unlimited access with print and download

Free

Continue searching

Vocalizations of children with autism spectrum disorders late in the second year of life

Dissertation
Author: Allison M. Plumb
Abstract:
Impairment in communication is a primary diagnostic feature of autism spectrum disorders (ASD), however relatively little is known about the early communication development of children with ASD. Vocalizations, which typically begin to emerge within the first year of life, are therefore an area of communication prime for research as a potential behavioral marker of ASD which can be assessed early in life. The purpose of this study was to compare differences in the vocalizations of children between 18 and 24 months of age later diagnosed with ASD ( n = 50), a mentally and chronologically age-matched group of children with developmental delay (DD; n = 25), and a chronologically age-matched group of children with typical development (TD; n = 50). In addition, for children with ASD, concurrent and predictive relationships between vocalizations and developmental level were investigated. Precise measures of vocalizations were obtained through systematic observation of videotaped behavior samples from the Communication and Symbolic Scales Developmental Profile (Wetherby & Prizant, 2002). Children with ASD used significantly fewer vocalizations and a lower proportion of vocalizations with consonants than children with TD. In addition, children with ASD used a significantly higher proportion of atypical vocalizations than children with TD and a significantly higher proportion of distress vocalizations than both children with TD and DD. For the ASD group, the frequency of vocalizations and the frequency of vocalizations containing recognizable speech sounds correlated significantly with developmental levels in both the second and third years. In addition, communicative vocalizations late in the second year were found to uniquely predict expressive language outcome at age 3 over and above noncommunicative vocalizations. Overall, the results of this study indicated that by 18 to 24 months of age many vocalization measures significantly differentiate children with ASD from children with TD, with a higher proportion of distress vocalization differentiating children with ASD from both children with TD and DD. The importance of communicative vocalizations for later language development is highlighted by the results of this investigation. This study will contribute to the understanding of early vocal communication in children with ASD and assist in early identification.

v TABLE OF CONTENTS

List of Tables ................................................................................................................ .. viii List of Figures ............................................................................................................... ..... ix Abstract…….. .................................................................................................................. ....x

1. INTRODUCTION ..........................................................................................................1 Vocalizations in Typical Development ....................................................................1 Vocalizations of Preschool Children with ASD ......................................................3 Profiles of Communicative Functions in Young Children with ASD .....................5 Vocalizations of Children with ASD Under Two Years of Age .............................7 Analyses of home videotapes .......................................................................7 Systematic observation ................................................................................7 Relationship Between Vocalizati ons and Later Spoken Language .........................9 Child-Driven and Transactional Theo ries of Language Development ..................11 Summary and Rationa le for the Study ...................................................................12 Statement of Purpose and Research Questions ......................................................13 Expected Outcomes ...............................................................................................14

2. METHOD .................................................................................................................... 15 Participant Recruitment .........................................................................................15 ASD group .................................................................................................16 DD group ...................................................................................................16 TD group ....................................................................................................16 Participant demographics ...........................................................................17 Participant developmental characteristics ..................................................18 Second Year Measures ...........................................................................................20 CSBS DP behavior samples .......................................................................20 Vocalizations..............................................................................................20 Categories of vocalizations ........................................................................21 Transcribable vocalizations .......................................................................21

vi Nontranscribable vocalizations ..................................................................22 Communicative Acts ..................................................................................23 Interrater Reliability ...............................................................................................23 Transcribable and nontranscribable vocalizations .....................................23 Transcribable vocalization categories ........................................................24 Nontranscribable vocalization categories ..................................................24 Communicative acts ...................................................................................25 3. RESULTS ....................................................................................................................26 Vocalization Measures derived from the Noldus Pro Observer ® ..........................26 Preliminary data analyses ..........................................................................27 Vocalizations..............................................................................................27 Communicative acts ...................................................................................28 Determination of analysis procedures ........................................................29 Rate of Vocalizations in Structured and Unstructured Contexts ...........................29 Group Differences in Vocalizations .......................................................................29 Total Vocalizations ....................................................................................30 Transcribable vocalizations .......................................................................30 Nontranscribable vocalizations ..................................................................35 Group Differences in Communicative Vocalizations ............................................36 Total vocalizations in communicative acts ................................................37 Transcribable vocalizations in communicative acts ..................................37 Nontranscribable vocalizations in communicative acts .............................40 Noncommunicative vocalizations ..............................................................40 Concurrent Relations with Development in the Second Year ...............................42 Vocalizations..............................................................................................42 Communicative and noncommunicative vocalizations .............................42 Predictive Relationships with Development Outcome at Age 3 ............................44 Vocalizations..............................................................................................44 Communicative and noncommunicative vocalizations .............................44 Unique contribution of communicative vocalizations ...............................46

vii 4. DISCUSSION ...............................................................................................................48 Communication Sampling in Structured and Unstructured Contexts ....................49 Vocalizations Between 18 and 24 months of age ..................................................49 Communicative Vocalizations Between 18 and 24 Months of Age ......................51 Concurrent Relations with Developmental Measure in the Second Year ..............51 Predictive Relations with Developmental Outcome at Age Three ........................52 Unique Contribution of Communicative Vocalizations ........................................53 Clinical Implications ..............................................................................................53 Limitations .............................................................................................................55 Future Directions ...................................................................................................56 Summary ................................................................................................................57

APPENDICES ...................................................................................................................59 Appendix A: Vocalizations Coding Definitions ...................................................59 Appendix B: Human Participants Approval ..........................................................63

REFERENCES .................................................................................................................70

BIOGRAPHICAL SKETCH .............................................................................................77

viii LIST OF TABLES

Table 1: Summary of Participant Demographics ...............................................................17

Table 2: Summary of Devel opmental Characteristics .......................................................18

Table 3: Reliability Kappas and Percent Agreement .........................................................25

Table 4: Rate of Vocalizations in Structur ed Compared to Unstructured Contexts ..........30

Table 5: Group Differences in Fre quency Measures of Vocalizations ..............................31

Table 6: Group Differences in Propor tion Measures of Vocalizations ..............................32

Table 7: Group Differences in Frequency Measures of Communicative Vocalizations ...38

Table 8: Group Differences in Proportion M easures of Communicative Vocalizations ...39

Table 9: Correlation Among the Vocalization Measures and Developmental Measures in the Second Year and at Age 3 for Child ren with Autism Spectrum Disorders .................43

Table 10: Correlations Among the Vocalizatio n Measures and Developmental Measures and Developmental Measures in the Second Y ear and at Age 3 for Children with Autism Spectrum Disorders ............................................................................................................ 45

Table 11: Hierarchical Regression Resu lts for Communicative Vocalizations .................47

ix LIST OF FIGURES

Figure 1: Frequency of vocalizations by type ....................................................................33

Figure 2: Proportion of TVs and NTVs out of total vocalizations ....................................33

Figure 3: Proportion of TV levels out of total TVs ...........................................................34

Figure 4: Proportion of single syllable a nd multisyllable TVs out of total TVs................35

Figure 5: Proportion of NTV Cate gories out of total NTVs ..............................................36

Figure 6: Proportion of communicative vocaliz ations out of total vocalizations ..............37

Figure 7: Proportion of communicative TV s in acts for BR, SI and JA ............................40

Figure 8: Proportion of communicative NT Vs in acts for BR, SI, and JA ........................41

Figure 9: Proportion of nonco mmunicative vocalizations .................................................41

x ABSTRACT

Impairment in communication is a primar y diagnostic feature of autism spectrum disorders (ASD), however relatively little is known about the early communication development of children with ASD. Vocaliz ations, which typically begin to emerge within the first year of life, are therefore an area of communication prime for research as a potential behavioral marker of ASD which can be assessed early in life. The purpose of this study was to compare differences in th e vocalizations of ch ildren between 18 and 24 months of age later diagnosed with ASD ( n = 50), a mentally and chronologically age- matched group of children with developmental delay (DD; n = 25), and a chronologically age-matched group of children with typical development (TD; n = 50). In addition, for children with ASD, concurrent and predictiv e relationships between vocalizations and developmental level were investigated. Precise measures of vocali zations were obtained through systematic observation of vide otaped behavior samples from the Communication and Symbolic Scales Developmental Profile (Wetherby & Prizant, 2002). Children with ASD used significantly fewer vocalizations a nd a lower proportion of vocalizations with consonants than children with TD. In addi tion, children with ASD used a significantly higher proportion of atypical vocalizations than children with TD and a significantly higher proportion of distress vo calizations than both children with TD and DD. For the ASD group, the frequency of vocalizations and the frequency of vocalizations containing recognizable speech sounds correlated significantl y with developmental levels in both the second and third years. In addition, communicat ive vocalizations late in the second year were found to uniquely predict expressive language outcome at age 3 over and above noncommunicative vocalizations. Overall, the resu lts of this study indicated that by 18 to 24 months of age many vocalization measures significantly differentiate children with ASD from children with TD, with a hi gher proportion of di stress vocalization differentiating children with AS D from both children with TD and DD. The importance of communicative vocalizations for later la nguage development is highlighted by the results of this investigation. This study will contribute to the understanding of early vocal communication in children with ASD and assist in early identification.

1 INTRODUCTION

The positive outcomes associated with early identification and intervention for children with autism spectrum disorders (A SD) are well documented. A growing body of evidence exists suggesting that intervention beginning before 3½ years of age results in greater long term benefits th an intervention beginning late r in life (Daws on & Osterling, 1997; Fenske, Zalenski, Kr antz, & McClannahan, 1985;

Harris and Handleman, 2000; McGee, Morrier, & Daly, 1999; NRC, 2001). Desp ite this knowledge, the average age of diagnosis for autism is still above age 3 (Mandell, Novak, & Zubritsky, 2005; NRC, 2001; Wiggins, Baio, & Rice, 2006) . Compounding the difficulty in early identification is the lack of a biological marker for the disorder . As a result, it is even more important for research to continue identifyi ng behavioral markers that can be observed early in life. Communication impairments are one such area of potential. Impairment in communication is a primar y diagnostic feature of autism spectrum disorders (American Psychiatric As sociation, 2000). According to the Diagnostic and Statistical Manual of Mental Diso rders Fourth edition, text revisions (DSM-IV-TR; APA, 2000), deficits in communication are dem onstrated by delay in or total lack of spoken language and gestures, impairment in initiating or sustaining conversation, stereotyped and repetit ive use of language, and/or lack of pretend play. Because the average age of diagnosis is st ill over 3 years of age, relativ ely little is known about the early vocal development of children with ASD. In addition, some di agnostic features of ASD related to communication as listed by th e DSM-IV-TR are not applicable to very young children, such as sustaining conversati on and stereotyped language. Research is therefore important to unders tand the early communicative prof iles of children with ASD under 3 years of age. Vocalizations, which typica lly develop within the first year of life and are necessary for spoken language, ar e an area of communication prime for investigation as a potential behavioral marker of ASD which can be as sessed early in life. Vocalizations in Typical Development

Perhaps the most widely accepted descrip tions of early vocal development are those of Oller (1980) and Stark (1980). Both stage models are based on the commonalities of past research in the field of infant vocal development (McCarthy, 1952; Oller & Smith, 1977; Stark, 1978; Stark, Rose , & McLagen, 1975; Zlatin, Laufer, &

2 Horii, 1977) and continue to be supported by the robust findings of extensive longitudinal and cross-sectional investigations of infant vocalizations (Kent & Bauer, 1985; Menyuk, Liebergott, & Schultz, 1986; Mitchell & Kent, 1990; Oller, 1986; Oller, Eilers, Bull, & Carney, 1985; Smith, Brown-Sweeney, & Stoel-Gammon, 1989; Smith & Oller, 1981; Stark, Bernstein, & Demorest, 1993). Although large sample sizes were not obtained for a majority of these studies, the thoroughness of the data collection procedures and the large number of studies providing supporting evidence contribute to the robustness of the findings. Based on the convergent evidence from these well-designed studies, an understanding of typical speech development has emerged which demonstrates a continuity between birth and first words, but also acknowledges the potential for individual variability. Now referred to as an “infraphonological” perspective, Oller’s stages trace vocal development in terms of “speechiness” (1986). Vocalizations are evaluated in terms of speech-like properties that they do or do not possess and described in terms of concrete phonetic speech parameters, such as pitch, vocal quality, resonance pattern, timing, and amplitude. To a great degree, Stark’s stages of vocal development (1980; 1986) mirror those of Oller, however, in addition Stark includes classification for reflexive vocalizations (i.e., cries, coughs, hiccups), which cannot be considered “speechy,” and are therefore omitted by Oller. Both Oller’s and Stark’s stages of infant vocal development are based on the commonalities of previous research in infant vocalization. Although there are some differences in terminology and age of emergence estimates, a general overlap exists allowing early vocalizations to be evaluated in terms of typicality. Of particular relevance to the current study is the emergence of canonical syllables, which are typically found to emerge between 6 and 10 months of age (Oller, 1980). Entry into this stage is usually recognized by the use of reduplicated syllables (e.g., /mama/), although there is not yet indication of sound-meaning correspondence. The consonant inventory is still rather limited at this stage, with stops, nasals, and glides observed the most frequently. Nonreduplicated babbling begins to occur around 11 months of age (Stark, 1986) in typically developing children, in which infants begin to produce single or multisyllablic utterances containing differing consonants or vowels. At this time, the co-occurrence of first words and jargon can also be observed. Although syllables containing consonants and vowels typically begin to occur prior to one year of

3 age, it is important to consider that vowels remain the most frequently observed sound type until 16 to 20 months (Kent & Bauer, 1985; Nathani, Ertmer, & Stark, 2008). In addition to considering syllable shape and consonant use of transcribable vocalizations (TVs) that contain at least a vowel and are therefore considered syllabic, it is also valuable to consider nontranscribable vocalizations (NTVs) which do not contain vowels or contain vowels with atypical phonation. NTVs include sounds such as laughs, cries, growls, yells, and screams. The earlier developing NTVs, such as cries, emerge shortly after birth, where later developing NTVs, such as squeals and growls, typically emerge between about 4 and 7 months (Oller, 1980). Also included as NTVs are certain phonemes considered consonantal in languages other than English, such as bilabial trills (raspberries), uvular trills, alveolar trills, and tongue clicks. Although not considered consonantal in their native language, English speaking children may produce these sounds during vocal play or exploration and are therefore important to include in this category. With the exception of alveolar trills, these non-English NTVs emerge within the first year of life in typically developing children (Vihman, 1996). Alveolar trills, however, still may not be mastered by typically developing children at 4 years of age (Jimenez, 1987; Linares, 1981). Wetherby, Cain, Yonclas, and Walker (1988) studied the TVs and NTVs of 15 typically developing children during the prelinguistic, one-word and multiword stages. Results indicated that NTVs were found to decrease significantly from the prelinguistic to the one-word stage whereas the percentage of multisyllable vocalizations increased significantly. In addition, the proportion of single and multisyllablic TVs containing consonants increased from the prelinguistic to the one-word stage for 11 of the 15 participants. By the one-word stage all of the participants used some TVs containing consonants and 13 of 15 used these as their primary vocalization type. Vocalizations of Preschool Children with ASD A growing number of studies have been conducted that examine the vocal communication of preschool and early school-age children with ASD. This research provides valuable evidence regarding a deficit in the vocal development of children with ASD, however the findings of each study could be strengthened with larger sample sizes. Wetherby, Yonclas, & Bryan (1989), for example, examined the syllabic shape of vocalizations used in the intentional communicative acts of 3 preschool children with

4 autistic behaviors and mental retardation, 4 preschool children with Down syndrome, and 4 preschool children with Specific Language Impairment (SLI). Analysis of 30 minute communication samples revealed that the children with characteristics of autism and mental retardation used a deficient proportion of vocal acts with a consonant when compared to normative data on typically developing children and also suggested that they displayed an excessive production of NTVs (Wetherby et al., 1988). Stone, Ousley, Yoder, Hogan, and Hepburn (1997) compared 14 children with ASD ranging from 27 to 38 months (M = 32.8, SD = 3.5) to 14 children with DD/Language Impaired (LI) matched on CA, MA, and expressive vocabulary and found no significant differences in the use of vocal communicative acts without gestures. Wetherby, Prizant, & Hutchison (1998) compared the Communication and Symbolic Behavior Scales Developmental Profile (CSBS DP) Behavior Samples of 22 participants with PDD ranging in age from 17 to 60 months (M = 35.9, SD = 11.72) to 22 participants with DD ranging in age from 19 to 66 months (M = 32.4, SD = 9.91). Based on the participants’ mean scaled scores, the PDD group used fewer syllables with consonants than the DD group, with a medium effect size obtained and fewer multisyllables with a small effect size achieved. Findings were consistent with those of Stone et al. (1997) who found that preschool children with ASD demonstrated similar use of communicative acts with vocalizations that did not involve gestures as children with DD. Although also limited by its small sample size, perhaps the most detailed study on vocalizations in children with ASD to date is that of Sheinkopf, Mundy, Oller, & Steffens (2000). These investigators compared the vocalizations of 11 preverbal children with developmental delay to 15 preschool children with ASD matched on nonverbal MA and verbal ability. Vocal samples were coded from video recordings of Early Social and Communication Scales (ESCS, 1996) samples. Vocalizations were coded to include breath-group based utterances, syllable structure, vocal quality, and frequency counts of utterance features associated with emotional states, such as laughing and crying. The results indicated that the ASD group used a comparable proportion of syllables containing consonants as the DD group, but a significantly higher proportion of syllables with atypical vocal quality, such as yells, squeals, and growls (M =.25, SD=.14). Despite being limited by small sample sizes, the convergent findings of these studies involving preschool children with ASD have documented deficits in vocalizations

5 when compared to peers with typical development. These results highlight the need for research on the vocalization development of children with ASD under 2 years of age in order to better understand their early communicative profiles and whether these deficits may be determined at a younger age. Profiles of Communicative Functions in Young Children with ASD There is a growing body of research demonstrating a unique profile of communicative functions or reasons for communicating in young children with ASD. To better understand the vocal abilities of a child with ASD it is important to consider how vocalization deficits may manifest in relation to this unique profile of communicative functions. By the end of the first year of life typically developing children begin to communicate for three primary functions: behavior regulation, social interaction, and joint attention (Bruner, 1981; Wetherby & Prizant, 2002). Behavior regulation involves communicating in order to regulate the behavior of another for a desired outcome, such as requesting or protesting. Communicating for social interaction involves drawing another’s attention to oneself, such as greeting or showing off, while joint attention involves drawing another’s attention to an object or event for the purpose of sharing it with others (e.g., requesting information, commenting). In typically developing children these behaviors develop concurrently (Carpenter, Mastergeorge, & Coggins, 1983; Crais, Douglas, & Campbell, 2004; Wetherby, 1986), however in children with ASD differences in the emergence of these functions and their rate of use have been found. Research has documented that communicating for behavior regulation is an area of strength for children with ASD, while communicating for joint attention is a deficit (McEvoy, Rogers, & Pennington, 1993; Mundy, Sigman, & Kasari, 1990, 1994; Sigman, Mundy, Sherman, & Ungerer, 1986; Wetherby & Prutting, 1984; Wetherby et al., 1998). Research is emerging on children under 2 years of age showing this same profile of communicative functions. Using a large sample drawn from the same participant pool as the current investigation, Shumway (2006) compared 50 children with ASD in the second year of life to 50 chronological age (CA) matched TD children and 25 mental age (MA) matched and CA matched DD children using precise coding of communicative acts with the Noldus Pro Observer © software during CSBS DP Behavior Samples. Results indicated that children with ASD communicated for behavior regulation at a significantly lower

6 rate (M = .76, SD = .43) than the TD group and a significantly lower rate for joint attention (M = .26, SD = .45) than the TD and DD groups. The three groups were similar on the rate of acts for social interaction (M = .24, SD = .26) however, this may be in part due to the low occurrence rate of this behavior across all groups. Because the ASD group used significantly fewer total acts (M = 22.42, SD = 14.91) than both the TD and DD groups, proportions were also used to measure differences between groups on the use of functions. Based on proportion measures, the ASD group used a significantly higher proportion of acts for behavior regulation (M = 65.34, SD = 21.89) and social interaction (M = 19.97, SD = 20.65) than the TD group, but a lower proportion of acts for joint attention (M = 14.69, SD = 14.91) compared to TD and DD. In a study using the same sample as Shumway (2006), Wetherby, Watt, Morgan, & Shumway (2007) examined social communication profiles from CSBS DP Behavior Samples videotaped between 18 to 24 months of age (M = 21.36; SD = 1.90) for 50 children with ASD, 23 with DD, and 50 with TD. Although the samples were identical for these studies it is important to note the difference in their measures. Wetherby and colleagues (2007) based results on the mean scores on the standardized clinical scoring of the CSBS DP Behavior Samples. Shumway (2006) provided additional detail by precisely coding samples using Noldus Pro Observer © software version 5.0 by Noldus Information Technology for each communicative act and its function. Despite the difference in methodology, the findings were similar to those of Shumway (2006) in that children with ASD scored significantly lower on joint attention than children with DD and TD and significantly higher on behavior regulation and social interaction than children with TD. In summary, previous research provides consistent evidence that communication for joint attention is a core deficit in young children with ASD, with communication for behavior regulation presenting as a relative strength. Therefore, a lower frequency of vocalizations used in acts for JA would potentially be likely during communication samples of children with ASD. To address this profile of strengths and weaknesses across communicative functions, research utilizing proportion measures of vocalizations used for different functions would be important. This will provide information on whether children with ASD are as likely to vocalize for the functions of BR, SI, and JA, as both the TD and DD groups when using vocalizations communicatively.

7 Vocalizations of Children with ASD under 2 Years of Age Evidence is beginning to emerge on the vocalizations of children with ASD under 2 years of age. This evidence has come from two primary sources: analysis of home videotapes and systematic observation. Although some of these studies are limited by weaknesses in methodology, they have provided important information that help to direct the study of early vocalizations of young children with ASD. Analyses of home videotapes. The examination of home videotapes collected from first birthday parties is one methodology used to investigate early social and communication abilities in very young children with ASD, but has provided mixed results in terms of vocalization findings. In one such study, Osterling, Dawson, & Munson (2002) did not find group differences when they compared duration of time spent vocalizing in the first birthday videotapes of 20 children with ASD, 14 children with DD and 20 children with TD, however this may in part be due to a small sample size. In a later investigation by Werner and Dawson (2005) significant differences were detected when they compared the first and second birthday videotapes of 36 children with ASD, 21 with early onset symptoms and 15 with reported regression, and 18 children with TD. The early onset ASD group differed significantly from the group with regression and the group with TD on measures of complex babbling at 12 months. By 24 months both ASD groups differed from TD on complex babbling. In addition to the use of a larger sample size, this study also measured vocalizations by the frequency of simple and complex babbling. This coding scheme was more detailed than that used by Osterling and colleagues (2002), which did not differentiate vocalizations by levels of complexity. These distinctions may account for their ability to find group differences. Systematic observation. Research is also emerging for children with ASD under 2 years of age based on direct, systematic observation. One major cohort consists of children identified from a general population screen using the CSBS DP. Wetherby and colleagues (2004) used the Systematic Observation of Red Flags (SORF) for Autism Spectrum Disorders in Young Children (Wetherby & Woods, 2002) to investigate a subset of this cohort in the second year of life and found that a lack of communicative vocalizations was one of 13 red flags that differentiated 18 children with ASD from 18 children with TD. The information provided by this study is valuable, but is negated by a

8 relatively small sample and large age range (13.1 to 26.9 months). These factors, in addition to the SORF’s restricted scoring range (e.g., 0, 1, 2), limit the study’s ability to find group differences. To better understand the communication abilities in young children with ASD, research with larger sample sizes and narrower age ranges is needed. Results of Wetherby and colleagues’ investigation (2007) of the communicative profiles of children with ASD, DD, and TD late in the second year of life indicated that a lack of well-formed syllables containing consonants differentiated children with ASD from children with TD, however not from DD. Building on what was learned from these investigations and using children from the same cohort, Plumb, Wetherby, Kahn, and Barber (2007) investigated vocalizations in the latter half of the second year for 50 children with ASD, 25 children with DD, and 50 children with TD. All vocalizations observed during the videotaped Behavior Samples were precisely coded as TVs or NTVs to determine their frequency of use by type. Supporting the findings of Wetherby & Woods (2002) and Wetherby and colleagues (2007), children with ASD were found to produce significantly fewer TVs than children with TD, but not DD. Although group differences in NTVs were not found, the specific types of NTVs (i.e., laugh, cry, squeal, tongue click) were not included in the coding scheme, rather vocalizations were coded dichotomously as transcribable or nontranscribable. Future research is needed to look more precisely at transcribable and nontranscribable vocalizations, specifically at consonant use and the number of syllables in TVs and specific categories of NTVs used by children with ASD in the second year of life. Investigating in more detail will provide needed information on the early vocalizations of children with ASD, as well as provide additional opportunity to illustrate potential group differences on NTVs that have been demonstrated by previous research with small sample sizes (Sheinkopf et al., 2000; Wetherby et al., 1989). In addition to examining communicative acts by function, Shumway (2006) also looked at the means of these acts, including both gesture and vocalization use. Results of this investigation indicated that the ASD group used a significantly lower proportion of acts with vocalizations than the TD group. Plumb and colleagues (2007) also looked at the communicative use of vocalizations and found that the ASD group demonstrated significantly fewer TVs in acts than the TD group, however there were no significant differences between the ASD group and the DD group. Proportion measures similarly

9 indicated that the ASD group used significantly lower proportions of TVs within communicative acts than their age matched TD peers. What is not known from this investigation is whether there were group differences in the TVs in terms of syllable structure and consonant use. TVs were coded uniformly and not examined to describe a hierarchy of developmental complexity. Further investigation with increased detail is recommended to provide precise information on the communicative use of specific vocalization types in young children. In summary, there is a general convergence of evidence that children with ASD use fewer TVs communicatively than children with TD. In addition, evidence is beginning to emerge which indicates that the presence of atypical NTVs, such as squeals, growls, and tongue clicks, can distinguish young children with ASD from young children with DD, making it especially important for early identification. Detailed research is now needed to look at the specific phonologic and syllabic composition of TVs, as well as the types of NTVs being used by children with ASD late in the second year of life. A careful investigation into these areas utilizing a large sample size, a narrow age range, and systematic observation will make it more likely to detect group differences, providing a clearer picture of the vocal communication abilities of young children with ASD. Relationship between Vocalizations and Later Spoken Language Although initially debated in the field, research has long supported the theory of continuity existing between prelinguistic babble and speech in typically developing children. Jakobson’s (1968) view that children begin with a diverse set of speech sounds during babble and subsequently lose those sounds only to start over with a limited repertoire of consonants and vowels in first words has since been refuted by evidence that the syllable structure, place, and manner of articulation, and vocalization length used most commonly in the canonical babbling period are also found in first word productions (Stoel-Gammon, 1985; Stoel-Gammon & Cooper, 1984; Vihman, Ferguson, & Elbert, 1987; Vihman, Macken, Miller, Simmons, & Miller, 1985). In addition to studies providing evidence on the continuity of phonetic patterns of early babble and speech, longitudinal studies have also provided evidence supporting a relationship between prelinguistic vocalizations and language skills later in childhood (Menyuk, Liebergott, & Schultz, 1986; Stoel-Gammon, 1992, 1998; Vihman & Greenlee, 1987). In a recent investigation of 160 typically developing children, Watt, Wetherby, and Shumway (2006)

Full document contains 88 pages
Abstract: Impairment in communication is a primary diagnostic feature of autism spectrum disorders (ASD), however relatively little is known about the early communication development of children with ASD. Vocalizations, which typically begin to emerge within the first year of life, are therefore an area of communication prime for research as a potential behavioral marker of ASD which can be assessed early in life. The purpose of this study was to compare differences in the vocalizations of children between 18 and 24 months of age later diagnosed with ASD ( n = 50), a mentally and chronologically age-matched group of children with developmental delay (DD; n = 25), and a chronologically age-matched group of children with typical development (TD; n = 50). In addition, for children with ASD, concurrent and predictive relationships between vocalizations and developmental level were investigated. Precise measures of vocalizations were obtained through systematic observation of videotaped behavior samples from the Communication and Symbolic Scales Developmental Profile (Wetherby & Prizant, 2002). Children with ASD used significantly fewer vocalizations and a lower proportion of vocalizations with consonants than children with TD. In addition, children with ASD used a significantly higher proportion of atypical vocalizations than children with TD and a significantly higher proportion of distress vocalizations than both children with TD and DD. For the ASD group, the frequency of vocalizations and the frequency of vocalizations containing recognizable speech sounds correlated significantly with developmental levels in both the second and third years. In addition, communicative vocalizations late in the second year were found to uniquely predict expressive language outcome at age 3 over and above noncommunicative vocalizations. Overall, the results of this study indicated that by 18 to 24 months of age many vocalization measures significantly differentiate children with ASD from children with TD, with a higher proportion of distress vocalization differentiating children with ASD from both children with TD and DD. The importance of communicative vocalizations for later language development is highlighted by the results of this investigation. This study will contribute to the understanding of early vocal communication in children with ASD and assist in early identification.