Dynamic interactions between perception and production: An integrated experimental and observational study

Lead Research Organisation: University of York
Department Name: Psychology

Abstract

Abstracts are not currently available in GtR for all funded research. This is normally because the abstract was not required at the time of proposal submission, but may be because it included sensitive information such as personal details.

Publications

10 25 50
publication icon
McGillion M (2013) Supporting Early Vocabulary Development: What Sort of Responsiveness Matters? in IEEE Transactions on Autonomous Mental Development

 
Description We know little about the ongoing relationship between emergent abilities to perceive speech and produce speech-like vocalisations (babble), and we also lack clear evidence regarding the relationship of those abilities to the onset of word learning and use. This study explored this relationship in a longitudinal design, following 59 children in the development of word recognition and segmentation, consistent consonant production, and word production. The model is based upon Dynamic Systems theory (Thelen & Smith, 1994), which assumes that when relatively simple skills combine, more complex behavioural patterns can emerge. In this case, basic vocal and perceptual skills eventually lead to the onset of an early lexicon.
We evaluated production skills based on weekly recordings from age 9 months to the point where the child makes reliable and consistent use of two consonants or Vocal Motor Schemes (VMS). Perception capacity was assessed on the basis of two headturn tasks, word-form recognition at 10 months and segmentation at 11 months. In both tasks words likely to be familiar from the home were contrasted with words which unlikely to be familiar. Advances in word production were assessed based on monthly home recordings up to age 18 months. Age at onset of first word use is defined as the age at which 4 different spontaneous words are identifiable in a 30-minute naturalistic observational session.
To summarize our findings:
1. Lexicon size at age 9 months (based on parental report) correlated significantly with performance on the two experimental tasks, with a larger lexicon correlating with low 10-month word recognition scores and with high 11-month segmentation scores. The fact that the relationship is with low scores in the one case and with high scores in the other gives further credence to our interpretation of both types of preference as 'success' (or a mark of relative infant advance).
2. The infants who had two VMS already by the time they participated in the word recognition task were significantly more variable and more extreme in their preferences than the infants who did not yet produce consonants consistently at the time of the experiment. This indicates that the infants who were more advanced in production were also more advanced in word recognition.
3. Both age of attaining two VMS and performance on the word recognition task are important for segmentation success: The only group among whom the majority of infants succeeded at the segmentation task was the group of infants who had both (a) attained two VMS by the time of the word recognition experiment and (b) succeeded at word recognition.
4. The only variable related to the age at which the children started to produce words was the age at attaining two VMS, which proved a strong predictor of age at first word use. Neither experimental task showed any relationship with the age of production of the first words.
Exploitation Route We are planning a collaboration with colleagues in Sheffield to continue exploring predictors of first word production.
Sectors Healthcare