Infant prosodic expressions in mother-infant communication
Prosody, generally defined as any perceivable modulation of duration, pitch or loudness in the voice that conveys meaning, has been identified as part of the linguistic system, or compared with the sound system of Western classical music. This thesis proposes a different conception, namely that prosody is a phenomenon of human expression that precedes, and to a certain extent determines the form and function of utterances in any particular language or music system. Findings from studies of phylogenesis and ontogenesis are presented in favour of this definition. Consequently, prosody of infant vocal expressions, which are made by individuals who have not yet developed either language or musical skills, is investigated as a phenomenon in itself, with its own rules. Recognising theoretical and methodological deficiencies in the linguistic and the Piagetian approaches to the development of infant prosodic expressions, this thesis supports the view that the origins of language are to be sought in the expressive dialogues between the mother and her prelinguistic child that are generated by intuitive motives for communication. Furthermore, infant vocalisations are considered as part of a system of communication constituted by all expressive modalities. Thus, the aim is to investigate the role of infant prosodic expressions in conveying emotions and communicative functions in relation to the accompanying non vocal-behaviours. A crossectional Pilot Study involving 16 infants aged 26 to 56 weeks and their mothers was undertaken to help in the design of the Main Study. The Main Study became a case description of two first born infants and their mothers; a boy (Robin) and a girl (Julie) both aged 30 weeks at the beginning of the study. The infants were filmed in their home every fortnight for five months in a structured naturalistic setting which included the following conditions: mother-infant free-play with their own toys, mother-infant play without using objects, the infant playing alone, motherinfant play with objects provided by the researcher, a 'car task' for eliciting cooperative play, and the mother staying unresponsive. Each filming session lasted approximately thirty minutes. In order to get an insight into the infants' 'meaning potential' expressed in their vocalisations, the mothers were asked to visit the department sometime in the interval between two filming sessions and, while watching the most recent video, to report what they felt their infant was conveyingif anything- in each vocalisation. Three types of analysis were carried out: a) An Analysis of Prosody - An attempt was made to obtain an objective, and not linguistically based account of infant prosodic features. First measurements were obtained of the duration and the fundamental frequency curve of each vocalisation by means of a computer programme for sound analysis. The values of fundamental frequency were then logarithmically transformed into a semitone scale in order to obtain measurements more sensitive to the mother's perception. b) A Functional Micro-Analysis of Non-Vocal Behaviours from Videos - The non vocal behaviours of mother and infant related with each vocalisation were codified without sound to examine to what extent the mothers relied for their interpretations on non-vocal behaviours accompanying vocalisations. c) An Analysis of the Mothers' Interpretations - The infants' messages were defined as perceived by their mother. The corpus comprised 713 vocalisations (322 for the boy and 391 for the girl) selected from a corpus of 864, and 143 minutes of video recording (64 for the boy and 79 for the girl). Correlations between the above three assessments were specified through statistical analysis. The findings from both infants indicate that between seven and eleven months prosodic patterns are not related one to one with particular messages. Rather, prosody distinguishes between groups of messages conveying features of psychological motivation, such as 'emotional', 'interpersonal', 'referential', 'assertive' or 'receptive'. Individual messages belonging to the same message group according to the analysis of prosody, are distinguished on the basis of the accompanying nonvocal behaviours. Before nine months, 'interpersonal' vocalisations display more 'alerting' prosodic patterns than 'referential' vocalisations. After nine months prosodic patterns in Robin's vocalisations differentiate between 'assertive' and 'receptive' messages, the former being expressed by more 'alerting' prosodic patterns than the latter. This distinction reflects a better Self-Other awareness. On the other hand, Julie's vocalisations occurring in situations of 'Joint Interest' display different prosodic patterns from her vocalisations uttered in situations of 'Converging Interest'. These changes in the role infant prosody reflect developments in the infants' motivational organisation which will lead to a more efficient control of intersubjective orientation and shared attention to the environment. Moreover, it was demonstrated that new forms of prosodic expression occur in psychologically mature situations, while the psychologically novel situations are expressed by mature prosodic forms. The above results suggest that at the threshold to language, prosody does not primarily serve identifiable linguistic functions. Rather, in spite of individual differences in form of their vocalisations, both infants use prosody in combination with other modalities as part of an expressive system, that conveys information about their motives. In this way prosody facilitates intersubjective and later cooperative communication, on which language development is built. To what extent such prelinguistic prosodic patterns are similar in form to those of the target language is a crucial issue for further investigation.