Vocalisation Repertoire at the End of the First Year of Life: An Exploratory Comparison of Rett Syndrome and Typical Development

Vol.:(0123456789) Journal of Developmental and Physical Disabilities https://doi.org/10.1007/s10882-022-09837-w 1 3 ORIGINAL ARTICLE Vocalisation Repertoire attheEnd oftheFirst Year ofLife: AnExploratory Comparison ofRett Syndrome andTypical Development KatrinD.Bartl‑Pokorny1,2 · FlorianB.Pokorny1,2 · DuniaGarrido3 · BjörnW.Schuller2,4 · DajieZhang1,5,6 · PeterB.Marschik1,5,6,7 Accepted: 23 January 2022 © The Author(s) 2022 Abstract Rett syndrome (RTT) is a rare, late detected developmental disorder associated with severe deficits in the speech-language domain. Despite a few reports about atypicalities in the speech-language development of infants and toddlers with RTT, a detailed analysis of the pre-linguistic vocalisation repertoire of infants with RTT is yet missing. Based on home video recordings, we analysed the vocalisations between 9 and 11months of age of three female infants with typical RTT and compared them to three age-matched typically developing (TD) female controls. The video material of the infants had a total duration of 424min with 1655 infant vocalisations. For each month, we (1) calculated the infants’ canonical babbling ratios with CBRUTTER, i.e., the ratio of number of utterances containing canonical syllables to total number of utterances, and (2) classified their pre-linguistic vocalisations in three non-canonical and four canonical vocalisation subtypes. All infants achieved the milestone of canonical babbling at 9months of age according to their canonical babbling ratios, i.e. CBRUTTER ≥ 0.15. We revealed overall lower CBRsUTTERand a lower proportion of canonical pre-linguistic vocalisations consisting of well-formed sounds that could serve as parts of target-language words for the RTT group compared to the TD group. Further studies with more data from individuals with RTT are needed to study the atypicalities in the pre-linguistic vocalisation repertoire which may portend the later deficits in spoken language that are characteristic features of RTT. Keywords Canonical babbling· Early vocalisations· Infant· Late detected developmental disorders· Rett syndrome· Speech-language impairment Dajie Zhang and Peter B. Marschikshare senior authorship * Florian B. Pokorny [email protected] Extended author information available on the last page of the article Journal of Developmental and Physical Disabilities 1 3 Introduction Rett syndrome (RTT; OMIM 312,750) is a severe developmental disorder mostly caused by de novo mutations in the MECP2 (methyl-CpG binding protein 2) gene on the long arm of the X chromosome (Xq28) (Amir etal., 1999; Zoghbi, 2005). RTT has a prevalence of approximately 1 of 10,000 live female births (Hagberg, 1985; Laurvick etal., 2006); male individuals with RTT are very rare (Christen & Hanefeld, 1995). Clinical diagnosis of RTT is based on four core consensus criteria (Neul etal., 2010): 1. partial or complete regression (i.e., loss) of already acquired purposeful hand skills, 2. regression of already acquired spoken language, 3. gait abnormalities, and 4. stereotypic hand movements. Most individuals with RTT have an onset of regression between 12 and 18months of age (Burford etal., 2003; Einspieler & Marschik, 2019; Lee etal., 2013). The classic form of RTT is currently diagnosed at a mean age of 2.7years (Tarquinio etal., 2015). Like for other late detected developmental disorders, the late diagnosis of RTT hinders the implementation of early, individually tailored interventions for affected children. This motivates research on early development to promote earlier identification of affected individuals. As deficits in the speech-language domain compose a core characteristic of RTT, a thorough investigation of the pre-linguistic speech-language development may reveal early signs that portend later associated impairments. Pre-linguistic vocalisations are typically produced throughout the first year of life, preceding the first referential words. Based on vocal data of typically developing (TD) children, specific vocalisation schemes have been developed with the aim to phonetically categorise these pre-linguistic vocalisations (Nathani etal., 2006; Papousek, 1994). The schemes define a number of vocalisation patterns and related studies investigated the onset and the proportional use of these patterns in the vocalisation repertoires of infants (Nathani etal., 2006; Oller, 1980; Papousek, 1994; Stark, 1980, 1981). One of the most salient pre-linguistic vocalisation patterns is canonical babbling, typically emerging between 5 and 10months of age (Lang etal., 2019; Morgan & Wren, 2018; Oller, 1980, 2000). Infants produce canonical syllables by combining consonant(-like) sounds and vowel(-like) sounds with fast formant transitions between consonant and vowel (Oller, 2000; Oller etal., 1999; Papousek, 1994). These canonical syllables often occur in series (e.g., /babu/, /mamama/, /dadama/) (Nathani et al., 2006). The ability of an infant to produce canonical syllables and practice them in various combinations is crucial for the production of referential words as these usually consist of canonical syllables (Lee etal., 2018; Morgan & Wren, 2018; Oller etal., 1998). Studies varied in their definitions of canonical babbling and in their parameters chosen to define the onset of canonical babbling (Lang etal., 2019; Molemans etal., 2012; Roche etal., 2018). Some studies defined canonical babbling to be acquired with the first occurrence of a canonical syllable or two combined canonical syllables (Bartl-Pokorny etal., 2013; Marschik etal., 2013; Schramm etal., 2009), whereas others defined a certain canonical babbling threshold (Lang etal., 1 3 Journal of Developmental and Physical Disabilities 2020; Molemans etal., 2012; Oller & Eilers, 1988; Oller etal., 1994; Patten etal., 2014). A widely used measure to determine the onset of canonical babbling as well as the proportion of this verbal pattern in an infant’s vocalisation repertoire at a given age is the canonical babbling ratio (CBR) (Oller & Eilers, 1988). It was originally defined as the ratio of canonical syllables to total number of utterances (CBRutt) (Oller & Eilers, 1988). In the last 30years, a number of different CBR measures were proposed; for an overview see Molemans and colleagues (2012). These differ mainly in the exact definition of a canonical syllable and in whether the number of canonical syllables is divided by the total number of syllables or by the total number of utterances (Lang etal., 2019). The only CBR measure that can be applied without counting the number of canonical syllables is CBRUTTER, the most recent CBR measure (Nyman & Lohmander, 2018). CBRUTTER is defined as the ratio of number of utterances containing canonical syllables to total number of utterances. Nyman and Lohmander (2018) and Lang and colleagues (2020) found high correlations between CBRUTTER and other CBR measures, yet CBRUTTER is less time-consuming to obtain. Thus, we decided to use CBRUTTER for the current study. A child is regarded as having reached the canonical babbling milestone if his or her CBR is found to be higher than a defined threshold. For CBRUTTER, the threshold is set at 0.15 (Nyman & Lohmander, 2018). Irrespective of different criteria used to define the onset of canonical babbling, not meeting this important milestone at 10months of age has been discussed as early indicator of atypical speech-language development (Lang etal., 2019; Lohmander etal., 2017; Oller etal., 1998, 1999; Yankowitz etal., 2019). For example, infants with profound hearing impairment (Löfkvist etal., 2020; Schauwers etal., 2004), Down syndrome (Lynch etal., 1995), autism spectrum disorder (Patten etal., 2014), or fragile X syndrome (Belardi etal., 2017) were found to have lower CBRs in comparison to TD infants and/or to fail to reach the defined CBR threshold for the onset of canonical babbling in time. The reported deviances in CBR of infants with developmental disorders lead to the assumption that deviances in CBR could also be present in infants with other late detected developmental disorders that are associated with deficits in the speechlanguage domain, such as RTT. To the best of our knowledge, the CBR has not been investigated so far for infants with RTT. Still, studies analysing home videos of infants with RTT indicated early speech-language peculiarities already in the preregression phase, including deviant canonical babbling (Bartl-Pokorny etal., 2013; Einspieler & Marschik, 2019; Einspieler etal., 2014; Marschik etal., 2011, 2013, 2014a, b; Pokorny etal., 2018; Townend etal., 2015). For example, Marschik and colleagues (2013) observed canonical babbling (as defined by at least one occurrence of two successive consonant–vowel combinations; e.g., /baba/) in only 5 out of the 10 children with RTT during the first two years of life. Bartl-Pokorny and colleagues (2013) reported that none of the sampled six 9to 12-month old infants with RTT was observed to use canonical babbling for communicative purposes such as directing attention of self or requesting an object. These findings, together with reports of individuals with other developmental disorders not achieving the canonical babbling milestone in time, led us to explore pre-linguistic vocalisations of individuals with RTT in more detail at an age (i.e., 9 to 11months) when TD infants Journal of Developmental and Physical Disabilities 1 3 are expected to achieve the milestone of canonical babbling. With the present study, we aimed to provide for the first time a meticulous comparison of the vocalisation repertoires of 9to 11-month old infants with RTT and TD infants. For this, we (i) provided the CBRs of the infants and (ii) classified the infants’ non-canonical and canonical pre-linguistic vocalisations in subtypes. We hypothesised that infants with RTT and TD infants differ in their CBRs and that infants with RTT and TD infants differ in the composition of their non-canonical and canonical pre-linguistic vocalisation repertoires. The aim of this exploratory study was to provide a starting point towards a better understanding of the pre-linguistic vocalisations of individuals with RTT. Materials andMethods We used a retrospective video analysis approach to compare the vocalisations of infants later diagnosed with RTT and of TD infants. The study was approved by the local research ethics committee. Participants For the present study, we included data of infants with RTT and of TD infants from our database. Inclusion criteria for the participants with RTT were: (a) a confirmed clinical diagnosis of typical RTT, (b) infant was brought up in a monolingual German-speaking family, (c) audio-video material was available between 9 and 11months (see also "Material"). Following these inclusion criteria, the available participants were three females with RTT (RTT1–RTT3). Genetic testing revealed the following pathogenic MECP2 mutations: p.R168X for RTT1, p.F157L for RTT2, and p.R106W for RTT3. The infants with RTT were matched with three TD infants (TD1–TD3) for gender, age at time of recording of video material, and family language. All infants were singletons and were born at term. Material Analysis was based on home video recordings that were taken by the infants’ parents during daily routines or special family events. At the time of recording, the parents of the participants with RTT were not aware of their children’s medical condition. The material was either provided by the participants themselves (TD group) or their parents (RTT group), who gave their informed consent for analysis of the data for research purposes and for publication of the results. In the present study, we included all available home video material taken from the participants when they were 9 to 11months old, to focus on the pre-linguistic period in which TD infants are expected to have achieved the milestone of canonical babbling (Lang et al., 2019; Lohmander etal., 2017; Oller etal., 1998, 1999; Yankowitz etal., 2019). The total duration of the included home video material was 424min (RTT1: 191min, RTT2: 81min, RTT3: 14min, TD1: 61min, TD2: 44min, and TD3: 33min). The 1 3 Journal of Developmental and Physical Disabilities videos of the TD group were recorded in the years 1991/1992 (TD1), 1988 (TD2), and 1998 (TD3). The videos of the RTT group were recorded between the years 1994 and 2003. A trained research assistant blind to the purpose of the project prepared the video material for analysis with the video coding system Noldus Observer XT (https:// www. noldus. com): First, the videos were annotated for scenes showing the infants in settings ‘with social interaction’ vs ‘without social interaction’. Second, the videos were marked for infant vocalisations by setting ‘start’ and ‘stop’ tags. A breathgroup criterion was used to segment vocalisations, i.e., segment boundaries were set in case of ingressive breathing (Nathani & Oller, 2001). Inspiratory sounds were defined as regular parts of a vocalisation and did not mark segment boundaries. Vegetative sounds (e.g., breathing sounds, sneezes, hiccups) were not segmented and were excluded from further analysis. Third, each vocalisation was exported as a separate audio clip, which was labelled with a randomly assigned numeric code. The audio clips prepared for subsequent coding did not include information on participant ID, age, and developmental outcome. As several studies have suggested that volubility and vocalisation patterns such as canonical babbling may be sensitive to social circumstances, e.g., interaction vs no interaction with caregiver (Goldstein & Schwade, 2008; Iyer etal., 2016; Lee etal., 2018), and for better comparability regarding recording situation, only those vocalisations produced in the ‘with social interaction’ settings (i.e., 96% of the entire vocalisations) were selected for further analyses. All segmented vocalisations were double-checked and verified for segment boundaries by the second author. The final dataset for analysis consisted of 1655 vocalisations (RTT1: 735, RTT2: 166, RTT3: 121, TD1: 197, TD2: 240, and TD3: 196). Vocalisation duration ranged from 0.28 to 12.26s (Mean = 1.64, SD = 1.12). The shortest vocalisation, uttered by TD3 at 10months of age, was a single vowellike sound. The longest vocalisation, uttered by TD2 at 11months of age, was a combination of several consonant-like and vowel-like sounds interspersed with short pauses without ingressive breathing. For RTT1 we had vocalisation data available for each of the three months of interest; for RTT2 data were available for 9 and 10months only, for RTT 3 data were available for 9months only, and for TD1, TD2, and TD3 we had vocalisation data available for each of the three months. Vocalisation Classification Vocalisations that were produced in a neutral mood, referred to as pre-linguistic vocalisations (PLVs) were classified according to the vocalisation scheme presented in Table1. The scheme was similar to the ‘Stark Assessment of Early Vocal Development-Revised’ (SAEVD-R) (Nathani et al., 2006) with alterations for our study purposes. We included vocalisation types that are characteristic for the investigated age. In particular, the following alterations compared to SAEVD-R were carried out to minutely represent the infants’ age-specific vocalisation repertoires and their stratified complexity: 1. We classified noncanonical vocalisations into three subtypes, i.e. PLV1–PLV3, instead of annotating all Level 1 to Level 3 vocalisation types defined by the SAEVD-R which are Journal of Developmental and Physical Disabilities 1 3 targeted on capturing the vocalisation repertoire of younger infants compared to our target group; 2. For the canonical realisations with 2 sounds, we defined a vocalisation subtype for the combinations of 1 vowel(-like) and 1 consonant(- like) sound with a rapid formant transition between the sounds, and with at least 1 of these sounds not conforming to the target language (i.e. PLV4), as a contrast to PLV5, which consists of both sounds conforming to the target language. Vowel-like and consonant-like sounds are not yet well-formed vowels and consonants that could serve as parts of target-language words (Oller, 2000) and therefore cannot be accurately transcribed with the International Phonetic Alphabet (IPA), a widely used system to transcribe speech sounds (International Phonetic Association, 2018); 3. For the canonical vocalisations with 3 or more sounds, we added the category PLV6, with at least 1 of the sounds not conforming to the target language, and PLV7, with all the sounds conforming to the target language. Note that in SAEVD-R, ‘VC’ syllables assigned to PLV5 in our scheme and all vocalisations assigned to PLV7 are defined identically as ‘complex syllables’ (CMPX), regardless of the number of sounds included. In our scheme, we separated the vocalisations with two sounds from those with three or more sounds marking different vocal complexities; 4. We excluded the subtype for consonant(-like)-vowel(-like) combinations with prolonged formant transitions (classified as ‘marginal babbling’ in SAEVD-R) as the corpus for the present study did not contain such realisations. Each vocalisation was first assigned to one of the three vocalisation types (i) pleasure (i.e., laughing, pleasure bursts), (ii) distress (i.e., fussing, crying), or (iii) PLV. Each PLV was then assigned to one of seven mutually exclusive vocalisation subtypes (i.e., PLV1–PLV7; see Table1). Vocalisations assigned to PLV1, PLV2, or PLV3 did not contain canonical syllables while vocalisations assigned to PLV4, PLV5, PLV6, or PLV7 contained canonical syllables (Table1). Vocalisation segmentation according to a breath-group criterion (see "Material") may result in two or more segments within one PLV (separated by pauses without ingressive breathing). Consequently, a PLV may include segments of different vocalisation subtypes, e.g., both a single vowel-like sound (i.e., PLV1) and a vowel-consonant combination (i.e., PLV5). If so, the PLV was assigned to the highest vocalisation subtype index included, ascending from PLV1 to PLV7. For example, a PLV including both a PLV1-segment and a PLV5-segment was coded as PLV5. Our approach to assign a PLV to the highest vocalisation subtype index included allowed us to capture all canonical babbling occurrences in the corpus as all canonical vocalisation subtypes (i.e., PLV4–PLV7) have higher subtype indexes than the non-canonical vocalisation subtypes (i.e., PLV1–PLV3). Three coders (first author, second author, third author) independently annotated all 1655 vocalisations according to the scheme (Table1). This annotation procedure resulted in a majority vote (i.e., at least two of the three coders agreed on the classification) for 1533 vocalisations, determining the final vocalisation type and subtype annotation for the respective vocalisations. The remaining 122 vocalisations for which no majority vote was available (7.4% of all vocalisations) were discussed within the team until consensus on the classification was achieved. 1 3 Journal of Developmental and Physical Disabilities Analysis We performed the following three steps to meet our research goals: First, we identified the number of vocalisations assigned to the respective vocalisation types (i.e., pleasure, distress, PLV) and the PLV subtypes (i.e., PLV1-PLV7) for each infant and month of age. Second, we computed the distribution (as percentage) of PLV1-PLV7. Third, we calculated the infants’ CBRs by using CBRUTTER (Nyman & Lohmander, 2018). The CBRUTTER can be easily derived from our vocalisation subtype analysis by dividing the sum of PLVs containing canonical syllables (i.e., PLV4-PLV7) by the total number of PLVs of the respective infant. The threshold for reaching the canonical babbling milestone was CBRUTTER ≥ 0.15, as defined by Nyman and Lohmander (2018). Results The vast majority of all infants’ vocalisations at 9, 10, and 11months were PLVs (see Table2). Figure1 illustrates the distribution (as percentage) of PLV1–PLV7 produced per infant and month. All infants produced both canonical (coloured and ruled bars, Fig.1) and non-canonical (grey shaded bars, Fig.1) PLVs, the latter forming the major component of most infants’ vocalisation repertoires. As indicated by the dashed line in Fig.1, all six infants achieved the canonical babbling milestone by 9months of age (i.e., CBRUTTER ≥ 0.15). Figure 1 presents a generally higher proportion of canonical PLVs in the TD than in the RTT group. The CBRsUTTER Table 1 Classification scheme of infant vocalisations. [c]/[v] = consonant-like/vowel-like sound, cannot be accurately transcribed with the International Phonetic Alphabet (IPA) (International Phonetic Association, 2018); [C]/[V] = consonant/vowel, included in the IPA; PLV = pre-linguistic vocalisation, produced in a neutral mood; ˇ = IPA diacritic to indicate an ascending pitch (rising contour); * = canonical babbling (Oller etal., 2000,1999) Journal of Developmental and Physical Disabilities 1 3 of the infants varied from month to month (for exact CBRUTTER values please refer to bottom of Table 2). Despite the variation, TD1 consistently had the highest CBRUTTER of the six infants from 9 to 11months, followed by TD2. The lowest observed CBRsUTTER of TD1 and TD2 (i.e., 0.42 and 0.35 at 10months) were higher than the highest CBRsUTTER of the remaining four infants (bottom of Table2) and were considerably higher than the threshold of 0.15. The lowest CBRsUTTER in this 10 20 30 40 50 60 70 80 90 100 100 90 80 70 60 50 40 30 20 10 % of PLVs with canonical syllables TD1 98 43 46 TD2 49 80 105 RTT2 111 44 RTT1 322 176 226 TD3 26 81 80 RTT3 120 15% PLV3 PLV2 PLV1 PLV4 PLV5 PLV6 PLV7 NA NA NA % of PLVs without canonical syllables Fig. 1 Distribution (as percentage) of the pre-linguistic vocalisation subtypes (PLV1–PLV7) produced by the participants at 9months of age (left bar), 10months (middle bar), and 11months (right bar). Each bar represents the total PLV repertoire (hundred-percent) analysed for the month. Numbers on top of the bars indicate the number of PLVs available for analysis of the month. The dashed line indicates the 0.15 threshold of reaching the canonical babbling milestone. PLV = pre-linguistic vocalisation; NA = no data available for the respective month Table 2 Number of vocalisations assigned to the respective vocalisation types and subtypes for the participants at 9, 10, and 11months of age. CBRUTTER = canonical babbling ratio; NA = no data available for the respective month; TLR = ratio of canonical PLVs conforming to the target language; * = canonical babbling (Oller etal., 2000, 1999) 1 3 Journal of Developmental and Physical Disabilities sample were observed in the RTT group (i.e., RTT3 at 9months, 0.18; and RTT2 at 10months, 0.11). TD3 and RTT1 demonstrated considerable similarities concerning their CBRsUTTER in all three months (Fig.1). A strong reduction in CBRUTTER was only observed for RTT2 (i.e., from 0.32 at 9months to 0.11 at 10months; Table2), who was also the only infant who ever presented a CBRUTTER lower than 0.15. As shown in Fig.1, the majority of canonical PLVs of all infants did not conform to the target language (i.e., vocalisations of PLV4 and PLV6). Except for RTT3, all infants were observed to produce canonical PLVs conforming to the target language (i.e., vocalisations of PLV5 and PLV7). Figure1 presents a generally higher proportion of canonical PLVs conforming to the target language in the TD than in the RTT group (for exact values please refer to bottom of Table2). TD1, the infant with the highest CBRUTTER across all months, was the one who produced the highest proportion of PLV5 and PLV7. Besides their comparable CBRsUTTER, TD3 and RTT1 demonstrated comparable distributions of both their canonical and non-canonical PLV subtypes across time. RTT3 demonstrated a distinct distribution of the non-canonical PLV subtypes at 9months, producing a much higher proportion of PLV3 than the other infants. Discussion In this study, using retrospective video analysis, we investigated and compared the pre-linguistic vocal behaviours of three infants later diagnosed with RTT with their age-matched TD peers. Following findings deserve special comments. Canonical Babbling Ratio Our results show that all infants with RTT and all TD infants reached the canonical babbling milestone by 9months, i.e. CBRsUTTER ≥ 0.15 (Table2, Fig.1). As canonical babbling is achieved by most children with typical outcome when they are between 5 and 10months of age (Lang etal., 2019; Morgan & Wren, 2018; Oller, 1980), the three infants with RTT met the canonical babbling milestone in time. This finding confirms that reaching the canonical babbling milestone in time alone is not sufficient to predict typical speech-language development (Lang etal., 2021; Oller etal., 1998, 1999). Even though all infants met the canonical babbling milestone in time, we found differences in the CBRsUTTER between the RTT and the TD group: Despite the varying and partly small number of available data per month, the TD group had overall considerably higher CBRsUTTER than the RTT group. The TD infants clearly and consistently exceeded the canonical babbling threshold (0.15) by producing a high proportion of canonical vocalisations. TD1 and TD2 had the highest CBRsUTTER of all infants in all three months, whereas the lowest CBRsUTTER were observed in RTT2 and RTT3. Our findings are in line with the studies focusing on other developmental disorders: Lower CBRs in comparison with TD infants were previously reported for infants with autism spectrum disorder (Patten etal., 2014) and fragile X Journal of Developmental and Physical Disabilities 1 3 with the preserved speech variant of Rett syndrome. Developmental Neurorehabilitation, 17(4), 284– 290. https:// doi. org/ 10. 3109/ 17518 423. 2013. 783139 Molemans, I., van den Berg, R., van Severen, L., & Gillis, S. (2012). How to measure the onset of babbling reliably? Journal of Child Language, 39(3), 523–552. https:// doi. org/ 10. 1017/ s0305 00091 10001 71 Morgan, L., & Wren, Y. E. (2018). A systematic review of the literature on early vocalizations and babbling patterns in young children. Communication Disorders Quarterley, 40(1), 3–14. https:// doi. org/ 10. 1177/ 15257 40118 760215 Nathani, S., Ertmer, D. J., & Stark, R. E. (2006). Assessing vocal development in infants and toddlers. Clinical Linguistics & Phonetics, 20(5), 351–369. https:// doi. org/ 10. 1080/ 02699 20050 02114 51 Nathani, S., & Oller, D. K. (2001). Beyond ba-ba and gu-gu: Challenges and strategies in coding infant vocalizations. Behavior Research Methods, Instruments, & Computers, 33(3), 321–330. https:// doi. org/ 10. 3758/ BF031 95385 Nathani, S., Oller, D. K., & Neal, A. R. (2007). On the robustness of vocal development: An examination of infants with moderate-to-severe hearing loss and additional risk factors. Journal of Speech, Language, and Hearing Research, 50(6), 1425–1444. https:// doi. org/ 10. 1044/ 10924388(2007/ 099) Neul, J. L., Kaufmann, W. E., Glaze, D. G., Christodoulou, J., Clarke, A. J., Bahi-Buisson, N., Leonard, H., Bailey, M. E. S., Schanen, N. C., Zappella, M., Renieri, A., Huppke, P., Percy, A. K., & RettSearch Consortium. (2010). Rett syndrome: Revised diagnostic criteria and nomenclature. Annals of Neurology, 68(6), 944–950. https:// doi. org/ 10. 1002/ ana. 22124 Nyman, A., & Lohmander, A. (2018). Babbling in children with neurodevelopmental disability and validity of a simplified way of measuring canonical babbling ratio. Clinical Linguistics & Phonetics, 32(2), 114–127. https:// doi. org/ 10. 1080/ 02699 206. 2017. 13205 88 Oller, D. K. (1980). The emergence of the sounds of speech in infancy. In G. H. Yeni-Komshian, J. F. Kavanagh, & C. A. Ferguson (Eds.).Child Phonology, (1sted.,). Academic Press.93–112 Oller, D. K. (2000). The Emergence of the Speech Capacity. Lawrence Erlbaum Associates. Oller, D. K., & Eilers, R. E. (1988). The role of audition in infant babbling. Child Development, 59(2), 441–449. Oller, D. K., Eilers, R. E., Neal, A. R., & Cobo-Lewis, A. B. (1998). Late onset canonical babbling: A possible early marker of abnormal development. American Journal of Mental Retardation, 103(3), 249–263. https:// doi. org/ 10. 1352/ 08958017(1998) 103% 3c0249: LOCBAP% 3e2.0. CO;2 Oller, D. K., Eilers, R. E., Neal, A. R., & Schwartz, H. K. (1999). Precursors to speech in infancy: The prediction of speech and language disorders. Journal of Communication Disorders, 32(4), 223–245. https:// doi. org/ 10. 1016/ S00219924(99) 00013-1 Oller, D. K., Eilers, R. E., Steffens, M. L., Lynch, M. P., & Urbano, R. (1994). Speech-like vocalizations in infancy: An evaluation of potential risk factors. Journal of Child Language, 21(1), 33–58. https:// doi. org/ 10. 1017/ S0305 00090 00086 67 Ozonoff, S., Iosif, A. M., Young, G. S., Hepburn, S., Thompson, M., Colombi, C., Cook, I. C., Werner, E., Goldring, S., Baguio, F., & Rogers, S. J. (2011). Onset patterns in autism: Correspondence between home video and parent report. Journal of the American Academy of Child and Adolescent Psychiatry, 50(8), 796-806.e1. https:// doi. org/ 10. 1016/j. jaac. 2011. 03. 012 Palomo, R., Belinchon, M., & Ozonoff, S. (2006). Autism and family home movies: A comprehensive review. Journal of Developmental and Behavioral Pediatrics, 27(2 Suppl), 59–68. https:// doi. org/ 10. 1097/ 00004 70320060 400200003 Papousek, M. (1994). Vom ersten Schrei zum ersten Wort : Anfänge der Sprachentwicklung in der vorsprachlichen Kommunikation. Hans Huber Verlag Patten, E., Belardi, K., Baranek, G. T., Watson, L. R., Labban, J. D., & Oller, D. K. (2014). Vocal patterns in infants with autism spectrum disorder: Canonical babbling status and vocalization frequency. Journal of Autism and Developmental Disorders, 44(10), 2413–2428. https:// doi. org/ 10. 1007/ s108030142047-4 Pokorny, F. B., Bartl-Pokorny, K. D., Einspieler, C., Zhang, D., Vollmann, R., Bölte, S., Gugatschka, M., Schuller, B. W., & Marschik, P. B. (2018). Typical vs. atypical: Combining auditory gestalt perception and acoustic analysis of early vocalisations in Rett syndrome. Research in Developmental Disabilities, 82, 109–119. https:// doi. org/ 10. 1016/j. ridd. 2018. 02. 019 Pokorny, F. B., Marschik, P. B., Einspieler, C., & Schuller, B. W. (2016). Does she speak RTT? Towards an earlier identification of Rett syndrome through intelligent pre-linguistic vocalisation analysis. In N. Morgan (Ed.), Proceedings Interspeech (pp. 1953–1957). IEEE. Roche, L., Zhang, D., Bartl-Pokorny, K. D., Pokorny, F. B., Schuller, B. W., Esposito, G., Bölte, S., Roeyers, H., Poustka, L., Gugatschka, M., Waddington, H., Vollmann, R., Einspieler, C., & Marschik, P. B. (2018). 1 3 Journal of Developmental and Physical Disabilities Early vocal development in autism spectrum disorder, Rett syndrome, and fragile X syndrome: Insights from studies using retrospective video analysis. Advances in Neurodevelopmental Disorders, 2(1), 49–61. https:// doi. org/ 10. 1007/ s412520170051-3 Schauwers, K., Gillis, S., Daemers, K., De Beukelaer, C., & Govaerts, P. J. (2004). Cochlear implantation between 5 and 20 months of age: The onset of babbling and the audiologic outcome. Otology & Neurotology, 25(3), 263–270. https:// doi. org/ 10. 1097/ 00129 49220040 500000011 Schramm, B., Bohnert, A., & Keilmann, A. (2009). The prelexical development in children implanted by 16 months compared with normal hearing children. International Journal of Pediatric Otorhinolaryngology, 73(12), 1673–1681. https:// doi. org/ 10. 1016/j. ijporl. 2009. 08. 023 Stark, R. E. (1980). Stages of speech development in the first year of life. In G. H. Yeni-Komshian, J. F. Kavanagh, & C. A. Ferguson (Eds.).Child Phonology, (1sted.,). Academic Press. 73–92 Stark, R. E. (1981). Infant vocalization: A comprehensive view. Infant Mental Health Journal, 2(2), 118– 128. https:// doi. org/ 10. 1002/ 10970355(198122) 2:2% 3c118:: AIDIMHJ2 28002 0208% 3e3.0. CO;2-5 Tarquinio, D. C., Hou, W., Neul, J. L., Lane, J. B., Barnes, K. V., O’Leary, H. M., Bruck, N. M., Kaufmann, W. E., Motil, K. J., Glaze, D. G., Skinner, S. A., Annese, F., Baggett, L., Barrish, J. O., Geerts, S. P., & Percy, A. K. (2015). Age of diagnosis in Rett syndrome: Patterns of recognition among diagnosticians and risk factors for late diagnosis. Pediatric Neurology, 52(6), 585–591. e2. https:// doi. org/ 10. 1016/j. pedia trneu rol. 2015. 02. 007 Townend, G. S., Bartl-Pokorny, K. D., Sigafoos, J., Curfs, L. M., Bölte, S., Poustka, L., Einspieler, C., & Marschik, P. B. (2015). Comparing social reciprocity in preserved speech variant and typical Rett syndrome during the early years of life. Research in Developmental Disabilities, 43–44, 80–86. https:// doi. org/ 10. 1016/j. ridd. 2015. 06. 008 Yankowitz, L. D., Schultz, R. T., & Parish-Morris, J. (2019). Preand paralinguistic vocal production in ASD: Birth through school age. Current Psychiatry Reports, 21(12), 126. https:// doi. org/ 10. 1007/ s119200191113-1 Zoghbi, H. Y. (2005). MeCP2 dysfunction in humans and mice. Journal of Child Neurology, 20(9), 736– 740. https:// doi. org/ 10. 1177/ 08830 73805 02000 90701 Publisher’s Note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations. Authors and Affiliations KatrinD.Bartl‑Pokorny1,2 · FlorianB.Pokorny1,2 · DuniaGarrido3 · BjörnW.Schuller2,4 · DajieZhang1,5,6 · PeterB.Marschik1,5,6,7 1 iDN – interdisciplinary Developmental Neuroscience, Division ofPhoniatrics, Medical University ofGraz, Graz, Austria 2 EIHW – Chair ofEmbedded Intelligence forHealth Care andWellbeing, University ofAugsburg, Augsburg, Germany 3 Mind, Brain, andBehaviour Research Centre, University ofGranada, Granada, Spain 4 GLAM – Group onLanguage, Audio, & Music, Department ofComputing, Imperial College London, London, UK 5 Child andAdolescent Psychiatry andPsychotherapy, Systemic Ethology andDevelopmental Science, University Medical Center Göttingen, Georg-August University Göttingen, Göttingen, Germany 6 Leibniz ScienceCampus Primate Cognition, Göttingen, Germany 7 Center ofNeurodevelopmental Disorders (KIND), Department ofWomen’s andChildren’s Health, Karolinska Institutet, Stockholm, Sweden