Abstract
Background: Children who have reached the age of 2 years without having acquired a 50‐word vocabulary and/or who use no word combinations are referred to in the literature as ‘Late Talkers’. Research has not yet identified the factors that cause slow development of expressive language; in particular, relatively little research has been carried out on the characteristics of the maternal linguistic input received by Late Talkers. In fact, the majority of studies in this area have focused principally on the mothers' semantic, pragmatic and interactive verbal behaviour.
Aims: The objective was to verify whether the input addressed to a group of Late Talkers differs to that addressed to a group of typically developing peers. It was expected that there would be differences in the prosodic and structural/lexical properties of the input, which by creating a ‘less than optimal environment’ for language acquisition processes, could affect the children's rate of linguistic development.
Methods & Procedures: The spontaneous maternal speech of nine Italian‐speaking mothers of late‐talking toddlers (LT) at 20 months of age was examined and compared with that of nine mothers of age‐matched children with typical language development (Late Starters). Data were collected during video‐recorded play sessions, which were transcribed and coded based on the structural–lexical and acoustical characteristics of the mothers' speech. Group differences were tested for the following variables: number of utterances per minute, mean number of words per utterance (MLUw), number of nouns per minute, number of different nouns per minute, common nouns as a percentage of overall words, type/token ratio for common nouns, one‐word utterances as a percentage of the total number of utterances, one‐word utterances formed by a common noun as a percentage of the total number of utterances, mean syllable duration, duration of the final syllable of the utterance, F0 maximum (the highest F0 in the utterance), F0 minimum (the lowest F0 in the utterance) and F0 range (by subtracting F0 min from F0 max).
Outcomes & Results: The mothers of LT produced common nouns and mark nouns by a pitch peak with a significantly lower frequency than the mothers of the children with typical language development. The mothers of LT also used flat pitch contours more often than the mothers of the typically developing peers.
Conclusions: The study provides evidence that maternal input addressed to Late Talkers can contribute to slowing the process of language acquisition. Based on these findings, additional investigations that examine the contribution of maternal verbal behaviour to the development of language delay are warranted.
Notes
1. The pitch of a sound is determined by the frequency of the vibration of the vocal cords (fundamental frequency, or F0). The faster the vocal cords vibrate, the higher the pitch. Variation of intonation corresponds to variation in the fundamental frequency and yields an F0 contour (De Boysson‐Bardies, Citation1999: 230).
2. Some subsequent sessions had to be cancelled due to infants' illness or other external factors.
3. If it was not possible to analyse a randomly selected utterance after digitalization due to technical impairments (particularly, low amplitudes sometimes impede the software to extract pitch contour), the random selection was repeated.
4. This procedure is more reliable than the automatic calculation of the highest (or lowest) F0 by the software in so far as it allows the researcher (1) to control that the value obtained corresponds to the mother's emission of voice (and not, for example, to an environmental noise) and (2) to calculate an interrater agreement.
5. A two‐tailed test was taken into consideration for this analysis only as we did not have a precise hypothesis on the effect of different MLU on the process of language acquisition.