Abstract
This study analyzed spontaneous language samples of 10 late talking toddlers and 11 typically developing toddlers using Dore’s primitive speech act analysis (1974). Late talkers did not produce as many pragmatic functions, and the ones that they do produce did not prompt continued interaction with a communication partner.
Introduction
Toddlers who are late talkers demonstrate reduced expressive vocabulary in the absence of physical, social, cognitive, or sensory impairment; they are usually identified at age 2, when they produce fewer than 50 words and do not combine words (Rescorla, 1989). Much of the research describes late talkers’ long-term linguistic outcomes (e.g., Rescorla, 2005, 2009). These studies reveal that as many as half are later diagnosed with specific language impairment (Leonard, 2000), whereas others who catch up and are labeled late bloomers (Thal, Tobias, & Morrison, 1991) demonstrate later difficulties with reading skills into adolescence (Rescorla, 2009).
The expressive language delay in late-talking toddlers affects semantics, phonology, syntax, and morphology (e.g., Carson, Klee, Carson, & Hime, 2003; Ellis Weismer & Evans, 2002; Paul & Jennings, 1992; Rescorla, Roberts, & Dahlsgaard, 2000; Williams & Elbert, 2003). In contrast, we know relatively little about pragmatic development among late talkers. Pragmatic skills are critical to children’s social development and yet, only one study has evaluated pragmatic development in late-talking toddlers. Bonifacio and colleagues (2007), using a parent-report questionnaire, found delays in assertive and responsive conversational skills when compared with age-matched peers. This study focused on only two pragmatic functions, yet we know that young children are expected to demonstrate a range of pragmatic functions. The present study examined pragmatic language functions in late-talking toddlers and their typically developing, age-matched peers using Dore’s primitive speech acts (1974) to understand the frequency and diversity of early pragmatic skills. If we learn that late talkers show pragmatic differences or delays, we may be able to intervene and, in turn, affect their communicative effectiveness, social interaction, and (potentially) their later academic performance.
Method
Participants
A total of 21 toddlers, aged 21 to 25 months (M = 23.2, SD = 1.3), participated in this study. Ten (8 males, 2 females) were late talkers and 11 (9 males, 2 females) served as age- and gender-matched, typical developing controls. All participants came from monolingual English-speaking homes. All completed standardized language and cognitive testing to determine eligibility; caregivers completed a Social-Economic Status (SES) questionnaire (adapted from Hollingshead, 1975) and a case history form. As noted in Table 1, late talkers produced fewer than 50 words on the MacArthur-Bates Communicative Development Inventories, Second Edition, Words and Sentences (CDI-2; Fenson et al., 2007), scored below average on the expressive communication subtest of Preschool Language Scale (4th ed.; PLS-4; Zimmerman, Steiner, & Pond, 2002), and average on the auditory comprehension subtest of the PLS-4 and the Bayley Scales of Infant Development, Mental Scale (BSID; Bayley, 2006). Typically developing controls produced at least 200 words expressively on the CDI-2 and had average scores on the PLS-4 (expressive and auditory comprehension) and the BSID. All toddlers passed a hearing screening. The typically developing toddlers did not differ from the late-talking toddlers in age, gender, SES, and maternal education. All participants were part of a larger investigation examining word learning in toddlers.
Language and Cognitive Measures
Note: PLS-4 = Preschool Language Scale, Fourth Edition; CDI 2 = MacArthur-Bates Communicative Development Inventories, Second Edition, Words and Sentences; BSID-3 = Bayley Scales of Infant Development, Third Edition, Mental Scale.
Auditory comprehension standard score.
Expressive communication standard score.
Procedures
Spontaneous play-based language samples were obtained in a research laboratory with a familiar adult (the first author) using a uniform set of toys: a puzzle, a book, Mr. Potato head, and pretend food items. Each language sample was audio and video recorded using a Sony Handycam DCR-DVD408 with an external Sony ECM-WH1R wireless microphone worn by the adult. Retherford’s guideline (2000) for 100 utterances was met for the typically developing toddlers. In contrast, no late talker met the criterion of 100 utterances, despite at least 30 min of play. A trained research assistant, blind to group membership, transcribed the utterances using Dore’s system of primitive speech acts (1974). Specifically, each utterance was classified as one of the following primitive speech acts: labeling, repeating, answering, requesting action, requesting answer, calling, greeting, protesting, practicing, or babble.
Data Analysis
The percentage of utterances in each category was calculated, which served as the dependent variable. Independent variables included participant group (late talker vs. typically developing) and pragmatic category type (i.e., each of the 10 pragmatic categories from Dore, 1974). ANOVA was used with post hoc Fisher Least Significant Difference (LSD) testing to examine interactions where p < .05. For all analyses, effect sizes are reported in
Results
Toddlers with typical language development produced an average of 5.9 pragmatic functions (SD = 1.0). Toddlers who were late talkers produced fewer pragmatic functions than their peers (M = 4.7, SD = 1.2). No participant (late talker or typically developing) produced utterances categorized as calling; therefore, this pragmatic category was not included in the analyses. A one-way ANOVA indicated that typically developing toddlers produced significantly more pragmatic functions than toddlers who were late talkers, F(1, 19) = 6.623, p = .02,
Figure 1 shows the average number of utterances per pragmatic category. For the typically developing group, the most frequent categories were labeling (31.88%), answering (29.89%), and repeating (13.34%). Similar results were found from the late talkers; the most frequent categories were labeling (31.22%), repeating (25.04%), and answering (24.19%). A repeated measures ANOVA examined differences in the frequency of the nine pragmatic categories. A main effect of pragmatic category, F(8, 152) = 26.05, p < .0000001,
Speech acts in late talkers and toddlers with typical language development
Post hoc Fisher LSD testing was performed to examine differences between groups for each pragmatic function. Repeating (p = .004) and babble (p = .03) were more frequent in the toddlers who were late talkers. Requesting answer was significantly greater in the toddlers with typical language development (p = .05). No significant differences were found between the typically developing and the late-talking groups in labeling, answering, requesting action, greeting, protesting, and practicing (p > .05).
Discussion
This study found that late-talking toddlers show pragmatic interactions that are similar and different to their age-matched peers. Both late talkers and typically developing toddlers frequently used labeling and answering. These pragmatic functions were frequently observed by Dore (1974), suggesting that toddlers who are late talkers show some pragmatic behaviors that are age appropriate. In contrast, toddlers who were late talkers used fewer pragmatic functions than their typically developing peers.
Although both late talkers and typically developing toddlers frequently used repeating, it was observed more often in late talkers. The increased use of repeating in late talkers may reflect difficulties with word learning (Ellis Weismer & Evans, 2002). In addition, babble was used more frequently in late talkers. This is not surprising because late-talking toddlers show delays and disordered phonological development patterns (e.g., Williams & Elbert, 2003). It is possible that the differences observed in late talkers’ phonological development are expressed in their pragmatic language. That is, late talkers attempt to produce words, but because of an immature phonological system, the listener perceives these utterances as unintelligible (i.e., babble).
The frequently used pragmatic functions in late talkers (i.e., babble and repeating) may result in a less direct influence on their communication partner, when compared with requesting answer, which was observed frequently in typically developing toddlers. Requesting answer prompts a conversational turn and thus continues a social interaction. In contrast, pragmatic functions used by late talkers did not encourage continued interaction with a communication partner, suggesting that they may be less proactive in using pragmatic functions that create early language-learning social opportunities.
It is unclear whether the pragmatic language differences in late-talking toddlers reflect an underlying disorder in social interaction skills or the other language domains (e.g., semantics, phonology). Intervention should focus not only on increasing vocabulary and syntax but also on pragmatic functions, especially if the children do not have a means to maintain and continue communication-based interactions.
Footnotes
This research was completed in part by Ilana Kaufman to fulfill Science Research Seminar at Hunter College High School. A portion of these data was presented at the New York City Science and Engineering Fair.
Declaration of Conflicting Interests
The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.
Funding
This study was funded in part by the Graduate Center, CUNY, Doctoral Students Dissertation Grant awarded to Michelle MacRoy-Higgins.