Verbal Agreement Inflection in German Children With Down Syndrome Purpose The study aims to explore whether finite verbal morphology is affected in children/adolescents with Down syndrome (DS), whether observed deficits in this domain are indicative of a delayed or deviant development, and whether they are due to phonetic/phonological problems or deficits in phonological short-term memory. Method An ... Research Article
Open Access
Research Article  |   September 19, 2018
Verbal Agreement Inflection in German Children With Down Syndrome
 
Author Affiliations & Notes
  • Martina Penke
    Department of Special Education and Rehabilitation, University of Cologne, Germany
  • Disclosure: The author has declared that no competing interests existed at the time of publication.
    Disclosure: The author has declared that no competing interests existed at the time of publication. ×
  • Correspondence to Martina Penke: martina.penke@uni-koeln.de
  • Editor-in-Chief: Sean Redmond
    Editor-in-Chief: Sean Redmond×
  • Editor: Jan de Jong
    Editor: Jan de Jong×
Article Information
Special Populations / Cultural & Linguistic Diversity / Genetic & Congenital Disorders / Speech, Voice & Prosody / Language / Research Articles
Research Article   |   September 19, 2018
Verbal Agreement Inflection in German Children With Down Syndrome
Journal of Speech, Language, and Hearing Research, September 2018, Vol. 61, 2217-2234. doi:10.1044/2018_JSLHR-L-17-0241
History: Received June 20, 2017 , Revised October 6, 2017 , Accepted April 20, 2018
 
Journal of Speech, Language, and Hearing Research, September 2018, Vol. 61, 2217-2234. doi:10.1044/2018_JSLHR-L-17-0241
History: Received June 20, 2017; Revised October 6, 2017; Accepted April 20, 2018

Purpose The study aims to explore whether finite verbal morphology is affected in children/adolescents with Down syndrome (DS), whether observed deficits in this domain are indicative of a delayed or deviant development, and whether they are due to phonetic/phonological problems or deficits in phonological short-term memory.

Method An elicitation task on subject–verb agreement, a picture-naming task targeting stem-final consonants that also express verbal agreement, a nonword repetition task, and a test on grammar comprehension were conducted with 2 groups of monolingual German children: 32 children/adolescents with DS (chronological age M = 11;01 [years;months]) and a group of 16 typically developing children (chronological age M = 4;00) matched on nonverbal mental age.

Results Analyses reveal that a substantial number of children/adolescents with DS are impaired in marking verbal agreement and fail to reach an acquisition criterion. The production of word-final consonants succeeds, however, when these consonants do not express verbal agreement. Performance with verbal agreement and nonword repetition are related.

Conclusions Data indicate that a substantial number of children/adolescents with DS display a deficit in verbal agreement inflection that cannot be attributed to phonetic/phonological problems. The influence of phonological short-term memory on the acquisition of subject–verb agreement has to be further explored.

Down syndrome (DS) is typically caused by a third exemplar of all or parts of chromosome 21 (trisomy 21) and is the most common neurodevelopmental disorder causing intellectual disability. It occurs in approximately one in 900 live births in Europe (Loane et al., 2013) and is associated with a number of physiological characteristics and intellectual disability that may range from mild to severe. Language is often reported to be a domain of particular difficulties in individuals with DS, considerably lagging behind expectations on the basis of their nonverbal mental development (Abbeduto, Warren, & Connors, 2007; Chapman, Seung, Schwartz, & Kay-Raining Bird, 1998; Fowler, Gelman, & Gleitman, 1994). Within language abilities, morphosyntactic abilities have been found to be particularly affected, leading to the assumption that language production in DS might be characterized by a generalized deficit in grammatical morphology (Chapman et al., 1998). A closer look at the literature, however, reveals that findings with respect to the production of inflectional morphology have not unanimously supported a deficit in this domain. Thus, although some studies have reported such deficits (Bol & Kuiken, 1990; Eadie, Fey, Douglas, & Parsons, 2002; Galeote, Soto, Sebastian, Checa, & Sanchez-Palacios, 2014; Laws & Bishop, 2003; Rutter & Buckley, 1994; Schaner-Wolles, 2004), others have found inflectional morphology to be unimpaired (Christodoulou & Wexler, 2016; Levy & Eilam, 2013; Ring & Clahsen, 2005; Stathopoulou & Clahsen, 2010). Based on an analysis of spontaneous speech recordings, Eadie et al. (2002), for instance, reported a significantly weaker performance of 10 children with DS (mean age = 7 years) with the regular English past tense marker -ed and the third person marker -s in comparison to a group of 10 typically developing (TD) children (aged 3 years) matched for mean length of utterance. A deficit with these two inflectional markers was also observed by Laws and Bishop (2003)  in an elicitation task performed with 14 individuals with DS (aged 10 to 19 years) and 11 TD children (aged 4 to 7 years) matched for nonverbal mental age. Deficits with inflectional morphology in individuals with DS have also been reported for Italian (Fabbretti, Pizzuto, Vicari, & Volterra, 1997), Spanish (Galeote et al., 2014), Dutch (Bol & Kuiken, 1990), and German (Schaner-Wolles, 2004). In contrast, Ring and Clahsen (2005)  stated that the eight English-speaking adolescents with DS (aged 12 to 14 years) who performed several elicitation tasks on past tense and noun plural inflection displayed no signs of a specific deficit with inflectional morphology. In a grammaticality judgment task performed with eight Greek-speaking adolescents (aged 12 to 18 years), Stathopoulou and Clahsen (2010)  reported the perfective past tense formation of existing verbs to be parallel to TD control children aged 5 to 7 years. In an extensive study of case inflections produced by 16 Cypriot Greek adults with DS, Christodoulou and Wexler (2016)  detected ceiling performance with case inflection and reasoned that this inflectional system was intact. Performance at ceiling level for nominal and verbal inflections was also observed by Levy and Eilam (2013)  in a longitudinal and cross-sectional study of nine Hebrew-speaking children with DS, even for the five youngest participants (mean age = 4;06 [years;months]) who had a mean length of utterance between 1 and 1.5 morphemes.
It is difficult to evaluate the different findings with respect to the production of inflectional morphology. A factor that might have contributed to different outcomes is the huge variability in language performance observed in individuals with DS. Whereas some individuals have been found to display performance that is within the normal range for their chronological age or is even considered exceptional (Rondal & Edwards, 1997, pp. 87–98), others have been found to be severely limited with a language production that does not exceed one-word or two-word utterances (Fowler, 1995; Fowler et al., 1994). A huge variability in performance has also been reported with respect to the acquisition of inflectional morphology in children with DS. In a longitudinal study (duration = 2;06 years) on the acquisition of 14 grammatical morphemes in 12 English-speaking children with DS (aged 12 to 38 months), Rutter and Buckley (1994)  found that only the progressive marker -ing was acquired by all children, whereas grammatical morphemes such as plural -s or the regular past tense marking were only acquired by some of the children. As most studies on language abilities in individuals with DS have investigated relatively small numbers of participants (usually between 10 to 16), the variability observed in individuals with DS poses a problem to studies that rely on comparing group means between individuals with DS and TD children. Thus, it might be that some of the divergent findings are due to sampling biases, that is, to accidentally include fewer or more participants in the DS group who happen to have or have not acquired the specific language phenomenon under investigation.
Related to the issue of whether or not the acquisition of inflectional morphology is affected in children/adolescents with DS is the question of whether observed problems are indicative of a delayed or a deviant development. Since the work of Lenneberg, Nichols, and Rosenberger (1964), studies have repeatedly described the language development in individuals with DS as delayed but following the typical developmental course, although at a slower rate and with some variability as to the ultimate attainment (e.g., Fowler et al., 1994; Levy & Eilam, 2013; Rondal & Edwards, 1997, Chapter 4.3). A delay without an indication of a qualitative deviance has also been proposed with respect to the acquisition of morphosyntax. On the basis of parental reports on 92 Spanish-speaking children with DS (aged 1;08 to 5;11), Galeote et al. (2014)  concluded that morphosyntactic development was similar to that observed in a control group of 92 TD children matched for mental age, although proceeding at a slower pace. In contrast, Laws and Bishop (2003)  and Eadie et al. (2002)  stressed the similarity between children with specific language impairment (SLI) and children with DS and took this similarity as indicative of a qualitative developmental deficit in children with DS. Within the neuroconstructivist framework to language development (e.g., Karmiloff-Smith, 2009), the deviation from a typical developmental trajectory that is observed in children with DS due to a later onset and a slower pace of acquisition is considered as suggestive of an atypical deviant development as developmental milestones are not achieved within sensitive periods in development (Levy & Eilam, 2013). Also, developmental delays may only affect some components of language, leading to an overall asynchronous language development that is qualitatively different from typical development (Levy & Eilam, 2013; Schaner-Wolles, 2004).
Related to the matter of whether to characterize language impairments as delayed or deviant is the issue on how to explain observed impairments. Thus, it might be the case that children with DS suffer from a language-specific limitation in their abilities to acquire grammatical morphology. Several studies have, for instance, invoked the extended optional infinitive hypothesis (Wexler, Schütze, & Rice, 1998) to account for deficits with inflectional morphology in children/adolescents with DS (Eadie et al., 2002; O'Neill & Henry, 2002; Ring & Clahsen, 2005). Originally, the extended optional infinitive hypothesis was advocated to capture deficits with tense and agreement inflection in children with SLI (Wexler et al., 1998). According to this hypothesis, an optional infinitive results when the child leaves the tense and/or agreement features of the syntactic head INFL(ection) underspecified. In this case, the verb displays nonfinite inflection and does not undergo verb movement. In children with SLI, the time period where the tense feature may be left underspecified is said to be extended. Research, however, has indicated that the deficit with grammatical morphology observed in children/adolescents with DS is not restricted to tense/agreement inflection, thus challenging the suitability of this hypothesis to capture grammatical deficits in DS (Eadie et al., 2002; O'Neill & Henry, 2002; Ring & Clahsen, 2005).
Impairments with grammatical morphology might also be due to peripheral factors related to hearing and articulation or to differences in language processing. Many children with DS suffer from frequent episodes of otitis media (e.g., Davies, 1996). It thus seems feasible that the mild to moderate conductive hearing loss associated with such episodes might affect the acquisition of grammatical morphemes as these are typically of lower phonetic/phonological saliency compared with open class lexical words, that is, they are unstressed, often realized by consonants, and typically situated in word-final position (Booij, 2000; Bybee, 2000; Dryer, 2005). Studies, however, did not find hearing level to be related to performance with grammatical morphology (e.g., Chapman et al., 1998) or reported inconsistent findings, such as a relationship to one but not other inflectional markers (Laws & Bishop, 2003) in individuals with DS.
Differences in the structure and functioning of articulators, such as tongue and palate, as well as muscle hypotonia, might affect speech production in individuals with DS (Abbeduto et al., 2007; Barnes, Roberts, Mirrett, Sideris, & Misenheimer, 2006; Kent & Vorperian, 2013; Stoel-Gammon, 2001). Indeed, several studies have found phonological development to be affected in children and adolescents with DS (Cleland, Wood, Hardcastle, Wishart, & Timmins, 2010; Roberts et al., 2005; Rupela, Manjula, & Velleman, 2010). Among the most frequent phonological processes observed in children/adolescents with DS are the reduction of consonant clusters and the deletion of word-final consonants (Cleland et al., 2010; Roberts et al., 2005; Rupela et al., 2010). As inflectional markers are often realized as consonants that are affixed at the right edge of a word (e.g., Booij, 2000; Bybee, 2000; Dryer, 2005), they should be particularly susceptible to these phonological processes. Accordingly, it has been suggested that deficits in producing inflectional morphemes in individuals with DS might be due to phonological problems. In two recent studies, Christodoulou (2015)  and Christodoulou and Wexler (2016)  have argued for this view. On the basis of extensive data on language production from 16 adult Cypriot Greek individuals with DS, Christodoulou (2015)  and Christodoulou and Wexler (2016)  suggested that observed problems with case morphology are mainly due to articulatory and phonological limitations that only mimic a deviant case system.
Deficits in processing the language input have also been made responsible for observed problems with inflectional morphology in individuals with DS (Caselli, Monaco, Trasciani, & Vicari, 2008; Yoder, Camarata, Camarata, & Williams, 2006). The prime candidate for such a deficit is phonological short-term memory. Phonological short-term memory, a component of Baddeley's model of working memory (Baddeley, 1986; Baddeley, Gathercole, & Papagno, 1998), is a capacity-limited storage buffer responsible for the temporary storage of phonological representations until the information contained in the material can be processed. Deficits in phonological short-term memory co-occur with impairments affecting inflectional morphology in children with SLI and in children with hearing loss (e.g., Hansson, Sahlén, & Mäki-Torkko, 2007; Norbury, Bishop, & Briscoe, 2001; Penke & Rothweiler, 2018). Indeed the co-occurrence of deficits in phonological short-term memory and language impairments is so consistent in these syndromes that it has been discussed whether deficits in phonological short-term memory underlie the language impairments observed (e.g., Gathercole & Baddeley, 1990; Leonard et al., 2007; Penke & Rothweiler, 2018; Tuller & Delage, 2014), for instance, because the language input cannot be kept in working memory long enough to extract the grammatical information necessary to acquire an inflectional affix. Impairments in phonological short-term memory have also been reported for individuals with DS (e.g., Cairns & Jarrold, 2005; Jarrold, Baddeley, & Phillips, 2002; Laws, 2004), raising the possibility that deficits in phonological short-term memory are connected to impairments with grammatical morphology in individuals with DS as well (Caselli et al., 2008; Yoder et al., 2006). Thus, Yoder et al. (2006)  have suggested that individuals with DS suffer from a generalized slowing in the processing of linguistic input. This slowing is assumed to lead to a “triage process” by which purely grammatical information expressed by inflectional markers might be sacrificed in favor of processing semantic information. As a consequence, children with DS might not have processed enough grammatical morphemes to initialize or complete their acquisition of inflectional markers.
To address the aforementioned issues, the article presents data on the production of subject–verb agreement markers in German-speaking individuals with DS and a control group of TD children matched on nonverbal mental age.
German Subject–Verb Agreement Inflection
German subject–verb agreement inflection marks the morphosyntactic dimensions person (first, second, third) and number (singular and plural) and is completely regular for main/thematic verbs. Three coronal consonants (/s/, /t/, and /n/) and the vowel /ə/ are employed to express the six different specifications with respect to person and number. Table 1 illustrates the agreement paradigm. Colloquial spoken German differs from standard written German with respect to the inflected forms produced. Thus, the first singular marker -e is usually omitted in spoken German where the stem form serves as the first singular form (see Table 1). The second singular affix -st is typically reduced to -s in spoken German and is henceforth notated as -s(t). Also, the first and third plural ending, although written as -en, is usually reduced to /n/ in articulation and is notated as -n in the following.
Table 1. Paradigm of German subject–verb agreement.
Paradigm of German subject–verb agreement.×
Value
Verb form
Verb stem on consonant (example: kauf- “buy”)
Verb stem on vowel (example: bau- “build”)
Person/number Stem-suffix Spoken form Stem-suffix Spoken form
1/singular kauf-e kauf bau-e bau
2/singular kauf-st kaufs bau-st baus
3/singular kauf-t kauft bau-t baut
1/plural kauf-en kau.fn bau-en baun
2/plural kauf-t kauft bau-t baut
3/plural kauf-en kau.fn bau-en baun
Note. Dot indicates syllable boundary.
Note. Dot indicates syllable boundary.×
Table 1. Paradigm of German subject–verb agreement.
Paradigm of German subject–verb agreement.×
Value
Verb form
Verb stem on consonant (example: kauf- “buy”)
Verb stem on vowel (example: bau- “build”)
Person/number Stem-suffix Spoken form Stem-suffix Spoken form
1/singular kauf-e kauf bau-e bau
2/singular kauf-st kaufs bau-st baus
3/singular kauf-t kauft bau-t baut
1/plural kauf-en kau.fn bau-en baun
2/plural kauf-t kauft bau-t baut
3/plural kauf-en kau.fn bau-en baun
Note. Dot indicates syllable boundary.
Note. Dot indicates syllable boundary.×
×
Most verbal stems in German end in one or two consonants. Inflecting a verb with the suffixes -s(t) or -t, hence, leads to an accumulation of word-final consonants (see Table 1). The German syllable structure allows up to three obstruents to follow the syllable's nucleus: one in the coda and two in the coronal appendix outside the rhyme (Grijzenhout & Penke, 2005; Vennemann, 1988). 1   In verb forms suffixed with -s(t) or -t, these suffixes appear in the coda or in the appendix of the verb form's final syllable, depending on the presence of stem-final obstruents in the verb. Thus, for the verb bau- exemplified in Table 1, the inflectional affixes -t and -s(t) would appear in the coda position in colloquial spoken German (e.g., baut or baus). For a verb such as kauf- in Table 1 that ends in a stem-final obstruent (e.g., /f/), these inflectional affixes are situated in the appendix (e.g., kauft or kaufs). Also, in verbs with stem-final consonants (see kauf- in Table 1), producing a verb form suffixed with -n leads to the addition of a reduced syllable, which encompasses the inflectional affix in the syllable's nucleus (kauf + n: /kaʊ.fn̩/). By a process of resyllabification, the stem-final consonant(s) (/f/ in the example) appears as the onset of the reduced syllable.
Aim of the Article
The aim of this article is to provide new evidence on the controversies regarding inflectional deficits in individuals with DS by presenting data on the production of verbal agreement markers obtained from a group of 32 German-speaking children and adolescents with DS and a control group of 16 TD children matched on nonverbal mental age. The comparison with a group of TD children matched for nonverbal mental age allows for differentiating a deficit in language acquisition in children/adolescents with DS from a language performance that is expected given the level of cognitive development achieved. Language performance in individuals with DS that differs quantitatively and/or qualitatively from the performance of mental age–matched control children is indicative of a language impairment because performance in this case cannot be accounted for by the level of cognitive development achieved by these individuals.
Specifically, the article addresses the following questions: (a) Is the production of inflectional markers impaired in children/adolescents with DS? (b) If so, is the impairment indicative of a quantitative delay or a qualitative deviance? (c) How can the impairment be accounted for?
Several factors make German subject–verb agreement a good candidate to investigate whether children and adolescents with DS display impairments in the acquisition of inflectional markers and whether a potential impairment is due to phonetic/phonological problems. For one, deficits in verbal agreement inflection have been repeatedly found in children/adolescents with DS (e.g., Bol & Kuiken, 1990; Eadie et al., 2002; Laws & Bishop, 2003), including German-speaking individuals (Schaner-Wolles, 2004). Also, the acquisition of the German verbal agreement paradigm has been found to be affected in children with SLI and in children with hearing loss (e.g., Clahsen, Bartke, & Göllner 1997; Penke & Rothweiler, 2018), suggesting that this grammatical domain might be particularly vulnerable for developmental language disorders in German-speaking children. Given the observed similarities between English-speaking children with SLI and individuals with DS (Eadie et al., 2002, Laws & Bishop, 2003), one might, hence, assume that verbal agreement inflection is also susceptible to impairments in German-speaking individuals with DS.
Moreover, the acquisition of the verbal agreement paradigm proceeds between 2 and 3 years of age in TD German-speaking children and is usually accomplished by the beginning of the fourth year of life (e.g., Clahsen, Eisenbeiss, & Penke, 1996; Clahsen & Penke, 1992). The acquisition of this inflectional system, thus, proceeds early enough to discover delays or deficits in children/adolescents with DS who will often achieve a nonverbal mental age that is above age 3 years. If individuals with DS with a mental age well above 3 years display problems in marking subject–verb agreement, these problems are indicative of a deficit in acquiring this grammatical domain.
Furthermore, German subject–verb agreement makes use of the phonemes /s/, /t/, and /n/ that are also employed by Greek inflectional systems. For these markers, Christodoulou and Wexler (Christodoulou 2015; Christodoulou & Wexler, 2016) argued that errors are mainly due to phonetic/phonological reasons as errors affect these phonemes independent of whether they serve as inflectional markers or not. To test the assumption that /s/, /t/, and /n/ are omitted independent of grammatical function, the article presents data from a picture-naming task exploring the production of /s/, /t/, and /n/ as stem-final consonants in nouns. If problems in producing verbal agreement markers were phonetically/phonologically determined, problems should also affect the production of stem-final consonants where these phonemes do not carry a morphosyntactic content.
To explore the relation between the acquisition of verbal agreement morphology and phonological short-term memory, all participants performed a nonword repetition task. Nonword repetition tasks have been considered a sensitive tool to test for phonological short-term memory capacities (Gathercole & Alloway, 2006; Gathercole & Baddeley, 1990). If a deficit in phonological short-term memory affects the acquisition of agreement morphology in German individuals with DS, performance in the two tasks should be related.
Moreover, performance with subject–verb agreement is compared to a standardized measure of grammar comprehension that is assumed to provide some indication on the general language level achieved by participants with DS. A performance in subject–verb agreement marking that dissociates from the general language level achieved might indicate an asynchronous development within language components or even a specific vulnerability of inflectional morphology within language.
The article will also explore the extent to which factors such as chronological age, mental age, IQ, gender, frequency and length of episodes of otitis media, socioeconomic status, and amount of speech and language therapy have an impact on the performance with respect to the production of agreement morphology in children and in adolescents with DS.
Method
Participants
Two groups of monolingual German children participated in the study: (a) 32 children and adolescents diagnosed with DS and (b) 16 TD children matched for nonverbal mental age to the participants with DS.
Children and Adolescents With DS
Children and adolescents with DS were recruited through local parent support groups and postings on websites or in journals of organizations concerned with DS. Participants were included in the study if they were monolingual German, if oral language was their primary means of communication, and if they produced at least two-word utterances in spontaneous speech production as evidenced by parent's report and confirmed by a speech recording of the first introductory meeting of the participant and the experimenter. Individuals with additional medical or behavioral conditions (such as epilepsy, psychosis, or anxiety disorder) were excluded from the study, as were individuals with a hearing loss exceeding 25 dB. In total, 32 children and adolescents with DS (13 females, 19 males) aged 4;07 to 19;00 (M = 11;01, SD = 3;06) participated in the study (see Table 2). Parents of the children and adolescents with DS were asked to fill out a questionnaire to provide information on biographical and medical background, including questions on diagnosis of DS, hearing and ear infections, socioeconomic factors, and speech and language therapy. According to parental report, of the 32 participants, 27 have a karyotype with a free trisomy 21; one participant has a mosaic form; and another has been diagnosed with a translocation. For three participants, parents were unaware of the precise diagnosis of the chromosomal abnormality causing DS. One of the children participating was wearing a hearing aid to compensate for a slight hearing loss (22 dB in the better ear). For all other participants, the parents stated that hearing was regularly monitored and unimpaired; also, vision was normal or corrected to normal. All participants lived with their families. They were either enrolled in regular school classes or visited schools for children with special educational needs. All participants had received special intervention, including speech and language therapy, from birth or early childhood.
Table 2. Characteristics of children with Down syndrome (DS) and typically developing (TD) control children.
Characteristics of children with Down syndrome (DS) and typically developing (TD) control children.×
Group N Sex Chronological age (years;months) Nonverbal mental age (years;months) IQ a
DS 32 13 females, 19 males M = 11;01 M = 4;05 M = 59.2
SD = 3;06 SD = 0;11 SD = 13.1
Range = 4;07–19;00 Range = 3;05–6;05 Range = 48–97
TD 16 7 females, 9 males M = 4;00 M = 4;05 M = 106
SD = 0;07 SD = 0;10 SD = 11.2
Range = 3;01–5;00 Range = 3;05–5;10 Range = 86–124
Comparison between groups (independent-samples t test) p < .001 p = .97 p < .001
Note. Nonverbal mental age and IQ values were determined on the basis of the reasoning scale of the Snijders–Oomen Non-Verbal Intelligence Test.
Note. Nonverbal mental age and IQ values were determined on the basis of the reasoning scale of the Snijders–Oomen Non-Verbal Intelligence Test.×
a IQ corrected for Flynn effect.
IQ corrected for Flynn effect.×
Table 2. Characteristics of children with Down syndrome (DS) and typically developing (TD) control children.
Characteristics of children with Down syndrome (DS) and typically developing (TD) control children.×
Group N Sex Chronological age (years;months) Nonverbal mental age (years;months) IQ a
DS 32 13 females, 19 males M = 11;01 M = 4;05 M = 59.2
SD = 3;06 SD = 0;11 SD = 13.1
Range = 4;07–19;00 Range = 3;05–6;05 Range = 48–97
TD 16 7 females, 9 males M = 4;00 M = 4;05 M = 106
SD = 0;07 SD = 0;10 SD = 11.2
Range = 3;01–5;00 Range = 3;05–5;10 Range = 86–124
Comparison between groups (independent-samples t test) p < .001 p = .97 p < .001
Note. Nonverbal mental age and IQ values were determined on the basis of the reasoning scale of the Snijders–Oomen Non-Verbal Intelligence Test.
Note. Nonverbal mental age and IQ values were determined on the basis of the reasoning scale of the Snijders–Oomen Non-Verbal Intelligence Test.×
a IQ corrected for Flynn effect.
IQ corrected for Flynn effect.×
×
Because subject–verb agreement inflection is mastered by the beginning of age 3 years in TD German children (Clahsen et al., 1996; Clahsen & Penke, 1992), we only included children with a nonverbal mental age above 40 months, that is, at a developmental age where subject–verb agreement would have been acquired in TD children. Nonverbal cognitive development was determined by the reasoning subscale of the Snijders–Oomen Non-Verbal Intelligence Test (SON-R; Tellegen, Laros, & Petermann, 2007). The mean nonverbal mental age of the participants with DS was 4;05 (SD = 0;11); their average IQ determined on the basis of the reasoning scale of the SON-R was 59.2 (SD = 13.1).
TD Children
A group of 16 TD monolingual children (seven females, nine males) was investigated as a control group. These children were recruited from local kindergartens. None of the children had a history of hearing or speech/language impairments, and the children displayed no evidence of physical or cognitive impairments. Their vision was normal or corrected to normal. All children demonstrated an age-appropriate language development according to a standardized German screening test (Sprachscreening für das Vorschulalter (SSV); Grimm, 2003). In addition, all children had normal intelligence and achieved a standard score of 85 or better on the reasoning scale of the SON-R intelligence test. The mean IQ for the children of this group was 106 (SD = 11.2). Children of the TD group were matched to the participants in the DS group for nonverbal cognitive development. Thus, we only included children within the nonverbal mental age range of the participants in the DS group. The mean nonverbal mental age of the TD children according to the reasoning scale of the SON-R was 4;05 (SD = 0;10). Their mean chronological age was 4;00 (SD = 0;07; see Table 2).
Statistical comparisons between the two groups of participants, conducted with independent-samples t tests (see Table 2), revealed no difference with respect to nonverbal mental age, t(46) = 0.04, p = .97, d = 0.01. Significant differences between the two participant groups were observed for chronological age, t(34) = 11.22, p < .001, d = 2.46, and for IQ, t(46) = −12.29, p < .001, d = 3.76. These, of course, were expected due to the higher age and greater age range and due to the lower IQ scores of the participants with DS.
Procedure
The research reported in this article is part of a larger research project exploring morphosyntactic skills in children and adolescents with DS. In total, participants attended four testing sessions of approximately 45 min each within 4 to 8 weeks. In this article, data from four tests are provided besides a measure of nonverbal cognition (SON-R): (a) data from an experiment eliciting subject–verb agreement inflection, (b) data from a picture-naming task, (c) a measure of phonological short-term memory, and (d) a standardized test on grammar comprehension (Test for Reception of Grammar - German (TROG-D); Fox, 2011) to provide an index of the general language level achieved by participants with DS. All tests were performed in a familiar setting in a quiet room where no other person except the participant and examiner was present. Typically, IQ testing was the first test performed. The experiment on subject–verb agreement, the picture-naming task, and the TROG-D were administered during the same session, typically within a month from IQ testing. The test on phonological short-term memory was applied last, usually within the same month. The same examiner applied all measures to the same participant after making herself and the situation familiar to the participant. Each experimental session was audiotaped and videotaped. The tapes were used to ensure that all tests were presented the same way and to transcribe participants' utterances for further analysis. The annotation tool ELAN was used to orthographically transcribe all utterances of participants. Pronunciations that deviated from spoken German were transcribed by using the International Phonetic Alphabet. All transcripts were checked against the video files by another experienced researcher. Cases where no agreement as to the produced utterance could be reached were excluded from further analysis.
The research reported in this article was performed according to the guidelines of the Declaration of Helsinki and approved by the Ethics Committee of the Medical Department of the University of Cologne (number of approval 12–033). All participants' parents gave prior written informed consent to their children participating in the study.
Measures
Test on Subject–Verb Agreement Inflection
To elicit inflected verb forms, participants were asked to describe the action shown in 30 short, silently presented video scenes. Actions were performed (a) by a single child, (b) by two children unknown to the participant, or (c) by the experimenter herself. Videos, for instance, show the experimenter writing a letter, a child swimming, or two children playing soccer. In describing these scenes, participants were expected to produce sentences in three target grammatical contexts:
i. Sentences with a third singular verb form (affix -t) for actions performed by a single child (e.g., boy swimming, expected response Der Junge schwimmt; “The boy is swimming.”).
ii. Sentences with a third plural verb form (affix -n) for actions performed by two children (e.g., two children playing soccer, expected response Die Kinder spielen Fussball; “The children are playing soccer.”).
iii. Sentences with a verb inflected for second singular (affix -s(t)) as a reaction to videos showing the experimenter acting (e.g., experimenter writing, expected response Du schreibs(t) einen Brief; “You are writing a letter.”).
For each of these three target contexts, 10 videos were presented in a randomized and previously determined order.
The actions depicted in the videos are describable by frequently occurring thematic verbs (mean lemma frequency: 700, according to the CELEX database; Baayen, Piepenbrock, & van Rijn, 1993) that are part of preschool children's vocabulary (Augst, 1984). The expected verb forms do not require stem changes and are controlled for syllable complexity. In half of the expected verb forms on -t and -s(t), the affix occupies the coda position of the inflected verb form (see Table 1: baut, baus); in the other half, the affix would take up the appendix (see Table 1: kauft, kaufs). In half of the expected verb forms on -n, the affixation adds a reduced syllable (see Table 1: kau.fn); in the other half, -n is added to the stem-final syllable (see Table 1: baun). Note, however, that the participants were free to describe the actions by using other than the target verbs and did so in some cases.
Videos were presented on a computer screen in front of which the participant and the experimenter were sitting side by side. Each video lasted about 10 s. After about 5 s, the experimenter asked, Was passiert da? “What is happening there?” or Was ist da los? “What is going on there?” to elicit the production of an utterance containing a finite verb form. Note that an elliptical nonfinite utterance is not licit as a response to these questions (e.g., Was passiert da? *Schwimmen “What is happening there? *Swimming”). In the relatively few cases (about 7% of the elicited utterances in each subject group) in which the participants responded with a different-than-expected subject (i.e., the woman instead of you, the girl instead of the children or the girl and the boy), the experimenter tried to elicit a sentence containing the expected subject by verbally highlighting the acting persons (“Look, it's me,” “Look there are a girl and a boy”). In case this succeeded, only the target utterance was evaluated.
The test phase was preceded by a short practice phase with three videos (one for each context) to familiarize the participant with the task. During this practice phase, the participant was introduced to the acting characters (“Look, here you can see me/a boy/two children in the picture. Tell me, what's happening?”) and was asked to describe the depicted scene. During familiarization, the experimenter provided positive feedback and the target sentence (“Yes, you're right, the girl is crying.”). During testing, no target sentence was provided.
Transcripts of the participants' utterances were evaluated by calculating the following measures for each individual participant: (a) First, the number of analyzable utterances was determined. Utterances were classified as analyzable if they were clearly related to the video scene and included both a thematic verb and an overt subject so that correctness of the agreement inflection could be evaluated without doubt. A minimum of three analyzable utterances per participant was set as a basis for further analyses. (b) Based on the number of analyzable utterances, the percentage of utterances displaying correct subject–verb agreement was calculated (overall accuracy score). (c) In a further step, the number of analyzable utterances and percentages of correct realizations were determined independently for each of the three tested grammatical contexts (second singular, third singular, and third plural). 2  
Test on the Production of Stem-Final Consonants
To investigate whether problems in producing verbal agreement markers are due to phonetic/phonological difficulties in producing the phonemes expressing these affixes, the ability of the participants to produce the relevant phonemes /s/, /t/, and /n/ in comparable syllabic positions in simplex nouns was examined. In these cases, /s/, /t/, and /n/ do not have morphosyntactic content but are part of the noun stem. Data were collected with a picture-naming task where participants had to name an object presented on a colored picture card. The investigator presented the pictures one after another and asked the participant to name the object depicted. The task was made clear with a practice card for which the target form was provided if necessary.
The test material consisted of pictures of six target nouns ending in /s/, six nouns ending in /t/ (including two words ending in the cluster /st/), and five nouns ending in /n/. The syllable position of the stem-final phonemes /s/, /t/, and /n/ was matched to the expected target verb forms in the subject–verb agreement experiment. Thus, comparable to the expected inflected verb forms on -t and -s(t), in half of the nouns, the critical consonants /s/ and /t/ were in coda position (e.g., Eis “ice,” Hut “hat”), in the other half of the items, /s/ and /t/ took up the appendix (e.g., Keks “cookie,” Stift “pencil”). Comparable to -n-inflected verbs, the stem-final /n/ in nouns either appeared in a reduced syllable (e.g., Ku.chen “cake,” three cases) or in the stem's offset (e.g., Huhn “hen,” two cases). All target nouns are monomorphemic simplex nouns that are part of preschool children's vocabulary (Augst, 1984) and occur frequently in the target language (mean lemma frequency: 211, according to the CELEX database). In total, 18 picture cards (including one practice card) were presented to the participants. Two individuals of the DS group did not finish the experiment. Their data are not included in the analysis.
As the goal of the experiment was to investigate the production of the stem-final consonants /s/, /t/, and /n/, cases where a participant produced a word that was not intended and that was not ending in one of the three critical consonants were excluded from further analysis (e.g., Wolf “wolf” instead of Fuchs “fox”). Based on the transcripts, it was then determined for each participant individually how often the stem-final consonants /s/, /t/, and /n/ were produced, omitted, or substituted by a different consonant. Note that realizations of /s/ as the interdental sibilant /θ/ were counted as correct because this articulation occurs very frequently in the speech production of TD German-speaking children, persisting well into school age (Fox, 2003, p. 63). Absolute numbers were then transferred into percentages.
Nonword Repetition Task
Phonological working memory was assessed by the Nonword-Repetition subtest of the Sprachentwicklungstest für drei- bis fünfjährige Kinder (SETK 3-5; Grimm, 2001). The nonword list of the SETK 3-5 has 18 items that display an increasing complexity with regard to the number of syllables (from two to five syllable words) and with respect to syllable complexity, that is, the number of consonants in onset and offset positions. The test was administered according to the manual. The nonwords were read one at a time to the participants who were instructed to repeat these words as accurately as possible. Following the procedure in the manual, a raw score for each participant was then computed by counting the absolute number of correctly repeated nonwords.
Measure of General Language Level
The TROG-D (Fox, 2011), a German adaption of the Test for Reception of Grammar (Bishop, 2003), was used as a standardized measure of general language level in participants with DS. The Test for Reception of Grammar is a well-established instrument for diagnostic and research purposes and tests the comprehension of a broad array of grammatical structures. Out of a choice of four pictures, the participant is required to identify the one that matches a word or sentence spoken by the experimenter. The test includes 21 blocks of four items each. Each block tests a different grammatical structure, increasing in grammatical complexity from simple content words to complex compound sentences. The test was administered and scored according to the manual. To determine whether individual scores obtained by the participants with DS were at the level expected for their level of cognitive development, individual raw scores were transformed into T-scores on the basis of the nonverbal mental age achieved by each participant with DS. A T-score of 40 or above indicates performance that is appropriate for the mental age achieved by the participant. Three participants with DS did not perform the test.
Data Analysis
Individual scores and percentages obtained by participants of the two subject groups were compared by parametric statistical tools. When Levene's test indicated unequal variances between subject groups, degrees of freedom were adjusted. For correlations between variables, the nonparametric Spearman rho test was chosen if Kolmogorov–Smirnov tests indicated that variables were not normally distributed. The level of statistical significance was set at .05.
A common finding of research on language performance in individuals with DS is that there is a huge interindividual variance. To take this interindividual variance into account, in addition to the statistical analyses comparing group performance, a criterion approach was employed to determine for each individual participant whether the system of subject–verb agreement marking was acquired. The criterion states that the system of subject–verb agreement has been acquired when accuracy scores surpass 90%. This criterion goes back to the seminal work of Brown (1973) . It has been applied in previous research on the acquisition of grammatical morphemes in participants with DS (Fowler et al., 1994; Rutter & Buckley, 1994). Moreover, this criterion has been proven valuable in determining whether the German verbal agreement system has been acquired in individual children with or without developmental language deficits (Clahsen et al., 1996; Clahsen & Penke, 1992; Penke & Rothweiler, 2018; Rothweiler, Chilla, & Clahsen, 2012).
Results
The results obtained by participants of the two subject groups in the four different tasks are summarized in Table 3.
Table 3. Experimental results—group comparisons.
Experimental results—group comparisons.×
Measure Group
p value
DS TD
Subject–verb agreement a
 Number of participants 27 16
 Number of analyzable responses 615 435
 Number of analyzable responses per participant M = 22.8, SD = 8.4 M = 27.2, SD = 2.54 p = .02
 Correct agreement in % M = 78.6%, SD = 22.1 M = 99.3%, SD = 1.4 p < .001
Picture naming
 Number of participants 30 16
 Number of analyzable responses 484 259
 Number of analyzable responses per participant M = 16.1, SD = 1.5 M = 16.2, SD = 1 p = .9
 Correct production of final consonant in % M = 91.9%, SD = 15 M = 98.8%, SD = 2.6 p = .02
Nonword repetition
 Number of participants 32 16
 Raw score M = 4.3, SD = 3.9 M = 7.3, SD = 2.1 p = .001
TROG-D
 Number of participants 29
T-score M = 46.3, SD = 9.2
Note. DS = Down syndrome; TD = typically developing; TROG-D = Test for Reception of Grammar - German.
Note. DS = Down syndrome; TD = typically developing; TROG-D = Test for Reception of Grammar - German.×
a Data of five participants with DS who produced fewer than three analyzable utterances each not included.
Data of five participants with DS who produced fewer than three analyzable utterances each not included.×
Table 3. Experimental results—group comparisons.
Experimental results—group comparisons.×
Measure Group
p value
DS TD
Subject–verb agreement a
 Number of participants 27 16
 Number of analyzable responses 615 435
 Number of analyzable responses per participant M = 22.8, SD = 8.4 M = 27.2, SD = 2.54 p = .02
 Correct agreement in % M = 78.6%, SD = 22.1 M = 99.3%, SD = 1.4 p < .001
Picture naming
 Number of participants 30 16
 Number of analyzable responses 484 259
 Number of analyzable responses per participant M = 16.1, SD = 1.5 M = 16.2, SD = 1 p = .9
 Correct production of final consonant in % M = 91.9%, SD = 15 M = 98.8%, SD = 2.6 p = .02
Nonword repetition
 Number of participants 32 16
 Raw score M = 4.3, SD = 3.9 M = 7.3, SD = 2.1 p = .001
TROG-D
 Number of participants 29
T-score M = 46.3, SD = 9.2
Note. DS = Down syndrome; TD = typically developing; TROG-D = Test for Reception of Grammar - German.
Note. DS = Down syndrome; TD = typically developing; TROG-D = Test for Reception of Grammar - German.×
a Data of five participants with DS who produced fewer than three analyzable utterances each not included.
Data of five participants with DS who produced fewer than three analyzable utterances each not included.×
×
Subject–Verb Agreement
Analyzable and Unanalyzable Utterances
The participants in the DS group produced a total of 618 utterances that were analyzable with respect to subject–verb agreement according to the criteria mentioned above (64.4% of 960 contexts). Five of the participants in the DS group, however, produced no (three participants) or less than three analyzable utterances in the experiment. Based on the criterion that at least three utterances should be produced per participant to validate further analyses on the accuracy of subject–verb agreement marking, the data of these five participants (in total three utterances) were excluded from further analyses. The data obtained for the remaining group of 27 participants with DS are presented in Table 3. With the five participants removed, each participant produced on average 23 (SD = 8) analyzable utterances for the 30 contexts tested. This was significantly different from the group of TD children who produced 27 analyzable utterances on average (SD = 3) per participant, t(33.25) = −2.54, p = .02, d = 0.64, totaling up to 435 analyzable responses out of 480 contexts (90.6%). With respect to the three tested grammatical contexts, the mean number of analyzable utterances per participant containing a second singular subject was seven in the group of participants with DS and eight in the group of TD children. The corresponding figures for third singular subjects were nine in the DS group versus 10 in the TD group and seven (DS) versus nine (TD) for utterances with a third plural subject. Comparisons of the number of analyzable utterances in the three tested grammatical contexts yielded no significant differences between the two subject groups (p > .05 each), indicating that the experiment was suitable for eliciting utterances in the three grammatical contexts in both subject groups.
Overall Accuracy Scores
In total, 524 out of 615 analyzable responses produced by the participants of the DS group displayed cor-rect subject–verb agreement. The mean accuracy score for participants with DS was 78.6% (SD = 22.1). Children in the TD group obtained a mean accuracy score of 99.3% (SD = 1.4). Only three of the 435 analyzable utterances produced by these children displayed incorrect subject–verb agreement. The difference in accuracy scores between the two subject groups was highly significant, t(26.38) = −4.84, p < .001, d = 1.17.
To determine whether the accuracy scores for subject–verb agreement were related to age, nonverbal cognitive development, or IQ, correlational analyses were conducted for the 27 participants of the DS group. As accuracy scores were not normally distributed (p = .03), correlation coefficients were computed with Spearman rho tests. No significant effects were found between overall accuracy scores for subject–verb agreement on the one hand and chronological age in months, r(25) = .08, p = .68, nonverbal mental age in months, r(25) = −.02, p = .94, or IQ, r(25) = −.03, p = .88, on the other hand. For TD children, no correlational analyses were calculated as these children performed at ceiling level with respect to subject–verb agreement.
To take individual variation into account, an acquisition criterion (over 90% correct marking) was applied to determine for each individual participant whether the system of subject–verb agreement had been mastered. The individual accuracy scores are displayed in Figure 1. According to this criterion, all TD children had acquired subject–verb agreement, with individual accuracy scores between 96% and 100%. In the group of participants with DS, 11 of the 27 participants surpassed the 90% criterion and achieved accuracy scores between 93% and 100% (M = 96.9, SD = 1.9). This group will henceforth be referred to as DS + AGR(AGR short for subject-verb agreement). The other 16 participants did not reach the acquisition criterion (henceforth, DS−AGR). Their mean accuracy score was at only 66.1% (SD = 20.9), ranging between 16.7% and 89%.
Figure 1.

Individual accuracy scores for subject–verb agreement for participants in the Down syndrome (DS) and the typically developing (TD) group. (Each symbol marks the score of one participant.)

 Individual accuracy scores for subject–verb agreement for participants in the Down syndrome (DS) and the typically developing (TD) group. (Each symbol marks the score of one participant.)
Figure 1.

Individual accuracy scores for subject–verb agreement for participants in the Down syndrome (DS) and the typically developing (TD) group. (Each symbol marks the score of one participant.)

×
To derive predictive factors distinguishing participants with DS who had or had not acquired subject–verb agreement, comparisons (t and chi-square tests) were calculated to evaluate the influence of individual factors, such as gender, age (chronological and mental), IQ, frequency of middle ear infections in the past, amount of speech and language therapy received in relation to chronological age, or socioeconomic background (measured by level of education of mother or father). However, only the amount of speech and language therapy received over the life span differentiated participants of the DS + AGR subgroup from those of the DS−AGR subgroup, with participants of the latter subgroup receiving significantly more hours of therapy in relation to their chronological age than participants of the DS + AGR subgroup, t(15) = −4.28, p = .001, d = 2.1. Table 4 summarizes the results.
Table 4. Comparison of individual factors in participants with Down syndrome who have (DS + AGR) or have not (DS−AGR) acquired subject–verb agreement.
Comparison of individual factors in participants with Down syndrome who have (DS + AGR) or have not (DS−AGR) acquired subject–verb agreement.×
Measure DS + AGR DS−AGR p value
Number of participants 11 16
Number of males and females 7 males, 4 females 8 males, 8 females p = .7
Chronological age (years;months) M = 11;03, SD = 4;01 M = 11;01, SD = 3;06 p = .92
Mental age (years;months) M = 4;06, SD = 1;01 M = 4;07, SD = 0;10 p = .84
IQ M = 61, SD = 12.8 M = 61, SD 14.2 p = .96
Number of participants with frequent (> 5 per year) or persistent (> 3 months) infections of the middle ear in the past 3 4 p = 1
Amount of speech and language therapy in relation to chronological age a M = 0.58, SD = 0.13 M = 0.8, SD = 0.08 p = .001
Education score b of mother M = 6.6, SD = 1.4 M = 5.3, SD = 2 p = .07
Education score b of father M = 6.3, SD = 1.5 M = 5.6, SD = 2.1 p = .35
Production score for stem-final consonants M = 98.4%, SD = 2.8 M = 94.8%, SD = 8.4 p = .13
Score for nonword repetition M = 6.8, SD = 2.6 M = 3.9, SD = 4 p = .04
Number of participants with TROG-D T-score > 40 9 of 11 11 of 15 p = .67
Note. AGR = subject-verb agreement; TROG-D = Test for Reception of Grammar - German.
Note. AGR = subject-verb agreement; TROG-D = Test for Reception of Grammar - German.×
a Months with speech and language therapy divided by chronological age in months).
Months with speech and language therapy divided by chronological age in months).×
b Education of mother and father were measured on the 9 level scale (ranging from 0 = early childhood education to 8 = doctoral or equivalent level) of the International Standard Classification of Education (United Nations Educational, Scientific and Cultural Organization, 2012).
Education of mother and father were measured on the 9 level scale (ranging from 0 = early childhood education to 8 = doctoral or equivalent level) of the International Standard Classification of Education (United Nations Educational, Scientific and Cultural Organization, 2012).×
Table 4. Comparison of individual factors in participants with Down syndrome who have (DS + AGR) or have not (DS−AGR) acquired subject–verb agreement.
Comparison of individual factors in participants with Down syndrome who have (DS + AGR) or have not (DS−AGR) acquired subject–verb agreement.×
Measure DS + AGR DS−AGR p value
Number of participants 11 16
Number of males and females 7 males, 4 females 8 males, 8 females p = .7
Chronological age (years;months) M = 11;03, SD = 4;01 M = 11;01, SD = 3;06 p = .92
Mental age (years;months) M = 4;06, SD = 1;01 M = 4;07, SD = 0;10 p = .84
IQ M = 61, SD = 12.8 M = 61, SD 14.2 p = .96
Number of participants with frequent (> 5 per year) or persistent (> 3 months) infections of the middle ear in the past 3 4 p = 1
Amount of speech and language therapy in relation to chronological age a M = 0.58, SD = 0.13 M = 0.8, SD = 0.08 p = .001
Education score b of mother M = 6.6, SD = 1.4 M = 5.3, SD = 2 p = .07
Education score b of father M = 6.3, SD = 1.5 M = 5.6, SD = 2.1 p = .35
Production score for stem-final consonants M = 98.4%, SD = 2.8 M = 94.8%, SD = 8.4 p = .13
Score for nonword repetition M = 6.8, SD = 2.6 M = 3.9, SD = 4 p = .04
Number of participants with TROG-D T-score > 40 9 of 11 11 of 15 p = .67
Note. AGR = subject-verb agreement; TROG-D = Test for Reception of Grammar - German.
Note. AGR = subject-verb agreement; TROG-D = Test for Reception of Grammar - German.×
a Months with speech and language therapy divided by chronological age in months).
Months with speech and language therapy divided by chronological age in months).×
b Education of mother and father were measured on the 9 level scale (ranging from 0 = early childhood education to 8 = doctoral or equivalent level) of the International Standard Classification of Education (United Nations Educational, Scientific and Cultural Organization, 2012).
Education of mother and father were measured on the 9 level scale (ranging from 0 = early childhood education to 8 = doctoral or equivalent level) of the International Standard Classification of Education (United Nations Educational, Scientific and Cultural Organization, 2012).×
×
Accuracy Scores per Context
A comparison of the mean accuracy scores for the three tested contexts (second singular, third singular, and third plural) obtained by the TD children, the participants of the DS + AGR subgroup, and the participants of the DS−AGR subgroup confirmed a significant effect for the factor subject group,F(2, 35) = 27.7, p < .001, ηp2 = .61, with TD subjects performing as DS + AGR subjects (p = 1), and both groups outperforming DS−AGR subjects (p < .001 each). In contrast, the factor grammatical context yielded no significant effect, F(1.7, 59.4) = .229, p = .76, ηp2 = .007, indicating that accuracy scores for the three tested grammatical contexts were similar within each subject group. The group of TD children and the 11 participants of the DS + AGR subgroup obtained mean accuracy scores well above 90% for each of the tested grammatical contexts. Also, the mean accuracy scores obtained by the 16 participants of the DS−AGR subgroup, although significantly lower compared with the accuracy scores of the TD children and the DS + AGR group, gave no indication that difficulties with subject–verb agreement were restricted to particular grammatical contexts. A repeated-measures analysis of variance on the accuracy scores obtained by these 16 participants in the three grammatical contexts tested yielded no significant difference, F(2, 22) = .32, p = .73, ηp2 = .03, between the mean accuracy scores for the tested second singular (M = 69%, SD = 37.2), third singular (M = 72.5%, SD = 22.7), and third plural (M = 65.9%, SD 35.9) contexts.
Error Analysis
The 16 participants of the DS−AGR subgroup produced a total of 81 analyzable utterances that displayed no agreement between subject and verb. The most frequent error type (28 cases, 34.6% of errors) was the incorrect substitution with the ending -t in grammatical contexts for the markings -n (17 cases) and -s(t) (11 cases). Substitutions with the ending -n made up 27.2% of the errors (22 cases). In 15 of these 22 cases (68%), the resulting -n-inflected form was clearly nonfinite, as indicated by the clause-final position of the verb. Thus, in these cases, participants produced an infinitive instead of the required finite verb form. Substitutions with -s(t) occurred in three cases, as did substitutions with the affix -e or with other phonemes that do not serve as inflectional markers in German. In 25 cases, that is, in 30.9% of the errors, the required affix was omitted.
Production of Stem-Final Consonants
Neither the TD children nor the participants with DS had any problems in performing the picture-naming task. The relevant data are presented in Table 3. In both subject groups, on average, 16 reactions to the 17 presented target pictures were analyzable with respect to the production of the critical stem-final consonants /s/, /t/, and /n/. In total, the 30 participants of the DS group produced 484 analyzable nouns. Analyzable productions for the 16 TD children amounted to 259 nouns. Children of the TD group produced 98.8% of the stem-final critical consonants correctly (SD = 2.6). There were only three cases in which the stem-final consonant was omitted. In all three cases, the item Wurst “sausage” was articulated as /vʊɐs/ without the stem-final /t/. Participants of the DS group produced on average 91.9% of the stem-final critical consonants correctly (SD = 15). Although slight, the performance difference of 6.9% between the two subject groups was significant, t(32.06) = −2.47, p = .02, d = 0.57.
Participants of the DS group displayed a significant difference with respect to the three stem-final consonants tested, F(2, 58) = 5.19, p = .01, ηp2 = .15. They performed equally well with the stem-final consonants /s/ and /n/ for which they achieved average production scores of 95.3% each. With a mean score of 84%, production of stem-final /t/ was significantly more affected in comparison. In total, the participants of the DS group omitted stem-final consonants in 27 cases. With 20 cases, omissions of the stem-final consonant /t/ made up the majority of these errors (74.1%). Omissions of stem-final /s/ (two cases) and /n/ (five cases) were comparably rare, as were substitutions of the stem-final consonant that occurred in six cases and only affected the consonants /s/ and /t/. In these six substitutions, the consonant /s/ was substituted by the fricatives /ç/ or /ʃ/ (three cases), and the consonant /t/ was replaced by /s/, /ʃ/, and /k/.
To investigate whether production scores for stem-final consonants were related to chronological age, mental age, or IQ, correlations were calculated. Because the production scores achieved by the DS group in the picture-naming task were not distributed normally (p < .001), Spearman rho correlations were computed. No significant correlations were observed for the production scores of stem-final consonants and chronological age, r(28) = −.16, p = .39; mental age, r(28) = −.07, p = .71; or IQ, r(28) = .08, p = .69. As the performance of the TD children was at ceiling, no corresponding analyses were calculated.
Relationship Between the Production of Word-Final Consonants and Subject–Verb Agreement
To evaluate the proposal that, in individuals with DS, problems with subject–verb agreement might be due to phonetic/phonological difficulties in producing the critical consonants /s/, /t/, and /n/ in word-final position, accuracy scores achieved in the subject–verb agreement task were correlated to the production scores obtained in the picture-naming task. The correlation, however, was not significant, r(25) = .23, p = .24. In addition, performance rates with respect to stem-final consonant production were compared for participants of the DS + AGR and the DS−AGR subgroups. In the picture-naming task, the 11 participants of the DS + AGR subgroup produced on average 98.4% of the stem-final critical consonants /s/, /t/, and /n/ correctly (SD = 2.8). The corresponding score for the 16 participants of the DS−AGR subgroup was 94.8% (SD = 8.4). This difference was not significant, t(19.5) = 1.58, p = .13, d = 0.53 (see Table 4). Both statistical results indicate that differences with respect to the accuracy scores in subject–verb agreement were not related to production scores of these same consonants in stem-final position in nouns. Specifically, production scores of the stem-final critical consonants were high, independent of whether or not difficulties in subject–verb agreement occurred. This was especially clear in the data of 10 participants with DS who performed flawlessly with respect to the stem-final consonants /s/, /t/, and /n/ but had not acquired subject–verb agreement, although the same consonants were involved in the same word-final and syllable-structural position in inflected verbs and in stem-final consonants of nouns.
Nonword Repetition
In the nonword repetition test, children in the TD group achieved an average score of 7.3 (SD = 2.1). This score was significantly higher than the average score of 4.3 (SD = 3.9) obtained by participants with DS, t(45.69) = −3.51, p = .001, d = 0.89.
Correlational analyses were conducted separately for both groups of participants to investigate whether nonword repetition scores were influenced by chronological or mental age or nonverbal IQ. As the nonword repetition scores obtained by the participants with DS were not normally distributed (p = .03), Spearman rho correlations were computed for this subject group. For the group of TD children, only mental age was correlated to scores in nonword repetition, r(14) = .56, p = .03. For the participants in the DS group, scores in nonword repetition were significantly correlated with nonverbal IQ, r(30) = .36, p = .04, but not with chronological age, r(30) = .01, p = .97, or mental age, r(30) = .29, p = .11.
Relationship Between Nonword Repetition and Results on Subject–Verb Agreement
Additional correlational analyses were run for the DS group to relate the scores obtained in nonword repetition to accuracy scores in subject–verb agreement. Because nonverbal IQ was correlated to scores in nonword repetition, the influence of this variable was partialled out. The analysis yielded a significant relation between nonword repetition scores and accuracy scores in subject–verb agreement, r(24) = .56, p = .003. The relation between nonword repetition and verbal agreement scores was supported by a comparison of the nonword repetition scores obtained by participants of the DS + AGR and the DS−AGR subgroups (see Table 4). With a mean nonword repetition score of 3.9 (SD = 4), the 16 individuals of the DS−AGR subgroup scored significantly below the mean nonword repetition score of 6.8 (SD = 2.6) achieved by the 11 participants of the DS + AGR subgroup, t(25) = 2.15, p = .04, d = 0.84.
Measure of General Language Level
In relation to their nonverbal mental age, the participants with DS obtained a mean T-score of 46.3 (SD = 9.2) in the TROG-D. Of the 29 subjects who participated in this test, only six achieved a T-score that was below the critical value of 40, indicating that grammar comprehension was below the level expected for the nonverbal cognitive development attained by these participants. The other 23 subjects reached T-scores between 41 and 65, indicating a level of grammar comprehension appropriate (or even better) for the level of nonverbal cognitive development achieved. 3  
Relationship Between Performance in the TROG-D and Results on Subject–Verb Agreement
The TROG-D provides some indication on the general language level achieved by the participants with DS. Performance with subject–verb agreement can be evaluated against this language baseline to determine whether performance in the two tests concurs or whether performance in agreement inflection dissociates from the measure of general language level provided by the TROG-D.
To determine whether the accuracy scores for subject–verb agreement were related to the T-scores obtained in the TROG-D, a correlation was computed using the Spearman rho test as accuracy scores in the agreement task were not normally distributed (p = .03). This analysis did not reach significance, r(26) = .39, p = .052, indicating that there is no strong relation between performance in the two tests.
The T-score of the TROG-D allows for classifying performance in grammar comprehension as appropriate or below the level expected by the nonverbal cognitive development achieved. The TD children's performance on agreement inflection illustrates that subject–verb agreement should have been acquired by the nonverbal mental age achieved by the participants with DS. Performance can thus be characterized according to the following two dimensions: (a) Performance in the TROG-D is or is not appropriate for the level of nonverbal cognitive development achieved, and (b) subject–verb agreement is or is not acquired. Given this matrix, the acquisition of subject–verb agreement should be associated with a T-score of above 40 in the TROG-D, indicating a mental age–appropriate language level in both tests. If, however, subject–verb agreement dissociates from general language level as measured by the TROG-D, the acquisition of subject–verb agreement may co-occur with a T-score below 40, or conversely, a T-score of above 40 may co-occur with a deficit in verbal agreement inflection.
Of the 11 participants of the DS + AGR group who have acquired subject–verb agreement, nine obtained a T-score above 40 in the TROG-D, and two scored below 40. Likewise, of the 15 participants of the DS−AGR subgroup who participated in the TROG-D, only four scored below 40, whereas 11 achieved a T-score above 40. These data indicate that the performance in subject–verb agreement and the performance in the TROG-D dissociate, because T-scores indicate mental age–appropriate general language levels independent of the presence of deficits in subject–verb agreement. Specifically, the finding that 11 subjects displayed deficits with subject–verb agreement despite a mental age–appropriate language level in the TROG-D suggests that performance in subject–verb agreement is weaker compared with the general language level.
Discussion
Individuals With DS Do Not Form a Homogeneous Group
The data on subject–verb agreement marking in German-speaking children and adolescents with DS revealed that individuals with DS did not form a homogeneous group but displayed a large variability in performance. Among the participants with DS tested, a subgroup of 11 individuals had successfully acquired subject–verb agreement and performed within the range of TD children of the same nonverbal mental age. In contrast, a subgroup of 16 individuals did not reach the acquisition criterion and performed significantly below mental age expectations. This finding conforms to previous observations in the literature that have also stated a large variability with respect to language performance in individuals with DS (e.g., Fowler et al., 1994). Specifically, the finding of large interindividual differences with respect to the acquisition of grammatical morphology is supported by Rutter and Buckley (1994)  who observed that the progressive marker -ing was the only marker to be acquired by all 12 English-speaking children with DS investigated in their study, whereas grammatical morphemes, such as plural -s or regular past tense -ed, were only acquired by some of the children.
The huge variability observed in individuals with DS cautions against the sole reliance on group means in exploring language capacities of individuals with DS because outcomes are dependent on the somewhat arbitrary factor of how many individuals in the studied DS group happen to have or have not acquired the specific language phenomenon under investigation. As most studies on language abilities in individuals with DS have investigated relatively small numbers of participants (usually between 10 to 16), the danger of underestimating or overestimating language abilities in a substantial number of affected individuals based on group data might be especially pronounced. This might also have contributed to the divergent findings that have been reported in the field with respect to deficits in grammatical morphology in individuals with DS, with some studies reporting such deficits (e.g., Caselli et al., 2008; Eadie et al., 2002; Laws & Bishop, 2003), whereas others have found inflectional morphology to be unimpaired (e.g., Christodoulou & Wexler, 2016; Levy & Eilam, 2013; Ring & Clahsen, 2005, Stathopoulou & Clahsen, 2010). The variability observed here within a group of individuals with DS suggests that a criterion-based approach that serves to categorize individuals with respect to the language phenomenon under investigation provides an important addendum to research on language abilities in individuals with DS over and above reporting group means and group statistics.
For a Subgroup of Individuals With DS the Acquisition of Subject–Verb Agreement Is Impaired
Although a substantial subgroup of individuals with DS have mastered the system of German subject–verb agreement, 16 of the participants tested here achieved accuracy scores for verbal agreement that did not surpass the acquisition criterion of 90% correct markings that was mastered by all children of the TD group. As both participant groups, TD and DS participants, were matched on nonverbal cognitive development, it can be concluded that, in these 16 participants of the DS group, the acquisition of verbal agreement markings was delayed in comparison to their cognitive development, indicating a language impairment in this domain. The finding that individuals with DS can be impaired with respect to inflectional morphology conforms to other studies that have observed similar deficits in individuals with DS. Thus, Rutter and Buckley (1994), Eadie et al. (2002), and Laws and Bishop (2003)  all observed deficits with regular past tense inflection and the third singular marker -s in English-speaking individuals with DS. Deficits with agreement inflection in individuals with DS have also been reported for Dutch (Bol & Kuiken, 1990) and, specifically, for German (Schaner-Wolles, 2004). The data of Schaner-Wolles (2004, Figure 2) who tested a large group of 40 German-speaking individuals with DS (mental age 2;05 to 4;11) with a sentence-repetition task indicate a mean error rate for agreement inflection of over 20%, a rate that is considerably higher than the error rate of less than 5% achieved by 20 TD children aged 2;03 to 3;11. In summary, the data presented here and in these previous studies provide strong evidence that verbal agreement inflection is impaired in a substantial number of individuals with DS.
Subject–verb agreement has been found to be affected in various types of developmental language disorders in German, including SLI or hearing loss (Clahsen et al., 1997; Penke & Rothweiler, 2018), marking this system as a vulnerable domain where language deficits strike independent of specific disorder syndromes. The aim of future research should be to investigate whether performance patterns specific to DS can be observed within this vulnerable domain. Such research would provide an important step in identifying the underlying causes of the observed deficit in affected individuals with DS, thus providing insights on the issue which individuals with DS are likely to experience impairments in verbal agreement marking and on how to target this deficit in therapeutic interventions.
Agreement Deficit Cannot Be Reduced to Phonetic/Phonological Problems
On the basis of data from 16 adult Cypriot Greek individuals with DS, Christodoulou and Wexler (Christodoulou, 2015; Christodoulou & Wexler, 2016) suggested that problems with inflectional morphology are due to phonetic/phonological difficulties affecting the articulation of consonants that act as inflectional markers, but that there is no evidence for morphosyntactic impairments affecting inflection. German verbal agreement inflection is particularly suited to investigate this proposal because it uses the same consonants /s/, /t/, and /n/ as inflectional markers as the system of Cypriot Greek case inflection investigated by Christodoulou and Wexler. However, the findings reported here differ substantially from the observations made for Cypriot Greek case inflection. Christodoulou and Wexler (Christodoulou, 2015; Christodoulou & Wexler, 2016) reported that /s/ was the most affected consonant and was frequently omitted in word-final position independently of whether it was a case marker or a stem-final consonant. In contrast, omissions of stem-final /s/ were only rarely observed in German-speaking individuals with DS, who rather omitted stem-final /t/. In contrast to the findings of Christodoulou and Wexler, in the German data, errors with the critical consonants were not independent of whether or not these consonants served as inflectional markers. Thus, while /t/ was frequently omitted in stem-final position in the picture-naming task, it was overapplied in the agreement task where incorrect substitutions with the affix -t in contexts for the affixes -n or -s(t) were the most frequent error. Because the verb forms expected in the agreement task and the nouns tested in the picture-naming task were matched for the syllable position of the critical final consonants /s/, /t/, and /n/, the relatively frequent omission of stem-final /t/ in the picture-naming task cannot be due to phonetic/phonological problems—after all, participants with DS were well able to produce this consonant in the very same structural positions as a verbal affix. Finally, statistical analyses did not reveal a relationship between the accuracy scores in verbal agreement and the production scores of the critical stem-final consonants in picture naming. In particular, individuals with DS who displayed problems with the verbal agreement system had no significantly lower production scores of stem-final consonants than individuals who displayed no deficits in agreement marking. In sum, these findings provide evidence against the assumption that the problems observed in the 16 German-speaking individuals of the DS−AGR subgroup might be due to phonetic/phonological problems affecting the articulation of the relevant inflectional markers. Rather, the observation that all three grammatical contexts tested were affected to a similar extent suggests that the morphosyntactic features expressed by the verbal agreement markers have not been acquired by the individuals of the DS−AGR subgroup.
A number of factors might account for the different findings obtained here and reported by Christodoulou and Wexler (Christodoulou, 2015; Christodoulou & Wexler, 2016). Christodoulou and Wexler investigated adult individuals with DS and were not targeting language development. Thus, it might be that their adult participants had already acquired the inflectional case system investigated. Also, case inflection might be differently affected in individuals with DS compared with verbal agreement inflection. Finally, the composition of the subject group studied by Christodoulou and Wexler might have included a larger proportion of participants with DS not experiencing language problems, affecting the group data presented in their work. Note also, that findings reported by Eadie et al. (2002)  support the assumption that inflectional deficits are independent from phonetic/phonological problems. In their study, participants with DS passed a phonological screening to assure that they were well able to articulate word-final /s/, /z/, /t/, and /d/. Nevertheless these participants displayed impaired performance in producing English third singular and regular past tense marking, expressed by the very same consonants. More research is needed to disentangle phonetic/phonological and morphosyntactic causes of inflectional deficits in individuals with DS across languages and inflectional systems.
Delay or Deviance?
A central question in research on language impairments in individuals with DS is whether observed problems are indicative of a delayed or a deviant development. If language acquisition were delayed in children with DS, it would follow the course observed in TD children despite a later onset and protracted development. A deviant development, in contrast, would be indicated if the language development in individuals with DS were qualitatively different in comparison to TD children. The critical criterion to distinguish a delayed development from a deviant development is whether or not the observed language behavior corresponds to an earlier acquisition stage of TD children. If it does, the observed difference is categorized as a developmental delay. If the observed behavior does, however, not correspond to an earlier acquisition stage observed in typical development, it is categorized as deviant. Given this criterion, two observations in the data of the individuals of the DS−AGR subgroup suggest a deviant development rather than a delay: (a) the dissociation of performance in the TROG-D and performance in the verbal agreement task and (b) the type of errors committed in subject–verb agreement.
Concerning (a) above, a comparison of the T-scores obtained in the TROG-D and the performance in the agreement task revealed that performance in the two tests dissociated. Although performance in the TROG-D was appropriate for the nonverbal mental age achieved by 11 of the 16 subjects of the DS−AGR subgroup, their accuracy scores in subject–verb agreement were below mental age expectations. This observation suggests that language development in these individuals is not delayed overall but that development is asynchronous, affecting subject–verb agreement more than grammar comprehension measured by the TROG-D. Such asynchronies are indicative of a deviant development that differs from the typical developmental course (e.g., Levy & Eilam, 2013).
Second, the frequent substitutions with the marker -t in contexts for the plural marker -n and the second singular marker -s(t) detected in the verbal agreement experiment constitute an error type rarely observed in typical development. The acquisition of the verbal agreement paradigm proceeds between 2 and 3 years of age in TD German-speaking children (e.g., Clahsen et al., 1996; Clahsen & Penke, 1992). During this time, two types of errors are frequent: substitutions with the marker -n and omissions of inflectional markers (e.g., Clahsen & Penke, 1992). In contrast, incorrect applications of the affix -t are extremely rare in the data of TD children who have not yet acquired the system of subject–verb agreement. Thus, an analysis of the longitudinal data of the German child Simone by Clahsen and Penke (1992)  yields that over a 9-month period before Simone acquired subject–verb agreement at age 2;04, she incorrectly applied -t in only 17 out of 637 -t inflected verb forms (i.e., 2.7%). In contrast, while omissions and substitution errors with -n did also appear in the data on individuals with DS reported here, the most frequent error type in the data was incorrect -t marking. This highly unusual error type occurred in the data of 10 of the 16 individuals of the DS−AGR subgroup. In total, of the 93 -t-markings produced by these 10 participants, 28 were incorrect (30.1%). What causes this unusual error is unclear at present. Thus, it is conceivable that input frequency plays a role, leading to substitutions with the marking that appears most frequently in the input. However, testing this idea requires data on child-directed speech that are not available for the participants with DS tested here. Also, the -n ending marks first and third plural and the infinitive. It is, thus, likely to appear in the input more often than the ending -t, suggesting that -n substitutions should occur more often than substitutions with -t, contrary to the data. Although this issue has to be left to future research, the frequent occurrence of the unusual -t-substitutions in individuals with DS suggests that the observed problem in acquiring subject–verb agreement might be due to a deviant development rather than to a delayed development. Taken together, observations (a) and (b) would classify the language development of 15 of the 16 individuals of the DS−AGR subgroup as deviant.
The high proportion of substitutions with the third singular marker -t is also in contrast to the suggestion that children/adolescents with DS display an extended optional infinitive stage (see Wexler et al., 1998). In contrast to the extended optional infinitive hypothesis, which states that errors in verbal inflection result from the substitution of finite by nonfinite forms, in these errors, a finite agreement marker -t replaces the agreement markers -s(t) and -n. This observation confirms earlier evaluations by O'Neill and Henry (2002), Eadie et al. (2002), and Ring and Clahsen (2005)  that also questioned the applicability of this hypothesis to capture inflectional impairments in individuals with DS.
More research employing fine-grained qualitative analyses of the language behavior observed in individuals with DS is needed to investigate whether language impairments in individuals with DS are indicative of a delayed or deviant language development, to confirm the supposition of a deviant development in verbal agreement marking in a subgroup of individuals with DS, and to explore the underlying causes for such a deficit.
Predictive Factors
An important issue raised by the research reported here is to determine factors that might allow for predicting whether or not language deficits are to appear in an individual with DS. Such predictive factors would be relevant in tailoring therapeutic interventions to the individual's needs. In an effort to detect such factors, a number of individual variables, including age (chronological, mental), IQ, gender, and socioeconomic status, were related to performance in verbal agreement. However, none of these factors differentiated individuals with DS who had acquired verbal agreement from those who had not. A number of other potentially relevant factors, such as the number of words in expressive and receptive vocabulary, reading abilities, or amount and quality of child-directed speech, might also have had some impact on the acquisition of verbal agreement markers, but were not assessed in this study. Whether and how these factors affect the acquisition of inflectional markers has to be left to future research.
The data suggest that the mostly mild and relatively short-dated hearing loss associated with episodes of otitis media did not affect the acquisition of subject–verb agreement, even when these episodes occurred frequently or persisted over several months (see Table 4). Note in this respect, that the one participant with DS who was wearing a hearing aid due to a slight but permanent conductive hearing loss achieved an accuracy score of 96.6% in verbal agreement, thus surpassing the acquisition criterion. Concurrent with the findings reported here, Chapman et al. (1998)  also reported no correlation between hearing loss and the acquisition of grammatical morphemes in 23 English-speaking adolescents with DS. Laws and Bishop (2003), in contrast, found a correlation between hearing threshold and use of the regular past tense marker in their sample of 14 English-speaking individuals with DS (aged 10 to 19 years). A correlation between hearing threshold and third singular marking was, however, not significant for this group. Unfortunately, hearing could not be evaluated in the German participants tested here, but we had to rely on prior diagnosis available to the parents. Thus, correlating the hearing level with the performance for individual agreement markers was not feasible here. Note, however, that a study on the acquisition of subject–verb agreement in German-speaking children with hearing loss indicated a very specific and characteristic performance pattern in these children (Penke & Rothweiler, 2018) that is different from the performance observed in the participants with DS studied here. Specifically, verbal agreement deficits in children with hearing loss selectively affected the second and third singular markers -s(t) and -t, whereas the marker -n was unaffected. In the children/adolescents with DS investigated here, in contrast, -n was as affected as the other markers. More research is needed to clarify the impact that a conductive hearing loss might have on the acquisition of subject–verb agreement.
The amount of speech and language therapy (in months) in relation to the chronological age of the participants with DS differentiated those participants who had acquired subject–verb agreement from those who had not. However, the relation was inverse, with participants of the DS−AGR subgroup receiving more speech and language therapy than participants of the DS + AGR subgroup who had acquired verbal agreement. This suggests that therapeutic interventions were intensified as reaction to the language problems observable in these participants.
A factor that could not be explored here is the type of trisomy as only two of the participants displayed a karyotype different from free trisomy. Of these, the individual with mosaic trisomy was one of the participants excluded from data analysis because he produced no analyzable utterances. The individual with a translocation displayed a deficit in verbal agreement marking with an accuracy score of 67%. Note in this respect, that according to Rondal (2001), existing studies did not uncover differences in language performance between the three main types of trisomy.
The one factor that clearly differentiated participants of the DS + AGR subgroup from participants of the DS−AGR subgroup was the performance in nonword repetition. Concurrent to the findings reported here, impairments in phonological working memory have been repeatedly reported for individuals with DS (e.g., Cairns & Jarrold, 2005; Jarrold et al., 2002; Laws, 2004; Næss, Lyster, Hulme, & Melby-Lervåg, 2011). Yoder et al. (2006)  have suggested that individuals with DS suffer from a generalized slowing in the processing of linguistic input that affects the acquisition of grammatical morphology. As a consequence, the processing of grammatical information expressed by inflectional markers is sacrificed in favor of processing semantic information. Children with DS might, hence, not have processed enough grammatical morphemes to initialize or complete the acquisition of these inflectional markers. According to Yoder et al., individual differences in speech processing might account for the large interindividual variance observed in individuals with DS with respect to language performance.
Although the proposal of Yoder et al. is certainly interesting, a note of caution is required. So far, few studies directly related performance in phonological short-term memory to production of inflectional morphology in individuals with DS, and those that have have come to different findings. Laws and Bishop (2003)  found a correlation between nonword repetition and use of the regular past tense marker -ed in their participants with DS, whereas nonword repetition was not significantly related to the third singular marker -s. Concordant with this latter finding is the observation of Lázaro, Garayzábal, and Moraleda (2013)  that difficulties in applying the Spanish plural marker in 30 children/adolescents with DS (age 6 to 18 years) were not related to performance in working memory as measured by the Wechsler Intelligence Scale for Children–Fourth Edition test (Wechsler, 2003). Presently, the co-occurrence of agreement deficits and phonological short-term memory deficits in German-speaking individuals with DS bears not enough substance to argue for any causal relationship between these deficits. More research, directly targeting the relationship between memory components and the acquisition of a variety of inflectional markers, is needed to explore whether and how deficits in phonological short-term memory affect the acquisition of inflectional markers in individuals with DS, or whether both deficits relate to a common basis (e.g., Lum, Conti-Ramsden, Page, & Ullman, 2012).
Before concluding, a note of caution is required in evaluating the results reported here. For one, the number of participants with DS that could be compared is still small, with 11 participants who have acquired agreement versus 16 participants who have not. Also, parents had to volunteer for their children to participate in this investigation, possibly biasing the sample toward parents who were more active in supporting the development of their children and, perhaps, also more anxious with regard to their language development. Although the sample might, thus, have been biased toward children who display problems in language development, the finding remains that, for a substantial proportion of German-speaking children and adolescents with DS, the acquisition of subject–verb agreement inflection is impaired.
Conclusion
Subject–verb agreement was found to be impaired in a substantial number of German-speaking children and adolescents with DS. In contrast to a control group of TD children of comparable nonverbal mental age, the affected individuals with DS did not achieve the acquisition criterion, indicating that the acquisition of subject–verb agreement marking was affected in these individuals. Production data collected in a picture-naming task indicated that difficulties in providing correct agreement markers cannot be accounted for by purely phonetic/phonological problems in articulating the respective inflectional markers in word-final position as the critical consonants /s/, /t/, and /n could be successfully produced as stem-final consonants of simplex nouns. The finding that all three tested agreement markers were similarly affected rather suggests an impairment in acquiring the morphosyntactic content of these inflectional markers. The frequent substitutions with the finite third singular marker -t, observed in a subgroup of 10 individuals with DS, are errors rarely detected in TD children. The occurrence of these highly unusual errors suggests that the development of subject–verb agreement might be deviant rather than delayed in individuals with DS. This assumption is supported by the observation of an asynchronous development of verbal agreement inflection and a general measure of grammar comprehension. Performance in a nonword repetition task was significantly related to performance in the agreement experiment. This suggests that impairments in verbal short-term memory might affect the acquisition of inflectional markers and might also account for the variability in performance in the group of participants with DS, with some participants displaying difficulties in agreement morphology while others have acquired this inflectional system. More fine-grained analyses of inflectional impairments and impairments in verbal working memory are needed to explore the nature and the determinants of inflectional impairments in children and adolescents with DS. To complement group data by individual data is an important step in achieving this endeavor.
Acknowledgments
The research reported in this article was supported by Grant PE 683/3-1 from the German Research Foundation (DFG) to the author. The author is grateful to the parents and children who undertook the effort of participating in the study. The author also thanks the members of the research team for their assistance in data collection and analysis. The article profited from comments to a previous version provided by Monika Rothweiler, Eva Wimmer, Bernadette Witecy, and two anonymous reviewers.
References
Abbeduto, L., Warren, S. E., & Connors, F. A. (2007). Language development in Down syndrome: From the prelinguistic period to the acquisition of literacy. Mental Retardation and Developmental Disabilities Research Reviews, 13, 247–261. [Article] [PubMed]
Abbeduto, L., Warren, S. E., & Connors, F. A. (2007). Language development in Down syndrome: From the prelinguistic period to the acquisition of literacy. Mental Retardation and Developmental Disabilities Research Reviews, 13, 247–261. [Article] [PubMed]×
Augst, G. (1984). Kinderwort: der aktive Kinderwortschatz [Children's words: The active vocabulary of children] . Frankfurt, Germany: Peter Lang.
Augst, G. (1984). Kinderwort: der aktive Kinderwortschatz [Children's words: The active vocabulary of children] . Frankfurt, Germany: Peter Lang.×
Baayen, H., Piepenbrock, R., & van Rijn, H. (1993). The CELEX lexical database (CD-ROM). Philadelphia, PA: Linguistic Data Consortium, University of Pennsylvania.
Baayen, H., Piepenbrock, R., & van Rijn, H. (1993). The CELEX lexical database (CD-ROM). Philadelphia, PA: Linguistic Data Consortium, University of Pennsylvania.×
Baddeley, A. D. (1986). Working memory. Oxford, United Kingdom: Clarendon Press.
Baddeley, A. D. (1986). Working memory. Oxford, United Kingdom: Clarendon Press.×
Baddeley, A. D., Gathercole, S., & Papagno, C. (1998). The phonological loop as a language learning device. Psychological Review, 105(1), 158–173. [Article] [PubMed]
Baddeley, A. D., Gathercole, S., & Papagno, C. (1998). The phonological loop as a language learning device. Psychological Review, 105(1), 158–173. [Article] [PubMed]×
Barnes, E. F., Roberts, J., Mirrett, P., Sideris, J., & Misenheimer, J. (2006). A comparison of oral structure and oral-motor function in young males with fragile X syndrome and Down syndrome. Journal of Speech, Language, and Hearing Research, 49(4), 903–917. [Article]
Barnes, E. F., Roberts, J., Mirrett, P., Sideris, J., & Misenheimer, J. (2006). A comparison of oral structure and oral-motor function in young males with fragile X syndrome and Down syndrome. Journal of Speech, Language, and Hearing Research, 49(4), 903–917. [Article] ×
Bishop, D. V. M. (2003). TROG-2—Test for Reception of Grammar–Second Edition. London, United Kingdom: Harcourt.
Bishop, D. V. M. (2003). TROG-2—Test for Reception of Grammar–Second Edition. London, United Kingdom: Harcourt.×
Bol, G., & Kuiken, F. (1990). Grammatical analysis of developmental language disorders: A study of the morphosyntax of children with specific language disorders, with hearing impairment and with Down's syndrome. Clinical Linguistics & Phonetics, 4(1), 77–86. [Article]
Bol, G., & Kuiken, F. (1990). Grammatical analysis of developmental language disorders: A study of the morphosyntax of children with specific language disorders, with hearing impairment and with Down's syndrome. Clinical Linguistics & Phonetics, 4(1), 77–86. [Article] ×
Booij, G. (2000). Morphology and phonology. In Booij, G., Lehmann, C., Mugdan, J., Kesselheim, W. & Skopetas, S. (Eds.), Morphology: An international handbook on inflection and word-formation (Vol. 1, pp. 335–344). Berlin, Germany: de Gruyter.
Booij, G. (2000). Morphology and phonology. In Booij, G., Lehmann, C., Mugdan, J., Kesselheim, W. & Skopetas, S. (Eds.), Morphology: An international handbook on inflection and word-formation (Vol. 1, pp. 335–344). Berlin, Germany: de Gruyter.×
Brown, R. (1973). A first language: The early stages. Cambridge, MA: Harvard University Press.
Brown, R. (1973). A first language: The early stages. Cambridge, MA: Harvard University Press.×
Bybee, J. L. (2000). Lexical, morphological and syntactic symbolization. In Booij, G., Lehmann, C., Mugdan, J., Kesselheim, W., & Skopetas, S. (Eds.), Morphology: An international handbook on inflection and word-formation (Vol. 1, pp. 370–377). Berlin, Germany: de Gruyter.
Bybee, J. L. (2000). Lexical, morphological and syntactic symbolization. In Booij, G., Lehmann, C., Mugdan, J., Kesselheim, W., & Skopetas, S. (Eds.), Morphology: An international handbook on inflection and word-formation (Vol. 1, pp. 370–377). Berlin, Germany: de Gruyter.×
Cairns, P., & Jarrold, C. (2005). Exploring the correlates of impaired non‐word repetition in Down syndrome. British Journal of Developmental Psychology, 23(3), 401–416. [Article]
Cairns, P., & Jarrold, C. (2005). Exploring the correlates of impaired non‐word repetition in Down syndrome. British Journal of Developmental Psychology, 23(3), 401–416. [Article] ×
Caselli, M. C., Monaco, L., Trasciani, M., & Vicari, S. (2008). Language in Italian children with Down syndrome and with specific language impairment. Neuropsychology, 22(1), 27–35. [Article] [PubMed]
Caselli, M. C., Monaco, L., Trasciani, M., & Vicari, S. (2008). Language in Italian children with Down syndrome and with specific language impairment. Neuropsychology, 22(1), 27–35. [Article] [PubMed]×
Chapman, R. S., Seung, H. K., Schwartz, S. E., & Kay-Raining Bird, E. (1998). Language skills of children and adolescents with Down syndrome: II. Production deficits. Journal of Speech, Language, and Hearing Research, 41, 861–873. [Article]
Chapman, R. S., Seung, H. K., Schwartz, S. E., & Kay-Raining Bird, E. (1998). Language skills of children and adolescents with Down syndrome: II. Production deficits. Journal of Speech, Language, and Hearing Research, 41, 861–873. [Article] ×
Christodoulou, C. (2015). Morphosyntactic illusions in Down syndrome: The role of phonetics and phonology. In Grillo, E., Jepson, K., & LaMendola, M. (Eds.), BUCLD 39: Proceedings of the 38th Boston University Child Language Development—Online Supplement. Available from http://www.bu.edu/bucld/supplementvol39/
Christodoulou, C. (2015). Morphosyntactic illusions in Down syndrome: The role of phonetics and phonology. In Grillo, E., Jepson, K., & LaMendola, M. (Eds.), BUCLD 39: Proceedings of the 38th Boston University Child Language Development—Online Supplement. Available from http://www.bu.edu/bucld/supplementvol39/ ×
Christodoulou, C., & Wexler, K. (2016). The morphosyntactic development of case in Down syndrome. Lingua, 184, 25–52. [Article]
Christodoulou, C., & Wexler, K. (2016). The morphosyntactic development of case in Down syndrome. Lingua, 184, 25–52. [Article] ×
Clahsen, H., Bartke, S., & Göllner, S. (1997). Formal features in impaired grammars: A comparison of English and German SLI children. Journal of Neurolinguistics, 10(2), 151–171. [Article]
Clahsen, H., Bartke, S., & Göllner, S. (1997). Formal features in impaired grammars: A comparison of English and German SLI children. Journal of Neurolinguistics, 10(2), 151–171. [Article] ×
Clahsen, H., Eisenbeiss, S., & Penke, M. (1996). Underspecification and lexical learning in early child grammars. In Clahsen, H. & Hawkins, R. (Eds.), Generative approaches to first and second language acquisition (pp. 129–159). Amsterdam, the Netherlands: John Benjamins.
Clahsen, H., Eisenbeiss, S., & Penke, M. (1996). Underspecification and lexical learning in early child grammars. In Clahsen, H. & Hawkins, R. (Eds.), Generative approaches to first and second language acquisition (pp. 129–159). Amsterdam, the Netherlands: John Benjamins.×
Clahsen, H., & Penke, M. (1992). The acquisition of agreement morphology and its syntactic consequences. In Meisel, J. (Ed.), The acquisition of verb placement (pp. 181–223). Dordrecht, the Netherlands: Kluwer Academic.
Clahsen, H., & Penke, M. (1992). The acquisition of agreement morphology and its syntactic consequences. In Meisel, J. (Ed.), The acquisition of verb placement (pp. 181–223). Dordrecht, the Netherlands: Kluwer Academic.×
Cleland, J., Wood, S., Hardcastle, W., Wishart, J., & Timmins, C. (2010). Relationship between speech, oromotor, language and cognitive abilities in children with Down's syndrome. International Journal of Language & Communication Disorders, 45(1), 83–95. [Article] [PubMed]
Cleland, J., Wood, S., Hardcastle, W., Wishart, J., & Timmins, C. (2010). Relationship between speech, oromotor, language and cognitive abilities in children with Down's syndrome. International Journal of Language & Communication Disorders, 45(1), 83–95. [Article] [PubMed]×
Davies, B. (1996). Auditory disorders. In Stratford, B. & Gunn, P. (Eds.), New approaches to Down syndrome (pp. 100–121). London, United Kingdom: Cassell.
Davies, B. (1996). Auditory disorders. In Stratford, B. & Gunn, P. (Eds.), New approaches to Down syndrome (pp. 100–121). London, United Kingdom: Cassell.×
Dryer, M. (2005). Prefixing versus suffixing in inflectional morphology. In Haspelmath, M., Dryer, M., Gil, D., & Comrie, B. (Eds.), The world atlas of language structures (pp. 110–114). Oxford, United Kingdom: Oxford University Press.
Dryer, M. (2005). Prefixing versus suffixing in inflectional morphology. In Haspelmath, M., Dryer, M., Gil, D., & Comrie, B. (Eds.), The world atlas of language structures (pp. 110–114). Oxford, United Kingdom: Oxford University Press.×
Eadie, P. A., Fey, M. E., Douglas, J. M., & Parsons, C. L. (2002). Profiles of grammatical morphology and sentence imitation in children with specific language impairment and Down syndrome. Journal of Speech, Language, and Hearing Research, 45(4), 720–732. [Article]
Eadie, P. A., Fey, M. E., Douglas, J. M., & Parsons, C. L. (2002). Profiles of grammatical morphology and sentence imitation in children with specific language impairment and Down syndrome. Journal of Speech, Language, and Hearing Research, 45(4), 720–732. [Article] ×
Fabbretti, D., Pizzuto, E., Vicari, S., & Volterra, V. (1997). A story description task in children with Down's syndrome: Lexical and morphosyntactic abilities. Journal of Intellectual Disability Research, 41(2), 165–179. [Article] [PubMed]
Fabbretti, D., Pizzuto, E., Vicari, S., & Volterra, V. (1997). A story description task in children with Down's syndrome: Lexical and morphosyntactic abilities. Journal of Intellectual Disability Research, 41(2), 165–179. [Article] [PubMed]×
Fowler, A. E. (1995). Linguistic variability in persons with Down syndrome: Research and implications. In Nadel, L. & Rosenthal, D. (Eds.), Down syndrome: Living and learning in the community (pp. 121–131). New York, NY: Wiley-Liss.
Fowler, A. E. (1995). Linguistic variability in persons with Down syndrome: Research and implications. In Nadel, L. & Rosenthal, D. (Eds.), Down syndrome: Living and learning in the community (pp. 121–131). New York, NY: Wiley-Liss.×
Fowler, A. E., Gelman, R., & Gleitman, L. R. (1994). The course of language learning in children with Down syndrome. In Tager-Flusberg, H. (Ed.), Constraints on language acquisition. Studies of atypical children (pp. 91–140). London, United Kingdom: Psychology Press.
Fowler, A. E., Gelman, R., & Gleitman, L. R. (1994). The course of language learning in children with Down syndrome. In Tager-Flusberg, H. (Ed.), Constraints on language acquisition. Studies of atypical children (pp. 91–140). London, United Kingdom: Psychology Press.×
Fox, A. V. (2003). Kindliche Aussprachestörungen [Articulatory deficits in children] . Idstein, Germany: Schulz-Kirchner.
Fox, A. V. (2003). Kindliche Aussprachestörungen [Articulatory deficits in children] . Idstein, Germany: Schulz-Kirchner.×
Fox, A. V. (2011). TROG-D. Test zur Überprüfung des Grammatikverständnisses. Idstein, Germany: Schulz-Kirchner.
Fox, A. V. (2011). TROG-D. Test zur Überprüfung des Grammatikverständnisses. Idstein, Germany: Schulz-Kirchner.×
Galeote, M., Soto, P., Sebastian, E., Checa, E., & Sanchez-Palacios, C. (2014). Early grammatical development in Spanish children with Down syndrome. Journal of Child Language, 41(1), 111–131. [Article] [PubMed]
Galeote, M., Soto, P., Sebastian, E., Checa, E., & Sanchez-Palacios, C. (2014). Early grammatical development in Spanish children with Down syndrome. Journal of Child Language, 41(1), 111–131. [Article] [PubMed]×
Gathercole, S. E., & Alloway, T. P. (2006). Practitioner review: Short-term and working memory impairments in neurodevelopmental disorders: Diagnosis and remedial support. The Journal of Child Psychology and Psychiatry, 47(1), 4–15. [Article] [PubMed]
Gathercole, S. E., & Alloway, T. P. (2006). Practitioner review: Short-term and working memory impairments in neurodevelopmental disorders: Diagnosis and remedial support. The Journal of Child Psychology and Psychiatry, 47(1), 4–15. [Article] [PubMed]×
Gathercole, S. E., & Baddeley, A. D. (1990). Phonological memory deficits in language disordered children: Is there a causal connection? Journal of Memory and Language, 29, 336–360. [Article]
Gathercole, S. E., & Baddeley, A. D. (1990). Phonological memory deficits in language disordered children: Is there a causal connection? Journal of Memory and Language, 29, 336–360. [Article] ×
Grijzenhout, J., & Penke, M. (2005). On the interaction of phonology and morphology in language acquisition and German and Dutch Broca's aphasia: The case of inflected verbs. In Booij, G. & van Marle, J. (Eds.), Yearbook of morphology 2005 (pp. 49–81). Dordrecht, the Netherlands: Springer.
Grijzenhout, J., & Penke, M. (2005). On the interaction of phonology and morphology in language acquisition and German and Dutch Broca's aphasia: The case of inflected verbs. In Booij, G. & van Marle, J. (Eds.), Yearbook of morphology 2005 (pp. 49–81). Dordrecht, the Netherlands: Springer.×
Grimm, H. (2001). Sprachentwicklungstest für drei- bis fünfjährige Kinder (SETK 3-5) [Test on language development in three- to five-year-old children] . Göttingen, Germany: Hogrefe & Huber.
Grimm, H. (2001). Sprachentwicklungstest für drei- bis fünfjährige Kinder (SETK 3-5) [Test on language development in three- to five-year-old children] . Göttingen, Germany: Hogrefe & Huber.×
Grimm, H. (2003). Sprachscreening für das Vorschulalter (SSV) [A screening of language development in preschool children] . Göttingen, Germany: Hogrefe & Huber.
Grimm, H. (2003). Sprachscreening für das Vorschulalter (SSV) [A screening of language development in preschool children] . Göttingen, Germany: Hogrefe & Huber.×
Hansson, K., Sahlén, B., & Mäki-Torkko, E. (2007). Can a ‘single hit’ cause limitations in language development? A comparative study of Swedish children with hearing impairment and children with SLI. International Journal of Language & Communication Disorders, 42(3), 307–323. [Article] [PubMed]
Hansson, K., Sahlén, B., & Mäki-Torkko, E. (2007). Can a ‘single hit’ cause limitations in language development? A comparative study of Swedish children with hearing impairment and children with SLI. International Journal of Language & Communication Disorders, 42(3), 307–323. [Article] [PubMed]×
Jarrold, C., Baddeley, A. D., & Phillips, C. E. (2002). Verbal short-term memory in Down syndrome: A problem of memory, audition, or speech? Journal of Speech, Language, and Hearing Research, 45(3), 531–544. [Article]
Jarrold, C., Baddeley, A. D., & Phillips, C. E. (2002). Verbal short-term memory in Down syndrome: A problem of memory, audition, or speech? Journal of Speech, Language, and Hearing Research, 45(3), 531–544. [Article] ×
Karmiloff-Smith, A. (2009). Nativism versus neuroconstructivism: Rethinking the study of developmental disorders. Developmental Psychology, 45(1), 56–63. [Article] [PubMed]
Karmiloff-Smith, A. (2009). Nativism versus neuroconstructivism: Rethinking the study of developmental disorders. Developmental Psychology, 45(1), 56–63. [Article] [PubMed]×
Kent, R. D., & Vorperian, H. K. (2013). Speech impairment in Down syndrome: A review. Journal of Speech, Language, and Hearing Research, 56, 178–210. [Article]
Kent, R. D., & Vorperian, H. K. (2013). Speech impairment in Down syndrome: A review. Journal of Speech, Language, and Hearing Research, 56, 178–210. [Article] ×
Laws, G. (2004). Contributions of phonological memory, language comprehension and hearing to the expressive language of adolescents and young adults with Down syndrome. The Journal of Child Psychology and Psychiatry, 45(6), 1085–1095. [Article] [PubMed]
Laws, G. (2004). Contributions of phonological memory, language comprehension and hearing to the expressive language of adolescents and young adults with Down syndrome. The Journal of Child Psychology and Psychiatry, 45(6), 1085–1095. [Article] [PubMed]×
Laws, G., & Bishop, D. V. M. (2003). A comparison of language abilities in adolescents with Down syndrome and children with specific language impairment. Journal of Speech, Language, and Hearing Research, 46(6), 1324–1339. [Article]
Laws, G., & Bishop, D. V. M. (2003). A comparison of language abilities in adolescents with Down syndrome and children with specific language impairment. Journal of Speech, Language, and Hearing Research, 46(6), 1324–1339. [Article] ×
Lázaro, M., Garayzábal, E., & Moraleda, E. (2013). Differences on morphological and phonological processing between typically developing children and children with Down syndrome. Research in Developmental Disabilities, 34(7), 2065–2074. [Article] [PubMed]
Lázaro, M., Garayzábal, E., & Moraleda, E. (2013). Differences on morphological and phonological processing between typically developing children and children with Down syndrome. Research in Developmental Disabilities, 34(7), 2065–2074. [Article] [PubMed]×
Lenneberg, E., Nichols, I., & Rosenberger, E. (1964). Primitive stages of language development in mongolism. Research Publications, Association For Research in Nervous and Mental Disease, 42, 119–157.
Lenneberg, E., Nichols, I., & Rosenberger, E. (1964). Primitive stages of language development in mongolism. Research Publications, Association For Research in Nervous and Mental Disease, 42, 119–157.×
Leonard, L., Weismer, S., Miller, C., Francis, D., Tomblin, B., & Kail, R. (2007). Speed of processing, working memory, and language impairment in children. Journal of Speech, Language, and Hearing Research, 50(2), 408–428. [Article]
Leonard, L., Weismer, S., Miller, C., Francis, D., Tomblin, B., & Kail, R. (2007). Speed of processing, working memory, and language impairment in children. Journal of Speech, Language, and Hearing Research, 50(2), 408–428. [Article] ×
Levy, Y., & Eilam, A. (2013). Pathways to language: A naturalistic study of children with Williams syndrome and children with Down syndrome. Journal of Child Language, 40(1), 106–138. [Article] [PubMed]
Levy, Y., & Eilam, A. (2013). Pathways to language: A naturalistic study of children with Williams syndrome and children with Down syndrome. Journal of Child Language, 40(1), 106–138. [Article] [PubMed]×
Loane, M., Morris, J., Addor, M., Arriola, L., Budd, J., Doray, B., … Dolk, H. (2013). Twenty-year trends in the prevalence of Down syndrome and other trisomies in Europe: Impact of maternal age and prenatal screening. European Journal of Human Genetics, 21(1), 27–33. [Article] [PubMed]
Loane, M., Morris, J., Addor, M., Arriola, L., Budd, J., Doray, B., … Dolk, H. (2013). Twenty-year trends in the prevalence of Down syndrome and other trisomies in Europe: Impact of maternal age and prenatal screening. European Journal of Human Genetics, 21(1), 27–33. [Article] [PubMed]×
Lum, J., Conti-Ramsden, G., Page, D., & Ullman, M. (2012). Working, declarative and procedural memory in specific language impairment. Cortex, 48(9), 1138–1154. [Article] [PubMed]
Lum, J., Conti-Ramsden, G., Page, D., & Ullman, M. (2012). Working, declarative and procedural memory in specific language impairment. Cortex, 48(9), 1138–1154. [Article] [PubMed]×
Næss, K.-A., Lyster, S.-A., Hulme, C., & Melby-Lervåg, M. (2011). Language and verbal short-term memory skills in children with Down syndrome: A meta-analytic review. Research in Developmental Disabilities, 32(6), 2225–2234. [Article] [PubMed]
Næss, K.-A., Lyster, S.-A., Hulme, C., & Melby-Lervåg, M. (2011). Language and verbal short-term memory skills in children with Down syndrome: A meta-analytic review. Research in Developmental Disabilities, 32(6), 2225–2234. [Article] [PubMed]×
Norbury, C., Bishop, D., & Briscoe, J. (2001). Production of English finite verb morphology: A comparison of SLI and mild-moderate hearing impairment. Journal of Speech, Language, and Hearing Research, 44, 165–178. [Article]
Norbury, C., Bishop, D., & Briscoe, J. (2001). Production of English finite verb morphology: A comparison of SLI and mild-moderate hearing impairment. Journal of Speech, Language, and Hearing Research, 44, 165–178. [Article] ×
O'Neill, M., & Henry, A. (2002). The grammatical morpheme difficulty in Down's syndrome. Belfast Working Papers in Language and Linguistics, 15, 64–73.
O'Neill, M., & Henry, A. (2002). The grammatical morpheme difficulty in Down's syndrome. Belfast Working Papers in Language and Linguistics, 15, 64–73.×
Penke, M., & Rothweiler, M. (2018). Comparing specific language impairment and hearing impairment: Different profiles in German verbal agreement morphology. Language Acquisition, 25, 39–57. https://doi.org/10.1080/10489223.2016.1204545 [Article]
Penke, M., & Rothweiler, M. (2018). Comparing specific language impairment and hearing impairment: Different profiles in German verbal agreement morphology. Language Acquisition, 25, 39–57. https://doi.org/10.1080/10489223.2016.1204545 [Article] ×
Ring, M., & Clahsen, H. (2005). Morphosyntax in Down's syndrome: Is the extended optional infinitive hypothesis an option? Stem-, Spraak- en Taalpathologie, 13(1), 3–13.
Ring, M., & Clahsen, H. (2005). Morphosyntax in Down's syndrome: Is the extended optional infinitive hypothesis an option? Stem-, Spraak- en Taalpathologie, 13(1), 3–13.×
Roberts, J., Long, S., Malkin, C., Barnes, E., Skinner, M., Hennon, E., & Anderson, K. (2005). A comparison of phonological skills of boys with fragile X syndrome and Down syndrome. Journal of Speech, Language, and Hearing Research, 48(5), 980–995. [Article]
Roberts, J., Long, S., Malkin, C., Barnes, E., Skinner, M., Hennon, E., & Anderson, K. (2005). A comparison of phonological skills of boys with fragile X syndrome and Down syndrome. Journal of Speech, Language, and Hearing Research, 48(5), 980–995. [Article] ×
Rondal, J. (2001). Language in mental retardation: Individual and syndromic differences, and neurogenetic variation. Swiss Journal of Psychology, 60(3), 161–178. [Article]
Rondal, J. (2001). Language in mental retardation: Individual and syndromic differences, and neurogenetic variation. Swiss Journal of Psychology, 60(3), 161–178. [Article] ×
Rondal, J., & Edward, S. (1997). Language in mental retardation. London, United Kingdom: Whurr Publishers Ltd.
Rondal, J., & Edward, S. (1997). Language in mental retardation. London, United Kingdom: Whurr Publishers Ltd. ×
Rothweiler, M., Chilla, S., & Clahsen, H. (2012). Subject–verb agreement in specific language impairment: A study of monolingual and bilingual German-speaking children. Bilingualism: Language and Cognition, 15(1), 39–57. [Article]
Rothweiler, M., Chilla, S., & Clahsen, H. (2012). Subject–verb agreement in specific language impairment: A study of monolingual and bilingual German-speaking children. Bilingualism: Language and Cognition, 15(1), 39–57. [Article] ×
Rupela, V., Manjula, R., & Velleman, S. (2010). Phonological processes in Kannada-speaking adolescents with Down syndrome. Clinical Linguistics & Phonetics, 24(6), 431–450. [Article] [PubMed]
Rupela, V., Manjula, R., & Velleman, S. (2010). Phonological processes in Kannada-speaking adolescents with Down syndrome. Clinical Linguistics & Phonetics, 24(6), 431–450. [Article] [PubMed]×
Rutter, T., & Buckley, S. (1994). The acquisition of grammatical morphemes in children with Down's syndrome. Down Syndrome Research and Practice, 2(2), 76–82. [Article]
Rutter, T., & Buckley, S. (1994). The acquisition of grammatical morphemes in children with Down's syndrome. Down Syndrome Research and Practice, 2(2), 76–82. [Article] ×
Schaner-Wolles, C. (2004). Spared domain-specific cognitive capacities? Syntax and morphology in Williams syndrome and Down syndrome. In Bartke, S. & Siegmüller, J. (Eds.), Williams syndrome across languages (pp. 93–124). Amsterdam, the Netherlands: John Benjamins.
Schaner-Wolles, C. (2004). Spared domain-specific cognitive capacities? Syntax and morphology in Williams syndrome and Down syndrome. In Bartke, S. & Siegmüller, J. (Eds.), Williams syndrome across languages (pp. 93–124). Amsterdam, the Netherlands: John Benjamins.×
Stathopoulou, N., & Clahsen, H. (2010). The perfective past tense in Greek adolescents with Down syndrome. Clinical Linguistics & Phonetics, 24(11), 870–882. [Article] [PubMed]
Stathopoulou, N., & Clahsen, H. (2010). The perfective past tense in Greek adolescents with Down syndrome. Clinical Linguistics & Phonetics, 24(11), 870–882. [Article] [PubMed]×
Stoel-Gammon, C. (2001). Down syndrome phonology: Developmental patterns and intervention strategies. Down Syndrome Research and Practice, 7, 93–100. [Article]
Stoel-Gammon, C. (2001). Down syndrome phonology: Developmental patterns and intervention strategies. Down Syndrome Research and Practice, 7, 93–100. [Article] ×
Tellegen, P., Laros, J., & Petermann, F. (2007). SON-R 2½ - 7. Non-verbaler Intelligenztest [Test of nonverbal intelligence] . Göttingen, Germany: Hogrefe & Huber.
Tellegen, P., Laros, J., & Petermann, F. (2007). SON-R 2½ - 7. Non-verbaler Intelligenztest [Test of nonverbal intelligence] . Göttingen, Germany: Hogrefe & Huber.×
Tuller, L., & Delage, H. (2014). Mild-to-moderate hearing loss and language impairment: How are they linked? Lingua, 139, 80–101. [Article]
Tuller, L., & Delage, H. (2014). Mild-to-moderate hearing loss and language impairment: How are they linked? Lingua, 139, 80–101. [Article] ×
United Nations Educational, Scientific and Cultural Organization. (2012). International standard classification of education: ISCED 2011. Montréal, Québec, Canada: UNESCO Institute for Statistics.
United Nations Educational, Scientific and Cultural Organization. (2012). International standard classification of education: ISCED 2011. Montréal, Québec, Canada: UNESCO Institute for Statistics.×
Vennemann, T. (1988). Preference laws for syllable structure and the explanation of sound change. Berlin, Germany: Mouton.
Vennemann, T. (1988). Preference laws for syllable structure and the explanation of sound change. Berlin, Germany: Mouton.×
Wechsler, D. (2003). Wechsler Intelligence Scale for Children–Fourth Edition—WISC-IV. San Antonio, TX: The Psychological Corporation.
Wechsler, D. (2003). Wechsler Intelligence Scale for Children–Fourth Edition—WISC-IV. San Antonio, TX: The Psychological Corporation.×
Wexler, K., Schütze, C., & Rice, M. (1998). Subject case in children with SLI and unaffected controls: Evidence for the Agr/Tns omission model. Language Acquisition, 7(2–4), 317–344. [Article]
Wexler, K., Schütze, C., & Rice, M. (1998). Subject case in children with SLI and unaffected controls: Evidence for the Agr/Tns omission model. Language Acquisition, 7(2–4), 317–344. [Article] ×
Witecy, B., & Penke, M. (2017). Language comprehension in children, adolescents, and adults with Down syndrome. Research in Developmental Disabilities, 62, 184–196. [Article] [PubMed]
Witecy, B., & Penke, M. (2017). Language comprehension in children, adolescents, and adults with Down syndrome. Research in Developmental Disabilities, 62, 184–196. [Article] [PubMed]×
Yoder, P., Camarata, S., Camarata, M., & Williams, S. (2006). Association between differentiated processing of syllables and comprehension of grammatical morphology in children with Down syndrome. American Journal on Mental Retardation, 111(2), 138–152. [Article] [PubMed]
Yoder, P., Camarata, S., Camarata, M., & Williams, S. (2006). Association between differentiated processing of syllables and comprehension of grammatical morphology in children with Down syndrome. American Journal on Mental Retardation, 111(2), 138–152. [Article] [PubMed]×
Footnotes
1 The term appendix goes back to Vennemann (1988)  who proposed a so-called coronal-appendix outside the rhyme to accommodate word-final alveolar obstruents in German. The appendix is not part of the syllable's rhyme. As a consequence, the sonority sequencing principle—which requires that sonority decreases toward the right edge of the rhyme—is not violated by the affixation of a final alveolar obstruent /s/ that will often lead to an increase in sonority (e.g., /pakst/ “pack, second singular”; the sequence /kst/ would be prohibited by the sonority sequencing principle if all obstruents were inside the rhyme).
The term appendix goes back to Vennemann (1988)  who proposed a so-called coronal-appendix outside the rhyme to accommodate word-final alveolar obstruents in German. The appendix is not part of the syllable's rhyme. As a consequence, the sonority sequencing principle—which requires that sonority decreases toward the right edge of the rhyme—is not violated by the affixation of a final alveolar obstruent /s/ that will often lead to an increase in sonority (e.g., /pakst/ “pack, second singular”; the sequence /kst/ would be prohibited by the sonority sequencing principle if all obstruents were inside the rhyme).×
2 In German, the third plural marker -n is homonymous to the infinitival marker -n. Whether a finite verb form (third plural) or a nonfinite verb form (the infinitive) is produced can often be determined by the position of the verb: Nonfinite verbs are placed clause-finally, finite verbs appear in V2 position. However, if an utterance only consists of a subject and a verb, such as the utterance Die Kinder rennen; “The children are running” the possibility that the verb might have been incorrectly inflected with the infinitival marker -n instead of the correct third plural marker -n cannot be ruled out, as the V2 position cannot be distinguished from the clause-final position. Despite this possibility, the -n inflection was scored as correct in these cases.
In German, the third plural marker -n is homonymous to the infinitival marker -n. Whether a finite verb form (third plural) or a nonfinite verb form (the infinitive) is produced can often be determined by the position of the verb: Nonfinite verbs are placed clause-finally, finite verbs appear in V2 position. However, if an utterance only consists of a subject and a verb, such as the utterance Die Kinder rennen; “The children are running” the possibility that the verb might have been incorrectly inflected with the infinitival marker -n instead of the correct third plural marker -n cannot be ruled out, as the V2 position cannot be distinguished from the clause-final position. Despite this possibility, the -n inflection was scored as correct in these cases.×
3 A detailed quantitative and qualitative analysis of the participants' performance in the TROG-D test (including its relation to age, IQ, and nonword repetition) has been published by Witecy and Penke (2017) . The reader is referred to this publication for more details regarding performance in this test. A reviewer of this article asked about information regarding the TROG-D T-scores the individuals with DS would have achieved given their chronological age. Unfortunately, the TROG-D is only standardized for German-speaking children up to an age of 10;11, and 16 of the 29 participants with DS who underwent this test were older than 10;11 at testing. Thus, a T-score related to their chronological age cannot be determined for these participants. Regarding the 13 subjects younger than 11 years, grammatical comprehension is deficient with respect to their chronological age, although (with two exceptions) appropriate for their nonverbal mental age.
A detailed quantitative and qualitative analysis of the participants' performance in the TROG-D test (including its relation to age, IQ, and nonword repetition) has been published by Witecy and Penke (2017) . The reader is referred to this publication for more details regarding performance in this test. A reviewer of this article asked about information regarding the TROG-D T-scores the individuals with DS would have achieved given their chronological age. Unfortunately, the TROG-D is only standardized for German-speaking children up to an age of 10;11, and 16 of the 29 participants with DS who underwent this test were older than 10;11 at testing. Thus, a T-score related to their chronological age cannot be determined for these participants. Regarding the 13 subjects younger than 11 years, grammatical comprehension is deficient with respect to their chronological age, although (with two exceptions) appropriate for their nonverbal mental age.×
Figure 1.

Individual accuracy scores for subject–verb agreement for participants in the Down syndrome (DS) and the typically developing (TD) group. (Each symbol marks the score of one participant.)

 Individual accuracy scores for subject–verb agreement for participants in the Down syndrome (DS) and the typically developing (TD) group. (Each symbol marks the score of one participant.)
Figure 1.

Individual accuracy scores for subject–verb agreement for participants in the Down syndrome (DS) and the typically developing (TD) group. (Each symbol marks the score of one participant.)

×
Table 1. Paradigm of German subject–verb agreement.
Paradigm of German subject–verb agreement.×
Value
Verb form
Verb stem on consonant (example: kauf- “buy”)
Verb stem on vowel (example: bau- “build”)
Person/number Stem-suffix Spoken form Stem-suffix Spoken form
1/singular kauf-e kauf bau-e bau
2/singular kauf-st kaufs bau-st baus
3/singular kauf-t kauft bau-t baut
1/plural kauf-en kau.fn bau-en baun
2/plural kauf-t kauft bau-t baut
3/plural kauf-en kau.fn bau-en baun
Note. Dot indicates syllable boundary.
Note. Dot indicates syllable boundary.×
Table 1. Paradigm of German subject–verb agreement.
Paradigm of German subject–verb agreement.×
Value
Verb form
Verb stem on consonant (example: kauf- “buy”)
Verb stem on vowel (example: bau- “build”)
Person/number Stem-suffix Spoken form Stem-suffix Spoken form
1/singular kauf-e kauf bau-e bau
2/singular kauf-st kaufs bau-st baus
3/singular kauf-t kauft bau-t baut
1/plural kauf-en kau.fn bau-en baun
2/plural kauf-t kauft bau-t baut
3/plural kauf-en kau.fn bau-en baun
Note. Dot indicates syllable boundary.
Note. Dot indicates syllable boundary.×
×
Table 2. Characteristics of children with Down syndrome (DS) and typically developing (TD) control children.
Characteristics of children with Down syndrome (DS) and typically developing (TD) control children.×
Group N Sex Chronological age (years;months) Nonverbal mental age (years;months) IQ a
DS 32 13 females, 19 males M = 11;01 M = 4;05 M = 59.2
SD = 3;06 SD = 0;11 SD = 13.1
Range = 4;07–19;00 Range = 3;05–6;05 Range = 48–97
TD 16 7 females, 9 males M = 4;00 M = 4;05 M = 106
SD = 0;07 SD = 0;10 SD = 11.2
Range = 3;01–5;00 Range = 3;05–5;10 Range = 86–124
Comparison between groups (independent-samples t test) p < .001 p = .97 p < .001
Note. Nonverbal mental age and IQ values were determined on the basis of the reasoning scale of the Snijders–Oomen Non-Verbal Intelligence Test.
Note. Nonverbal mental age and IQ values were determined on the basis of the reasoning scale of the Snijders–Oomen Non-Verbal Intelligence Test.×
a IQ corrected for Flynn effect.
IQ corrected for Flynn effect.×
Table 2. Characteristics of children with Down syndrome (DS) and typically developing (TD) control children.
Characteristics of children with Down syndrome (DS) and typically developing (TD) control children.×
Group N Sex Chronological age (years;months) Nonverbal mental age (years;months) IQ a
DS 32 13 females, 19 males M = 11;01 M = 4;05 M = 59.2
SD = 3;06 SD = 0;11 SD = 13.1
Range = 4;07–19;00 Range = 3;05–6;05 Range = 48–97
TD 16 7 females, 9 males M = 4;00 M = 4;05 M = 106
SD = 0;07 SD = 0;10 SD = 11.2
Range = 3;01–5;00 Range = 3;05–5;10 Range = 86–124
Comparison between groups (independent-samples t test) p < .001 p = .97 p < .001
Note. Nonverbal mental age and IQ values were determined on the basis of the reasoning scale of the Snijders–Oomen Non-Verbal Intelligence Test.
Note. Nonverbal mental age and IQ values were determined on the basis of the reasoning scale of the Snijders–Oomen Non-Verbal Intelligence Test.×
a IQ corrected for Flynn effect.
IQ corrected for Flynn effect.×
×
Table 3. Experimental results—group comparisons.
Experimental results—group comparisons.×
Measure Group
p value
DS TD
Subject–verb agreement a
 Number of participants 27 16
 Number of analyzable responses 615 435
 Number of analyzable responses per participant M = 22.8, SD = 8.4 M = 27.2, SD = 2.54 p = .02
 Correct agreement in % M = 78.6%, SD = 22.1 M = 99.3%, SD = 1.4 p < .001
Picture naming
 Number of participants 30 16
 Number of analyzable responses 484 259
 Number of analyzable responses per participant M = 16.1, SD = 1.5 M = 16.2, SD = 1 p = .9
 Correct production of final consonant in % M = 91.9%, SD = 15 M = 98.8%, SD = 2.6 p = .02
Nonword repetition
 Number of participants 32 16
 Raw score M = 4.3, SD = 3.9 M = 7.3, SD = 2.1 p = .001
TROG-D
 Number of participants 29
T-score M = 46.3, SD = 9.2
Note. DS = Down syndrome; TD = typically developing; TROG-D = Test for Reception of Grammar - German.
Note. DS = Down syndrome; TD = typically developing; TROG-D = Test for Reception of Grammar - German.×
a Data of five participants with DS who produced fewer than three analyzable utterances each not included.
Data of five participants with DS who produced fewer than three analyzable utterances each not included.×
Table 3. Experimental results—group comparisons.
Experimental results—group comparisons.×
Measure Group
p value
DS TD
Subject–verb agreement a
 Number of participants 27 16
 Number of analyzable responses 615 435
 Number of analyzable responses per participant M = 22.8, SD = 8.4 M = 27.2, SD = 2.54 p = .02
 Correct agreement in % M = 78.6%, SD = 22.1 M = 99.3%, SD = 1.4 p < .001
Picture naming
 Number of participants 30 16
 Number of analyzable responses 484 259
 Number of analyzable responses per participant M = 16.1, SD = 1.5 M = 16.2, SD = 1 p = .9
 Correct production of final consonant in % M = 91.9%, SD = 15 M = 98.8%, SD = 2.6 p = .02
Nonword repetition
 Number of participants 32 16
 Raw score M = 4.3, SD = 3.9 M = 7.3, SD = 2.1 p = .001
TROG-D
 Number of participants 29
T-score M = 46.3, SD = 9.2
Note. DS = Down syndrome; TD = typically developing; TROG-D = Test for Reception of Grammar - German.
Note. DS = Down syndrome; TD = typically developing; TROG-D = Test for Reception of Grammar - German.×
a Data of five participants with DS who produced fewer than three analyzable utterances each not included.
Data of five participants with DS who produced fewer than three analyzable utterances each not included.×
×
Table 4. Comparison of individual factors in participants with Down syndrome who have (DS + AGR) or have not (DS−AGR) acquired subject–verb agreement.
Comparison of individual factors in participants with Down syndrome who have (DS + AGR) or have not (DS−AGR) acquired subject–verb agreement.×
Measure DS + AGR DS−AGR p value
Number of participants 11 16
Number of males and females 7 males, 4 females 8 males, 8 females p = .7
Chronological age (years;months) M = 11;03, SD = 4;01 M = 11;01, SD = 3;06 p = .92
Mental age (years;months) M = 4;06, SD = 1;01 M = 4;07, SD = 0;10 p = .84
IQ M = 61, SD = 12.8 M = 61, SD 14.2 p = .96
Number of participants with frequent (> 5 per year) or persistent (> 3 months) infections of the middle ear in the past 3 4 p = 1
Amount of speech and language therapy in relation to chronological age a M = 0.58, SD = 0.13 M = 0.8, SD = 0.08 p = .001
Education score b of mother M = 6.6, SD = 1.4 M = 5.3, SD = 2 p = .07
Education score b of father M = 6.3, SD = 1.5 M = 5.6, SD = 2.1 p = .35
Production score for stem-final consonants M = 98.4%, SD = 2.8 M = 94.8%, SD = 8.4 p = .13
Score for nonword repetition M = 6.8, SD = 2.6 M = 3.9, SD = 4 p = .04
Number of participants with TROG-D T-score > 40 9 of 11 11 of 15 p = .67
Note. AGR = subject-verb agreement; TROG-D = Test for Reception of Grammar - German.
Note. AGR = subject-verb agreement; TROG-D = Test for Reception of Grammar - German.×
a Months with speech and language therapy divided by chronological age in months).
Months with speech and language therapy divided by chronological age in months).×
b Education of mother and father were measured on the 9 level scale (ranging from 0 = early childhood education to 8 = doctoral or equivalent level) of the International Standard Classification of Education (United Nations Educational, Scientific and Cultural Organization, 2012).
Education of mother and father were measured on the 9 level scale (ranging from 0 = early childhood education to 8 = doctoral or equivalent level) of the International Standard Classification of Education (United Nations Educational, Scientific and Cultural Organization, 2012).×
Table 4. Comparison of individual factors in participants with Down syndrome who have (DS + AGR) or have not (DS−AGR) acquired subject–verb agreement.
Comparison of individual factors in participants with Down syndrome who have (DS + AGR) or have not (DS−AGR) acquired subject–verb agreement.×
Measure DS + AGR DS−AGR p value
Number of participants 11 16
Number of males and females 7 males, 4 females 8 males, 8 females p = .7
Chronological age (years;months) M = 11;03, SD = 4;01 M = 11;01, SD = 3;06 p = .92
Mental age (years;months) M = 4;06, SD = 1;01 M = 4;07, SD = 0;10 p = .84
IQ M = 61, SD = 12.8 M = 61, SD 14.2 p = .96
Number of participants with frequent (> 5 per year) or persistent (> 3 months) infections of the middle ear in the past 3 4 p = 1
Amount of speech and language therapy in relation to chronological age a M = 0.58, SD = 0.13 M = 0.8, SD = 0.08 p = .001
Education score b of mother M = 6.6, SD = 1.4 M = 5.3, SD = 2 p = .07
Education score b of father M = 6.3, SD = 1.5 M = 5.6, SD = 2.1 p = .35
Production score for stem-final consonants M = 98.4%, SD = 2.8 M = 94.8%, SD = 8.4 p = .13
Score for nonword repetition M = 6.8, SD = 2.6 M = 3.9, SD = 4 p = .04
Number of participants with TROG-D T-score > 40 9 of 11 11 of 15 p = .67
Note. AGR = subject-verb agreement; TROG-D = Test for Reception of Grammar - German.
Note. AGR = subject-verb agreement; TROG-D = Test for Reception of Grammar - German.×
a Months with speech and language therapy divided by chronological age in months).
Months with speech and language therapy divided by chronological age in months).×
b Education of mother and father were measured on the 9 level scale (ranging from 0 = early childhood education to 8 = doctoral or equivalent level) of the International Standard Classification of Education (United Nations Educational, Scientific and Cultural Organization, 2012).
Education of mother and father were measured on the 9 level scale (ranging from 0 = early childhood education to 8 = doctoral or equivalent level) of the International Standard Classification of Education (United Nations Educational, Scientific and Cultural Organization, 2012).×
×