syllable omissions and dysprosody. Significant correlations were found between inaccuracy for non-words and inconsistency of errors, and between syllable omissions, inaccuracy for words and dysprosody. However, only the correlation between inaccuracy for non-words and inconsistency of errors retained statistical significance after correction for multiple comparisons (according to Benjamini and Hochberg’s approach). – II Aggregation: Oro-motor Level. This aggregation included both isolated and sequential non-verbal and verbal oro-motor skills, together with familial antecedents for speech and language disorders. Such correlations clearly identified oro-motor deficits as a highly specific component. – III Aggregation: Speech-motor and language level. This aggregation included both speech-motor variables (that is, DDK rate) and speech-language measures (as phonetic inventory, intelligibility expressive and receptive vocabulary, expressive grammar). The aggregation provides evidence in support of a strong association between speech-motor and language measures in CAS. As shown in Table 1, all the variables correlated significantly with each other. – IV Aggregation: Early developmental milestones. This included early motor delay and a set of early linguistic behaviors that typically developing children acquire within the first two years of life, following a rather predictable time-table. After correction for multiple comparisons, a statistically significant correlation was found between vocabulary increase and combinatory speech, and between age at first words and parental report of early motor delay (r = 0.682). Moreover, age at first words correlated positively with vocabulary increase (r = 0.313). The presence of MRI abnormalities and age at speech-language assessment did not emerge as components of any aggregation.

2.3.2. Inter-clouds correlations Regarding the first aggregation, word inaccuracy, dysprosody and syllable omissions were inversely correlated with all speechmotor and language measures belonging to the third aggregation (DDK rate, phonetic inventory, intelligibility, receptive and expressive vocabulary and expressive grammar). Inverse correlations were also found between inconsistency of errors, expressive grammar and (to a lesser degree) DDK. As shown in Table 1, after adjustment for multiple comparisons, not all of them retained statistical significance. A linear regression analysis showed that dysprosody explained the highest proportion of variance in intelligibility (R2 = 0.27), with a negative relationship between dysprosody and intelligibility (b = 10.1, F(1.29) = 11.319, p = 0.002). When examining the second aggregation, the only significant correlation was between isolated non-verbal oro-motor skills and DDK rate, whereas no significant correlation was observed between oro-motor measures and speech programming skills. In the third aggregation, correlations between familial antecedents for language difficulties and expressive and receptive vocabulary were found, but did not survive after correction for multiple comparisons. Finally, in the fourth aggregation, both age at vocabulary increase and onset of combinatory speech correlated negatively

180

A.M. Chilosi et al. / Brain & Language 150 (2015) 177–185

Fig. 1. Principal Component Analysis: projection of variables on the subspace of the first two components. The horizontal line is the Principal Component 1 while the vertical line is the Principal Component 2. Two components explain 46.5% of the variance of variables. Within each cloud variables with the highest correlation indices are represented. Four clouds were distinguished: (1) speech programming level, (2) oro-motor Level, (3) speech-motor and language level, and (4) early developmental milestones.

with receptive and expressive vocabulary and grammar, DDK rate and oro-motor skills.

3. Discussion This study investigated the behavioral and neurobiological characteristics of childhood apraxia of speech in Italian-speaking children. The results provide novel insights regarding speech and language profiles of Italian children with CAS and some additional information on the genetic and neuroanatomical correlates of the disorder. As for other developmental language disorders, a genetic basis for CAS was suggested by the prevalence of males (males to females ratio 4:1) and by the high familial aggregation rate for oral and written language disorders (70%). Results of the a-CGH analysis revealed the presence of genetic alterations in six cases: in three of them (whose parents underwent a-CGH analysis), the genetic abnormalities were inherited from a single parent, who appeared phenotypically normal. Chromosomal alterations did not involve the FOXP2 gene or other chromosomes that have been described in association with CAS. Only one child presented a de novo genetic abnormality characterized by a duplication in chromosome 9 (9p22.1p24.3) associated with a deletion affecting the long arm of chromosome 4 (4q35.1. q35.2). A 4q35.2 deletion has been described by Shriberg, Jakielski, and El-Shanti (2008) in three siblings with CAS, presenting an unbalanced 4q;16q translocation.

These findings, though limited to a small number of subjects, confirm that, from a genetic point of view, CAS is a highly heterogeneous condition (Laffin et al., 2012; Worthey et al., 2013), in which the pathogenetic link between genotype and phenotype is not a direct one. They also suggest that the inclusion of Array-Comparative Genomic Hybridization analysis in a clinical assessment protocol may help the diagnostic process, given the high incidence of genetic alterations in this population. The results of neuroradiological investigation extend the scarce MRI reports available in the literature, all showing a normal MRI or only minor abnormalities, as in all our cases. These findings confirm that the brain regions typically damaged in adult post-stroke apraxia of speech (e.g. Broca’s area, insula, ventral premotor cortex) generally do not show abnormalities in idiopathic CAS. As suggested by Liégeois and Morgan (2012), anatomical brain abnormalities associated with idiopathic CAS may be too subtle to be detected by standard clinical MRI sequencing methods. Preliminary results from an MRI Diffusion Tensor Imaging (DTI) study, conducted by our research group, indicate a neural dysfunction involving the bilateral connectivity of brain circuitries subserving speech in 15 children belonging to the present experimental sample (Chilosi, Cipriani, Graziosi, & Fiori, 2014; Fiori et al., submitted manuscript). Behavioral phenotype, indirect parental information and direct speech and language assessment revealed that all participants exhibited most typical features reported in CAS (ASHA, 2007; Murray et al., 2015; Shriberg et al. 2011). All the parents described

Age at combinatory speech

skills

Age at vocabulary increase

Familial antecedents

motor oro-motor skills

Isolated verbal oro-motor

Isolated non- verbal oro-

motor skills

Sequenced non-verbal oro-

Dysprosody

Word inaccuracy

Syllabic omission

Inconsistency

Non-word inaccuracy

Intelligibility

Receptive vocabulary

Expressive grammar

Phonetic inventory

Expressive vocabulary

Note: Statistically significant correlations at PCA analysis are in italics. Correlations retaining significance after adjustment for multiple comparisons (using the Benjamini and Hochberg’s approach) are in bold.

.760 (.000)

1.000

.303 (.046) .322 (.036) 1.000

1.000

.243 (.106)

.528 (.004)

.347 (.035)

.535 (.004)

.355 (032)

.294 (.081)

.332 (.042)

1.000

1.000

.482 (.003)

.385 (.022)

.341 (.028)

.441 (.006)

.365 (.028)

.356 (.023) .229 (.104)

.346 (.036) .554 (.001)

.241 (.092) .309 (.071)

.136 (.228)

.587 (.001)

.102 (.290)

.065 (.370) .044 (.413)

1.000

.014 (.469) .359 (.022)

1.000

1.000

.141 (.247)

.223 (.118) .254 (.088) .150 (.215) .207 (.136) .027 (.444) .062 (.372) .188 (.160) .374 (.021) 1.000

.010 (.479)

.205 (.134) .389 (.015)

.051 (.402)

.177 (.174) .102 (.296)

.100 (.313)

.180 (.166) .063 (.368)

.165 (.211)

.081 (.335) .185 (.164)

.089 (.332)

.097 (.302) .046 (.403)

.083 (.331)

.271 (.090) .223 (.137)

.263 (.080)

.523 (.001) .366 (.021)

.076 (.346)

.022 (.457) .299 (.069)

.236 (.105)

.365 (.022)

.623 (.000)

1.000

.016 (.467) .129 (.244)

1.000

1.000

.518 (.001)

.110 (.274)

.549 (.001) .572 (.000)

.169 (.178)

.577 (.000) .199 (.138)

.243 (.090)

.287 (.056) .288 (.055)

.103 (.288)

.332 (.032) .231 (.102)

.057 (.379)

.331 (.032) .208 (.127)

.077 (.337)

.118 (.261)

.561 (.000)

.401 (.011)

.417 (.009) .645 (.000)

.660 (.000)

.594 (.000) .571 (.000)

.653 (.000)

.442 (.006) .244 (.089)

.326 (.035)

.246 (.088)

.252 (.082)

.327 (.034)

.403 (.011)

.494 (.002)

.466 (.004)

.538 (.001)

.721 (.000)

1.000

1.000

1.000

.590 (.000)

.212 (.122)

.479 (.004)

.543 (.001) .586 (.000) .353 (.024) .274 (.064) .009 (.480) .142 (.219) .506 (.002) .399 (.012) .441 (.006) .165 (.183) .502 (.002) .464 (.004) .675 (.000) 1.000

.317 (.047)

.621 (.000)

.044 (.405)

.259 (.076) 155 (.199)

.538 (.001) .241 (.104)

.170 (.177) .187 (.153)

.275 (.075) .445 (.008)

.094 (.304) .329 (.033)

.192 (.159) .435 (.005)

.002 (.496) .037 (.421)

.355 (.029) .313 (.049)

.028 (.440) .052 (.389)

.305 (.054) .264 (.083)

.174 (.170) .187 (.153)

.209 (.125)

.205 (.130)

.748 (.000)

.069 (.354)

.430 (.010)

.047 (.399)

.604 (.000)

.461 (.006)

skills

.194 (.14)

1.000

1.000 Age (mts)

DDK

motor skills motor skills oro-motor

speech

increase antecedents verbal oroverbal oro-

combin

Age at vocab Familial Isolated Isolated non

inaccuracy omission

Syllabic Inconsistency

inaccuracy

Non-word Intelligibility Phonetic

inventory grammar

Expressive Receptive

vocabulary vocabulary

Expressive DDK

(mths)

Age

Table 1 Correlations between data from medical records, isolated and sequenced oro-motor skills, DDK, speech and language measures.

Word

Dysprosody

Sequenced

non verbal

Age at

A.M. Chilosi et al. / Brain & Language 150 (2015) 177–185

181

their children as quite silent in the first two years of life, indicating the absence or severe disruption of early vocal behavior, including spontaneous and imitative vocal play and babbling. The lack or severe alteration of reduplicated canonical babbling may be considered a very early sign of a specific deficit in transforming auditory input into motor output through vocal imitation (an ability that typically developing children acquire between 6 and 10 months of age). This symptom, often reported in literature (ASHA, 2007) has been interpreted as evidence for a motor dysfunction hypothesis. In addition, early language milestones were reported to be severely delayed with slow increase of vocabulary and late onset of combinatory speech. Moreover, speech was highly unintelligible, especially to unfamiliar adults during the preschool years. These early signs of speech-language deviance correlated with diadochokinesis and with later language proficiency. Our study, the first conducted in Italian, a language with a rich bound and free morphology, documents difficulties at the level of expressive grammar and delayed vocabulary skills, testifying that most Italian children with CAS have problems both at the speech and language level. Further evidence in support to this finding was provided by statistically significant correlations between language measures and speech-motor abilities. Results of behavioral assessment, thus, provide evidence of a multi-level dysfunction, affecting speech-programming as well as speech-motor and language abilities. Furthermore, most children showed associated signs of oral non-verbal apraxia. However, using Principal Component Analysis, we found that oro-motor difficulties did not significantly correlate with speechprogramming measures, highlighting the possible dissociation between oral and verbal apraxia. This result may carry some implications for intervention, suggesting that oral-motor rehabilitation may not necessarily help to improve speech-programming skills. On a speech programming level, correlations between measures of inaccuracy, inconsistency of errors and syllable omissions may reflect, as suggested by Ozanne (2005), a difficulty in assembling phono-articulatory motor plans, with poor maintenance of word structural length and complexity and suprasegmental alterations (dysprosody). Syllable omissions, which are considered a distinctive feature of CAS in English-speaking children (ASHA, 2007), were more frequent in more severely affected or younger participants and significantly correlated with language skills. As reported above, speech-programming measures did not correlate with oromotor skills, whereas they were inversely correlated with diadochokinesis (DDK), so that, as DDK rate increased, inaccuracy, inconsistency, syllable omissions and dysprosody decreased. DDK is a task that requires rapid sequential transitions between different articulatory plans and, as such, should be considered a clinically important test for the diagnosis of different types of Speech Sound Disorders (SSD), in particular dysarthria and dyspraxia (Thoonen, Maassen, Wit, Gabreels, & Schreuder, 1996). While in typical development, such ability appears to increase with age (William & Stackhouse, 2000), in our children we did not find any correlation between DDK rate and age. Moreover, the mean DDK rate was only 2 syllables per second, in agreement with the criteria for diagnosis of CAS according to Thoonen et al. (1996) (less than 3.4 syllables per second in children between 4 and 6 years of age, or failure to produce any correct repetition of /pataka/). This score was significantly lower even in comparison with the younger control group (mean age 4.3 yrs), who produced 3.4 mean syllables per second, a value similar to that reported by Rvachew et al. (2006) for English-speaking children aged 4-to-6 yrs. DDK rate was also correlated with isolated oro-motor skills and with language performances, that were below the mean for the children’s age at time of assessment.

182

A.M. Chilosi et al. / Brain & Language 150 (2015) 177–185

These findings suggest that diadochokinesis involves a complex set of skills that act at the interface between speech motor planning/execution and higher order levels of language organization. In conclusion, the above behavioral findings appear to confirm that CAS is a multi-deficit disorder affecting both motor-speech and language competence. According to some authors (Ozanne, 2005; Terband & Maassen, 2010; Velleman, 2011), the difficulties in planning and/or programming spatio-temporal parameters of movement sequences may produce a ‘‘cascade” effect that interferes with the development of phonological and lexical skills. As shown by classical studies on speech and language acquisition in typically developing and in late talking children (Bates, Thal, & Janowsky, 2002; Locke, 1994) speech normally develops in interaction with other language and cognitive functions. With regard to CAS, in a recent study on cognitive functions in childhood apraxia of speech, Nijland, Terband, and Maassen (2015) reported that children with CAS showed multiple complex sensorimotor and sequential memory deficits that significantly correlated with the severity of the speech impairment. At present, it is not clear whether the co-occurrence of speech- motor, language and cognitive difficulties can be interpreted as the effect of an unitary multilevel disorder, or as the association of distinct but co-morbid conditions. In this context, the critical review by Pennington and Bishop (2009) on comorbidity between Speech Sound Disorders, language impairment and reading disabilities may have interesting diagnostic, neuropsychological and etiological implications for a better understanding of the pathogenic mechanisms underlying CAS. The multiple overlapping risk factors model proposed by the Authors postulates that each of the above developmental disorders may arise as the consequence of a specific constellation of deficits, some of which act as a single (or separate) underlying cause, while others would play a more general (or shared) pathogenic role. According to this model, the observed clinical phenotype appears to depend on the interaction among genetic, cognitive and environmental risk factors. This topic is critical for genetic research, as suggested by a recent paper (Centanni et al. 2015) in which variants of CNTNAP2, a candidate gene for dyslexia, SLI and autism (Laffin et al., 2012), have been reported in two children affected by CAS without comorbidities. As regards the etiology of CAS in our patients, the results of the present study support the hypothesis of a genetic basis, although a specific gene alteration with a clear pathogenetic role was not identified. As expected in accordance with the diagnosis of idiopathic CAS, clinical structural MRI did not reveal any major neuroanatomical abnormality, suggesting that more sensitive techniques of grey and white matter analysis are necessary to determine the neural basis of CAS. Therefore, the neurobiological correlates of CAS still remain a challenging issue and further studies on larger, well-phenotyped samples, possibly using more sophisticated methods of genetic (Whole Genome Sequencing) and neuroradiological (Voxel Based Morphometry and DTI) analysis will be required to better understand the nature of the disorder. One of the clinical implications of this study concerns the diagnostic process, which requires both an accurate parental interview and a comprehensive clinical assessment. In particular, parental report of early disrupted vocal behaviour and severe delay of speech/language milestones, could alert clinicians to look for other diagnostic features specific to CAS, thus aiding early diagnosis and intervention. When feasible, MRI would be helpful for both diagnostic purposes and for selection of experimental samples, being the most reliable tool to exclude brain structural anomalies and, thus, to differentiate idiopathic from secondary CAS. However, given the prevalence still uncertain of brain abnormalities in CAS, and the

necessity of sedation to perform a brain MRI in most of these children, to date, an absolute recommendation for MRI cannot be supported. For what concerns genetic investigation, a growing number of different anomalies have been found in CAS, confirming the importance of molecular analysis (such as Array-CGH) to evaluate their pathogenic role and the specific relationship between genotype and phenotype. Therefore, the costs and benefits of MRI and genetic testing in the clinical management of childhood apraxia of speech should be established in larger samples, before these procedures can be recommended for assessment of all children with CAS. Finally, the finding that CAS children may show difficulties in grammar acquisition, points to the importance of addressing this aspect in the rehabilitative programs. 4. Methods 4.1. Participants The sample consisted of 32 children affected by idiopathic CAS, selected from a large population of subjects referred for speech and language disorders to the Neurolinguistic Laboratory at the IRCCS Stella Maris during the period January 2010–October 2013. Exclusionary criteria were the presence of orofacial structural abnormalities, known pathologies of neurological, neurometabolical and genetic etiologies, audiological deficits and epilepsy. CAS diagnosis was conducted by a multidisciplinary team in accordance with the three ASHA criteria (2007) and in the presence of at least 4 of 10 Strand’s features of CAS (Murray et al., 2015; Shriberg et al. 2011). The presence of specific diagnostic features was based on perceptual analysis performed by three independent observers (AC, IL, PC) on each child’s speech samples which were video recorded during spontaneous interaction and on execution of formal tests. Agreement reliability for diagnosing a child with apraxia of speech was 98%. For each patient we report in Table 1 of the Supplementary Material the list of features used for assigning the diagnosis of CAS. The final group included 26 males and 6 females (ratio 1:4.3), with a mean age at the time of assessment of 6.8 years (range 4.4–8.6; SD = 3.6). All children had been receiving speech and language therapy for a mean duration of 2.8 years (SD 0.9; range 2–4), but none underwent specific motor treatment for CAS (such as PROMPT, Dale & Hayden, 2013). Parental consent and child assent were obtained in all cases. The study was approved by the Ethics Committee of the IRCCS Fondazione Stella Maris (Number 13/2013). 4.2. Procedures 4.2.1. Biological investigation This included neurological assessment with evaluation of motor development, biochemical testing for serum and urinary amino acids and creatine metabolites, urinary mucopolysaccarids and organic acids; Array-Comparative Genomic Hybridization at 100 KB (platform Agilent 60 k with 100 kb resolution). Brain MRIs were performed on all patients using a 1.5 Tesla MR scanner (GE, Signa Horizon 1.5, Milwaukee, WI, USA). Several sequences were acquired as part of the clinical MRI protocol, including at least a high-resolution 3D T1 BRAVO sequence, 3D FLAIR sequence and Gradient Echo. The study was completed with a brain spectroscopy on the volume of white and/or grey matter.

Table 2 Description of behavioral assessment procedures. Procedure

Parental report

Phonetic inventory Inaccuracy

DDK rate (maximum performance task)

Inconsistency of errors Prosody

Expressive grammar

Family history, child’s pre-, peri- and post-natal clinically significant events, early vocal behavior and language milestones acquisition. Intelligibility of speech to familiar and unfamiliar adults (modified version of the questionnaire by Chilosi et al., 2009) Repetition of 21 syllables containing all the Italian consonantal sounds For words: standardized picture naming test (Bello, Caselli, Pettenati, & Stefanini, 2000); for non- words: repetition of six three-syllable non-words (/tapaka/, /pataka/, /takapa/, /kapata/, /pakata/ and /katapa/). Scoring based on percentage of erroneous productions in both tasks Same tasks as for inaccuracy. Scoring based on the percentage of words and non-words in which the correct number of syllables was not maintained (regardless of phonetic correctness) Fast repetition for 20 s of the trisyllabic non-word sequence / pataka/, scored as: (a) number of repeated pataka obtained within the time given and (b) number of single syllables (/pa/ , /ta/, /ka/) repeated per second Same tasks as for inaccuracy. Scoring based on the percentage of variable phonetic errors in three repeated productions of the same word or non-word stimuli Evaluation of fluency, of lexical stress assignment and of variations in speech rate by perceptual analysis of spontaneous speech samples Grid of Analysis of Spontaneous Speech (GASS) developed by Chilosi et al. (2013)

Number and mean age of subjects

Scoring

30 ss, mean age = 4.3 yrs (SD 0.47)

Mean number of phonemes: 19.2 (SD 0.9)

30 ss, mean age = 4.3 yrs (SD 0.47)

Word inaccuracy: mean 8.8% (SD 10.7) Non-word inaccuracy: mean 14% (SD 19)

30 ss, mean age = 4.3 yrs (SD 0.47)

Syllable omissions: mean 1% for both words and non-words

60 ss, mean age = 6.2 yrs (SD 2) Three age sub-groups: 1st group (32 ss): mean age = 4.3 yrs (SD 0.5) 2nd group (14 ss): mean age = 6.7 yrs (SD 0.6) 3rd group (14 ss): mean age = 8.5 yrs (SD 0.8) 30 ss mean age = 4.3 (SD 0.47)

Whole group: mean rep 25 (SD 4.7); syll/s 3.7 (SD 0.7) 1st group: mean rep 22.8 (SD 4.8); syll/s 3.4 (SD 0.7) 2nd group: mean rep 25.4 (SD 2.7); syll/s 3.8 (SD 0.4) 3rd group mean rep 27.8 (SD 5.9); syll/s 4.1 (SD 0.6)

Longitudinal sample: 6 children video recorded twice a month from 19 to 36 mths (Cipriani, Chilosi, Bottari, & Pfanner, 1993)

TFL: 240 Italian children from 2.6 to 6 yrs PPVT: 2400 Italian children from 3.9 to 11.6 yrs

19–25 mths: Presyntactic level, MLU = 1.4, emergence of twoand three-word combinations 20–26 mths: Telegraphic level, MLU = 2.2, emergence of morphosyntatctically incomplete subject –verb-object structures 24–31 mths: Grammatical stage 1, MLU = 2.6, full control of morphology in simple sentences 28–36 mths: Grammatical stage 2, MLU = 3.1, production of well-formed both simple and complex sentences Percentile scores Standard scores

TFL: 268 Italian children from 3 to 6 yrs One – Word Picture Vocabulary test: 154 children from 4.6 to 10.8 yrs

Percentile scores Separate z-scores for high (52 items) and low (52 items) frequency words

Cross-sectional sample: 50 children aged 26–44 mths (Chilosi et al., 2013)

Receptive vocabulary

Expressive vocabulary

Phono Lexical Test-TFL (Vicari, Marotta, & Luci, 2007) and/or Peabody Picture Vocabulary Test (PPVT-R –Dunn & Dunn, 1997; Italian standardization – Stella et al., 2000), depending on the child’s age and on the severity of the disorder Phono Lexical Test-TFL (Vicari et al., 2007) and/or One-Word Picture Vocabulary Test (Brizzolara, 1989), depending on the child’s age and on the severity of the disorder

Words: mean 0.4% (SD 1.3) Non-words: mean 3.2% (SD 5.9)

A.M. Chilosi et al. / Brain & Language 150 (2015) 177–185

Syllable omissions

Normal controls

Mean rep = mean number of repeated /pataka/ obtained in 20 s. Syll/sec = mean number of syllables repeated in one second. ss = subjects, mths = months, yrs = years.

183

184

A.M. Chilosi et al. / Brain & Language 150 (2015) 177–185

Structural MRIs and spectroscopies were acquired primarily to exclude white and grey matter abnormalities, with particular attention to brain regions involved in speech planning and execution (supplementary motor area, insula, frontal cortex, thalamus, caudate nucleus and cerebellum). MRI images were double checked by a child neuroradiologist (RP) and a child neurologist expert in MRIs for structural abnormalities (SF). 4.2.2. Case history and behavioral assessment To identify the specific characteristics of CAS including early signs of the disorder, the assessment protocol included: (a) Parental report on child and family history. Family history was considered positive if one or more of the participant’s nuclear family members had a history of any type of speech-language disorder and/or dyslexia. Intelligibility was evaluated on the basis of information provided by parents and was rated according to percentages of intelligible speech to both familiar and unfamiliar adults as follows: less than 25% of intelligible speech; between 25% and 50%; between 50% and 75%; more than 75%. (b) Evaluation of non-verbal intelligence (WPPSI-III and WISCIII or IV). (c) Assessment of isolated and sequenced volitional verbal and non-verbal oral movements (Bearzotti & Fabbro, 2003). (d) Speech tasks for assessment of phonetic inventory, inaccuracy and inconsistency of speech and syllable omissions. Since these measures were not standard scores from normreferenced tools, a small group of 30 typically developing children with a mean age of 4.3 yrs (SD 0.47) were used as normal controls. (e) Diadochokinesis (DDK) was assessed by a maximum performance task (see Table 2 for details). As this task lacks normreferenced scores, data were collected from a control sample of 60 typically developing children subdivided into three age sub-groups (4–5; 6–7.5; >7.6 yrs). (f) Perceptual analysis of prosody of spontaneous speech by three independent raters (AC, IL, PC). (g) Analysis of the level of grammar organization on speech samples collected during spontaneous verbal interaction or presentation of a pictured story. All assessment sessions were videotaped and coded by three independent observers (AC, IL, PC). (h) Administration of standardized tests of receptive and expressive vocabulary. For details on speech and language tasks and on control groups, see Table 2. 4.2.3. Data analysis Quantitative data were expressed as means ± SD, while categorical data were expressed as frequencies and percentages. Correlation analyses by Pearson Correlation were conducted on four subsets of variables: data from the medical records, isolated and sequenced volitional verbal and non-verbal oral movements, speech and language measures. Statistical significance of correlations was adjusted for multiple comparisons using the Benjamini and Hochberg’s approach. After a descriptive analysis of the data, a Principal Component Analysis (PCA) based on the correlation matrix was performed. PCA is a descriptive exploratory technique designed to analyze the pattern of relationships among the variables included in the analysis. To state the goal of a typical analysis, variables are represented in terms of the distances between them in a lowdimensional space. The display of the variables in the final coordinate system provides an indication of the nature of the

relationships between them. Rotated PCA was performed by varimax with normalization of Kaiser. One-way Anovas were carried out for comparison of CAS children and normal controls on speech measures and DDK scores. Bonferroni post hoc analysis was also performed for comparison between CAS children and the 3 subgroups of normal controls on DDK scores. A two-sided p value <0.05 was considered significant. Analyses were performed by SPSS version 21. Acknowledgments This research was supported by the Italian Ministry of Health (RRC-2013-2353257) and 5x1000 grant. We would like to thank the speech pathologists Barbara Cerri, Lorena Cittadoni, Lucia Pfanner and Renata Salvadorini for their help in collecting language data and psychologists Claudia Casalini and Alessandro Comparini for cognitive assessment of CAS children. Special thanks to Irina Podda, speech-pathologist and Italian instructor of PROMPT Institute, for suggestions and assistance and to Federica Ciaponi, psychologist for collecting data from normal controls. Finally, we would like to thank Laura Steven and Vincent Corsentino for reviewing the English of the manuscript. Appendix A. Supplementary material Supplementary data associated with this article can be found, in the online version, at http://dx.doi.org/10.1016/j.bandl.2015.10. 002. References American Speech-Language-Hearing Association (2007). Technical report on childhood apraxia of speech. Last accessed: 19th August 2015. Bates, E., Thal, D., & Janowsky, J. (2002). Early language development and its neural correlates. In I. Rapin & S. Segalowitz (Eds.), Handbook of neuropsychology (2nd ed.. Elsevier ed.. Vol. 8, Part II. Battini, R., Chilosi, A., Mei, D., Casarano, M., Alessandri, M. G., Leuzzi, V., ... Cioni, G. (2007). Mental retardation and verbal dyspraxia in a new patient with de novo creatine transporter (SLC6A8) mutation. American Journal of Medical Genetics, 143A(15), 1771–1774. Bearzotti, F., & Fabbro, F. (2003). Il test per la valutazione delle prassie orofacciali nel bambino. Giornale di Neuropsichatria dell’Età Evolutiva, 23, 406–417. Bello, A., Caselli, M. C., Pettenati, P., & Stefanini, S. (2010). PinG parole in gioco. Giunti O.S: Firene. Brizzolara, D. (1989). Test di vocabolario figurato, Technical Report of the Research Project 500.4/62.1/1134. Centanni, T. M., Sanmann, J. N., Green, J. R., Iuzzini-Seigel, J., Bartlett, C., Sanger, W. C., et al. (2015). The role of candidate-gene CNTNAP2 in childhood apraxia of speech and specific language impairment. American Journal of Medical Genetics. http://dx.doi.org/10.1002/ajmg.b.32325. Chilosi, A. M., Brizzolara, D., Lami, L., Pizzoli, C., Gasperini, F., Pecini, C., ... Zoccolotti, P. (2009). Reading and spelling disabilities in children with and without a history of early language delay: A neuropsychological and linguistic study. Child Neuropsychology, 15(6), 582–604. Chilosi, A. M., Cipriani, P., Graziosi, V., & Fiori, S. (2014). Clinical characteristics of CAS. Developmental Medicine & Child Neurology, 56(s4), 9. Chilosi, A. M., Cipriani, P., Pecini, C., Brizzolara, D., Biagi, L., Montanaro, D., ... Cioni, G. (2008). Acquired focal brain lesions in childhood: Effects on development and reorganization of language. Brain and Language, 106(3), 211–225. Chilosi, A. M., Comparini, A., Scusa, M. F., Orazini, L., Forli, F., Cipriani, P., et al. (2013). A longitudinal study of lexical and grammar development in deaf Italian children provided with early cochlear implantation. Ear and Hearing, 34(3), 28–37. Cipriani, P., Chilosi, A. M., Bottari, P., & Pfanner, L. (1993). L’acquisizione della morfosintassi in italiano: Fasi e processi. Padova: Unipress. Dale, P. S., & Hayden, D. A. (2013). Treating speech subsystems in childhood apraxia of speech with tactual input: The PROMPT approach. American Journal of SpeechLanguage Pathology, 22(4), 644–661. Dunn, L. M., & Dunn, L. M. (1997). Peabody picture vocabulary test-PPVT (3rd ed.). Circle Pines: American Guidance Service Inc (Italian adaptation by Stella G., Pizzoli C., & Tressoldi P. (2000). Torino: Omega Edition.). Fedorenko, E., Morgan, A., Murray, E., Cardinaux, A., Mei, C., Tager-Flusberg, H., ... Kanwisher, N. (2015). A highly penetrant form of childhood apraxia of speech

A.M. Chilosi et al. / Brain & Language 150 (2015) 177–185 due to deletion of 16p11.2. European Journal of Human Genetics. http://dx.doi. org/10.1038/ejhg.2015.149. Jul 15, Epub ahead of print. Fiori, S., Chilosi, A. M., Jhimli, M., Pannek, K., Pasquariello, R., Cipriani, P., ... Guzzetta, A. (submitted for publication). Neuroanatomical correlates of childhood apraxia of speech: A connectomic approach. Hurst, J. A., Baraitser, M., Auger, E., Graham, F., & Norell, S. (1990). An extended family with a dominantly inherited speech disorder. Developmental Medicine and Child Neurology, 32(4), 352–355. Laffin, J. J., Raca, G., Jackson, C. A., Strand, E. A., Jakielski, K. J., & Shriberg, L. D. (2012). Novel candidate genes and regions for childhood apraxia of speech identified by array comparative genomic hybridization. Genetics in Medicine, 14(11), 928–936. Lai, C. S. L., Fisher, S. E., Hurst, J. A., Vargha-Khadem, F., & Monaco, A. P. (2001). A forkhead-domain gene is mutated in a severe speech and language disorder. Nature, 413(6855), 519–523. Liégeois, F. J., & Morgan, A. T. (2012). Neural bases of childhood speech disorders: Lateralization and plasticity for speech functions during development. Neuroscience and Biobehavioral Reviews, 36(1), 439–458. Liégeois, F. J., Morgan, A. T., Connelly, A., & Vargha-Khadem, F. (2011). Endophenotypes of FOXP2: Dysfunction within the human articulatory network. European Journal of Paediatric Neurology, 15, 283–288. Locke, J. L. (1994). Gradual emergence of developmental language disorders. Journal of Speech and Hearing Research, 37(3), 608–616. Maassen, B., Nijland, L., & Terband, H. (2010). Developmental models of childhood apraxia of speech. In B. Maassen & P. van Lieshout (Eds.), Speech motor control (pp. 243–258). Oxford University Press. MacDermot, K. D., Bonora, E., Sykes, N., Coupe, A. M., Lai, C. S., Vernes, S. C., ... Fisher, S. E. (2005). Identification of FOXP2 truncation as a novel cause of developmental speech and language deficits. American Journal of Human Genetics, 76, 1074–1080. Murray, E., McCabe, P., Heard, R., & Ballard, K. J. (2015). Differential diagnosis of children with suspected childhood apraxia of speech. Journal of Speech, Language, and Hearing Research, 58(1), 43–60. Nijland, L., Terband, H., & Maassen, B. (2015). Cognitive functions in childhood apraxia of speech. Journal of Speech, Language, and Hearing Research, 58(3), 550–565. Ogar, J., Willock, S., Baldo, J., Wilkins, D., Ludy, C., & Dronkers, L. (2006). Clinical and anatomical correlates of apraxia of speech. Brain and Language, 97(3), 343–350. Ozanne, A. E. (2005). Childhood apraxia of speech. In B. Dodd (Ed.), Differential diagnosis and treatment of children with speech disorders (2nd ed., pp. 71–82). London: Whurr Pub. Ltd.. Pennington, B. F., & Bishop, D. V. M. (2009). Relations among speech, language, and reading disorders. Annual Review of Psychology, 60, 283–306. Peter, B., Matsushita, M., Oda, K., & Raskind, W. (2014). De novo microdeletion of BCL11A is associated with severe speech sound disorder. American Journal of Medical Genetics, 164A(8), 2091–2096. Raca, G., Baas, B. S., Kirmani, S., Laffin, J. J., Jackson, C. A., Strand, E. A., ... Shriberg, L. D. (2013). Childhood Apraxia of Speech (CAS) in two patients with 16p11.2 microdeletion syndrome. European Journal of Human Genetics, 21(4), 455–459.

185

Rvachew, S., & Grawburg, M. (2006). Correlates of phonological awareness in preschoolers with speech sound disorders. Journal of Speech, Language, and Hearing Research, 49(1), 74–87. Shriberg, L. D., Jakielski, K. J., & El-Shanti, H. (2008). Breakpoint localization using array-CGH in three siblings with an unbalanced 4q;16q translocation and childhood apraxia of speech (CAS). American Journal of Medical Genetics, 146a, 2227–2233. Shriberg, L. D., Lohmeier, H. L., Strand, E. A., & Jakielski, K. J. (2012). Encoding, memory, and transcoding deficits in Childhood Apraxia of Speech. Clinical Linguistics & Phonetics, 26(5), 445–482. Shriberg, L. D., Potter, N. L., & Strand, E. A. (2011). Prevalence and phenotype of childhood apraxia of speech in youth with galactosemia. Journal of Speech, Language and Hearing Research, 54(2), 487–519. Terband, H., & Maassen, B. (2010). Speech motor development in childhood apraxia of speech: Generating testable hypotheses by neurocomputational modeling. Folia Phoniatrica et logopaedica: Official Organ of the International Association of Logopedics and Phoniatrics, 63(3), 134–142. 277. Thevenon, J., Callier, P., Andrieux, J., Delobel, B., David, A., Sukno, S., ... Faivre, L. (2013). 12p13.33 microdeletion including ELKS/ERC1, a new locus associated with childhood apraxia of speech. European Journal of Human Genetics, 21(1), 82–88. Thoonen, G., Maassen, B., Wit, J., Gabreels, F., & Schreuder, R. (1996). The integrated use of maximum performance tasks in differential diagnostic evaluations among children with motor speech disorders. Clinical Linguistics and Phonetics, 10, 311–336. Vargha-Khadem, F., Gadian, D. G., Copp, A., & Mishkin, M. (2005). FOXP2 and the neuroanatomy of speech and language. Nature Reviews – Neuroscience, 6(2), 131–138. Velleman, S. L. (2011). Lexical and phonological development in children with childhood apraxia of speech. A commentary on Stoel-Gammon’s ‘Relationships between lexical and phonological development in young children’. Journal of Child Language, 38(1), 82–86. Vicari, S., Marotta, L., & Luci, A. (2007). Test TFL - Test fono lessicale valutazione delle abilità lessicali in età prescolare. Trento: Erickson Edition. William, P., & Stackhouse, G. (2000). Rate, accuracy and consistency: Diadochokinetic performance of young, normally developing children. Clinical Linguistics and Phonetics, 14(4), 267–293. Wechsler D. WISC-III (1992). Wechsler intelligent scale for children-3rd edition. San Antonio, TX: Psychological Corporation. Italian adaptation by Orsini, A. & Picone, L. (2006). Firenze: Giunti O.S. Wechsler D. WISC-IV (2003). Wechsler intelligence scale for children-4th edition. San Antonio: Psychological Corporation. Italian adaptation by Orsini, A., Pizzuti, L. & Picone, L. (2012). Firenze: Giunti O.S. Worthey, E. A., Raca, G., Laffin, J. J., Wilk, B. M., Harris, J. M., Jakielski, K. J., ... Shriberg, L. D. (2013). Whole-exome sequencing supports genetic heterogeneity in childhood apraxia of speech. Journal of Neurodevelopmental Disorders, 5, 29. Wechsler D. WPPSI-III (2002). Wechsler preschool and primary scale of intelligence3rd edition. San Antonio: Psychological Corporation. Italian adaptation by Fancello, G. S. & Cianchetti, C. (2008). Firenze: Giunti O.S.