Poster workshop
Chairs: Emmanuel Devouche and Maya Gratier
Discussant: Colwyn Trevarthen
All human experience is conditioned from within the subject by a rhythmic sense of time passing. All actions achieve prospective control by systematic anticipation within a space and time generated by and centred within the body and mind. Control of movements resides in a coherent programme of sensory-motor co-ordination that unites all modalities from the start, driving naive impulses of action, experience and communication. Musicality, a polyrhythmia of subjective activity with coherence as emotional narrative, serves as the basis for sympathetic awareness and communication of mental states and processes. All this is evident from the ways infants behave from birth -- the precise temporal co-ordination of the movements of their body parts, the spontaneous pulse of their actions, and their reactions to expressions of other persons. Most striking is the sensitivity of even the youngest infants for both the form of a human partner's expressions (as in neonatal imitations), and both the timing and emotional quality of human sounds, contacts and gestures. The motive state and physiological self-regulations of an infant are profoundly responsive to the pulse, modulated sequencing and affective contours of other persons movements appreciated by touch and mechanical stimulation of the body, by hearing the human voice, or by sight of the human face and hands. Auditory awareness of human expressive states is functional before birth, and reactions to the pulse and melody of a mother's speech and singing are strong throughout infancy and early childhood.In this Poster Workshop a new approach to the musical talents of infants and their mothers is strongly represented. Precise acoustic methods are added to micro-analysis of videos to observe the natural collaboration between infant and adult in joint 'performances'. These show that the enjoyment together of musical experiences is not just a matter of 'co-regulation' or 'mutual attunement'. There is a joining of an essential vitality of the core motive processes of two brains that are actively creating and anticipating an ordered flow of experience that is shared. Infants are very active and anticipating participants, evoking the sensitive intuitive musical 'narratives' from their mothers in appropriate tempos and moods, and being greatly affected by what their mothers present. The musical companionship is highly emotional and thrives on pleasurable or joyful states. It is profoundly interfered with by separation of the two minds, and by emotional states in either one that tend to create negative withdrawal.The concept of Communicative Musicality defined on the basis of acoustic analysis of mother-infant vocal interactions has given a firm basis for these new studies, of the emotions, habits and cultural significance of the human talent for appreciating music.
Details of individual items:
poster
Effective communication has been shown to be structured around precise and coherent interactive timing patterns, in verbal as well as in non-verbal communication between adults and infants. Timing patterns, moreover, which manifest themselves through the coordination of various means of expression: through voice, gesture, posture, gaze. We propose a method for studying the integration of interactive rhythmic expression in face-to-face mother-infant interaction. Studies of synchrony and entrainment suggest that adult-infant interaction is intermodal by nature (Stern, Beebe) but few studies have systematically highlighted the correspondences between the dynamic rhythms of voice and gesture in interaction (Condon & Sander). Entrainment is thought to occur bidirectionally between interacting partners (interactional synchrony) and also within each individual so that the same expression can be channeled through the voice, the gestures, or the gaze (self-synchrony). By 'sharing the beat', interacting partners spontaneously co-create rhythmic exchanges and variations which allow them to anticipate each other's expressions (Trevarthen, Fogel). We aim to show that interactions through voice and movement are structured around common beats organized hierarchically within an ongoing musical flow which is neither predictable nor random (Tronick, Fogel). We suggest that the rhythmic organization of mother-infant interaction is closest to the way musical rhythm is organized, and that the type of meanings derived from it is akin to musical meaning. We also aim to show that infants communicate rhythmically with their whole being rendering possible the sharing of feelings and motives between them (Trevarthen). We want to highlight the importance of variations around the beat in expressive rhythm between mothers and infants. The data employed in this study consist in simultaneous digital audio recordings and video recordings of face-to-face mother-infant interactions. Infants are all between 2 and 3 months old and dyads belong to one of three different cultural backgrounds: Indian, French and American. Analyses are made separately on different levels of the interaction: vocal, gestural and visual. The acoustic analysis of the vocal interaction is based on methods of beat extraction and segmentation into groups, as well as on pitch plotting techniques. The detection of rhythmic and melodic groups enables us to envisage the hierarchical rhythmic structure of the interaction. The analyses of gesture and gaze are done by frame-by-frame analysis (25 frames/second) of video recordings. By presenting the different analyses as time-series it becomes possible to compare the rhythmic patterns of each with the others. We expect to find significant rhythmic correspondences for each interacting partner between the different modes of expression (self-synchrony), as well as within the dyad between each partner and regardless of the mode of expression. We also expect to find that mothers and infants from different cultural contexts all follow rhythmic patterns but we are open to possible cultural variations in the particular rhythmic structures they create.
poster
This presentation will examine the idea that there is an intrinsic human musicality which facilitates communication and companionship when attentive and sympathetic persons meet. Supporting data will be drawn predominantly from studies of mother/infant interactions (Malloch, in press). Mother/infant vocalisations are examined using computer-based acoustic analysis techniques, and methods are introduced for the analysis of Pulse, Quality and Narrative in mother/infant vocalisations. With references to the disciplines of music therapy and biomusicology, as well as studies of time and human perception, I pose the hypothesis that human companionship is largely facilitated through a 'Communicative Musicality.' This is a human ability that allows people to share a jointly constructed narrative moving through time. This narrative may be formed from the timing and prosody of our speech, the timing and shapes of our bodily gestures, as well as the emotional and dramatic content conveyed through our words. Thus, Communicative Musicality is not concerned with what we generally understand as 'music,' but rather with those elements of human communication that allow a co-ordinated relationship to occur and be maintained with an other. And it is these very human abilities that enable the culturally specific instances of performed music to manifest. The vital engine of Communicative Musicality is Rhythm.
poster
Mother-infant interaction features a number of repetitive and rhythmic behaviours that help structure early communication independently of the referential content of language and thus the infant's cognitive development as a whole (Bruner, 1983; Papaeliou & Trevarthen, 1994). Prominent among these are repetitive actions with variations, rhythmic chants, and fully musical songs, sometimes combined with synchronous moving of the baby's limbs, dandling, or dancing with the baby held in the arms. Such repetitive or rhythmical sequences may be fundamental means of structuring early communicative interactions in that they provide a principal means for generating temporal expectations (e.g. Stern 1977). Of specific interest in this context is the nature and extent of behavioural entrainment between mother and infant. In collaboration with Colwyn Trevarthen we have collected a corpus of videotaped sessions in which 25 Swedish mothers engage in spontaneous play, song and dance with their infants. 11 of the dyads were filmed (SONY Digital Videocam, DSR-30P) when the infant was 3, 6, 9 and 12 months of age, and the remainder beginning at between 4 and 9 months of age. Thecorpus contains a rich assortment of rhythmic behaviours spanning a range of temporal scales from fractions of minutes (e.g. 'turn-taking') to fractions of seconds (e.g. marking the musical beat in children's songs).A number of methodological issues involved in defining and characterising the rhythmic content of behavioural sequences are illustrated with the help of transcriptions from the video materials. We then use sample transcriptions of such sequences to address the issue of entrainment between mother and infant, that is, the extent to which rhythmic coordination occurs between mother and infant at different ages, and the nature of such coordination. We discuss our results in relation to the question of the role of rhythmicity and entrainment in the mother-infant interaction and infant development more generally.
poster
Lullabies, a distinct musical style, are found throughout the world. They are often associated with rocking behavior and can be popular songs or classical music. Recent research has shown their universality and their importance in accompanying infants to sleep (Stork, 1993). Others have tried to determine their characteristics: For example, S. Trehub and coll have shown that adults were able to recognize a lullaby from a pair of musical excepts, based on soothing, softness and slow tempo. They also pointed out several musical regularity of lullabies: slow tempi, simple and decreasing contours.Surprisingly, no research has focused on the main function of lullabies, that of their capacity to send infants to sleep. The aim of the present study is to determine whether infants fall asleep faster when listening to lullabies compared with other music or no musical stimulus, when the mother is not present, and without rocking behavior.A preliminary study identified classical lullabies which could be identified as such by adults. Three classical lullabies and three corresponding classical pieces were tested, with two different methods, identification of the lullaby in a pair of stimuli (as used by Trehub & al) and random identification of each piece. The results were contradictory, indicating that cues which are used to recognize lullaby depend strongly on musical context. From these data a lullaby and a short musical piece by Grieg were chosen as stimuli for the main experiment. Two popular lullabies (one sung by a male voice and one by a female voice) were added, each with a corresponding play-song by the same singer.An analysis of the tempi of the stimuli verified previous data: the tempi of lullabies were slower than corresponding music and were situated inside the described tempo range (Baruch & Stork,1996). In the main experiment 12 infants (3 to 9 months of age, mean 5 month and a half) were filmed in day nurseries, during the afternoon nap. Each infant was observed at three different experimental sessions. At each, they heard one of the three possible stimuli during the beginning of the nap: a lullaby, a musical excerpt and no musical stimulus (i.e. the 'silence' of the nursery) in a counterbalanced order over the group. An observation sheet determine the delay to sleep.The results indicate that infants seem to fall asleep faster with music (all musical conditions), versus silence (706 sec. vs 994sec. marginaly significant (F(1,22)3D3.4; p<.10). Lullabies do not facilitate sleep comparing with corresponding music (815 sec. vs 598 sec.). The comparison is opposite to what was expected, even the difference is not significant, probably due to high variability of data in lullabies conditions. Nevertheless, if one examines for each infant the stimulus inducing sleep the fastest, 42% fell asleep faster with the lullaby, 33% with the corresponding music and only 25 % in the silent condition. In conclusion, if adults recognize well the lullabies, as previously shown, and consider that lullabies have a real soothing effect on infants, we could suppose that the real effect on infants is provoked by the musical environment by itself, be it a lullaby or another kind of corresponding music.
poster
Music appears in every culture and the sounds of music appeal to human beings at every stage of life, from the last trimester of pregnancy to old age (Blacking; Bjorkvold). Experimental studies of infants' musical perceptions and preferences reveal that they can discriminate various features of musical sound that are significant in western European music (Trehub and colleagues, and others). Research on mother-infant communication suggest that musicality, i. e. the production and appreciation of rhythmic and melodious patterns from voice or instruments, appears to be a central part of the interaction that occurs with caregivers as infants develop over the first year of life (Papousek and Papousek; Stern). In this longitudinal field study, 15 mother-infant pairs were each recorded 11 times in their homes from the 2nd to the 10th month, by VHS video and digital audio -- every 15 days from the 2nd to the 4th month, and thereafter monthly until 10 months. The aim was to find out how the infants (7 girls, 8 boys) reacted to recorded music when on their own, and to mother's rhythmic games and singing, and to observe developments in the infants' responses by body-movement and vocalisation, as well as the rhythms of their spontaneous movements in the absence of musical sounds. An observed episode lasted 12 minutes and consisted of 4 conditions. In Condition 1 (1 minute) the infant was filmed moving on his or her own while the mother was out of sight and silent. In Condition 2 (2 minutes), the infant was again alone, but this time a tape recording of a traditional Greek baby song was played at a moderate volume near the baby. The song presented was changed every two months, each!! new song having a slightly faster tempo than the preceding one. At the end of each song there was a pause of 10 seconds, and then the song started again. This was done to see infant's behaviour during the pause. In Condition 3 (6 minutes), a period of free play between mother and infant was recorded. The instruction given to the mother was 'please play with your baby as you usually do'. In Condition 4 (2 minutes), mother and infant played freely as in Condition 3, but now the same traditional song was heard as in Condition 2. Many mothers sang, danced or played rhythmic games for their infants. Data from micro-analysis of the video (with a video-logger to 0.04 seconds) and acoustic analysis of vocalisations and other sounds (Hypersignal and acoustic programs created by Malloch, in press) illustrates the development of spontaneous rhythmic behaviour, both in the absence of music and in response to music, as well as the development of musical interactions in mother-infant play, and the effect of recorded music on their behaviour. Infants' stilling to listen, orientation to the source of the music and expressions of pleasure preceded active participation by bouncing, waving limbs and vocalising. Even the youngest clearly recognised musical forms. The main results and implications for theory and practice of research, as well as for interpretation of development of infant consciousness and motives for self-expression and communication, are discussed.
poster
Mother-Infant communication that satisfies both partners involves various musical elements. In cases where the mother is suffering from postnatal depression, qualities of rhythmic attunement, reciprocity and overall satisfaction with the interaction all decline. This case study reports detailed acoustic analysis of vocal interactions (Malloch et al 1997; Malloch, 1999) between two mothers and their infants at eight weeks and six months. One mother had a clinical diagnosis of depression over this period. The other mother was entirely healthy (Murray et al, 1996). Results showed the depressed mother to produce quieter, lower pitched vocalisations, punctuated by longer pauses. Disruption was also evident in the depressed mother's turn-taking behaviour. Matching of pitch, low arousal, less 'joining in' and negative mood states in the infant of the depressed mother suggested corresponding low affect. These characteristics of the depressed dyad's communication improved as clin!!ical symptoms declined. Both dyads showed periodicity in timing of interactions although this was considerably slower, and less co-ordinated, in the depressed pair. The control dyad produced more evidence of reciprocal, happy communication with regular timing and 'singing' voice quality. These results present preliminary evidence that features of communicative musicality (Malloch, Trevarthen) are important in healthy, reciprocal interactions, and highlight the importance of an innate pulse and shared timing within a musical framework in the organisation and regulation of infant behaviours.The same acoustic methods have also been utilised in the analysis of classroom interactions between teachers and primary school pupils aged between seven and eleven years. Initial results have shown that disruption to consistent and regular timing with clear turn-taking cues results in less frequent, and shorter, contributions by the pupils. It seems likely that this may have implications for the pupils self-esteem and motivation for learning, besides feelings of satisfaction and
poster
Interaction studies have highlighted the degree to which infant and caregiver are coordinated in the timing of their expressions (Stern; Trevarthen, Tronick and Cohn). The study of vocal face-to-face interaction in particular suggests that the vocal exchanges between mothers and infants between two and three months of age are structured around timing units which enable them to anticipate each others' expressions as they unfold and to 'play around with the beat' (Fogel, Stern, Trevarthen). In this study we compare the rhythmic structures of vocal face-to-face interaction of a clinical sample with those of a non-clinical control group. Healthy mothers and infants seem to jointly partake in the creation of innovative as well as canonical expressions which have a hierarchically organized structure lending them narrative coherence. Deviations from the underlying rhythmic structure are central to its creative and innovative character, but rhythmic deviations or 'unlawful events' (Stern) must remain within a given range to be understood by both partners as embedded in the flow of the interaction. It is in this sense that mother-infant vocal interaction can be likened to improvisation in Jazz music. Studies have shown that the vocal interactions of mothers suffering from depression ad their infants tend to loose their dynamic, co-creative quality and thus their contingency (Field et al., Trevarthen). The aim of this study is to examine the rhythmic organization of the vocal interactions of mothers with Borderline Personality Disorder and their infants. We would like to ascertain whether these interactions are characterized, like those of depressed dyads, by a lack of contingency or whether the specific features of Borderline Personality Disorder such as sudden disruptions of mood and inconsistencies of identity might entail a relative disorganization at the rhythmic level resulting in a lack of hierarchical coherence, a disjointed rhythm.Digital audio recordings of three minutes of face-to-face vocal interaction of a group of dyads with mothers suffering from Borderline Personality Disorder and a non-clinical control group are analyzed and compared. Diagnoses are made on the basis of an interview conducted by a skilled psychiatrist in accordance with the DSM-IV criteria. Infants are all between 8 and 12 weeks old, were born full term and in good health. The acoustic analysis of the vocal interactions is based on methods of beat extraction and segmentation into groups as well as on pitch plotting techniques. The segmentation into groups of the vocalizations provides us with the basic time units, measured in milliseconds, which combine to form larger rhythmic and melodic groups structured hierarchically to form phrase-like expressions. We expect to find that the vocal interactions of healthy mother-infant dyads will be structured in a coherent manner, showing regularities of beat and patterns in the rhythmic-melodic groups, whereas the vocal interactions of mothers with Borderline Personality Disorder and their infants will lack this type of internal coherence.
poster
Interaction studies carried out in experimental settings have focussed primarily on healthy mother-infant dyads and on dyads in which the mother suffers from postnatal depression. Dyads in which a mother suffers from another psychopathology, especially those of Axis 2 (DSM4), have been largely ignored. From a very young age, infants display behaviors that appear to be highly organized as well as purposeful (Stern, Tronick). The different configurations of infants' expressions (Tronick, Weinberg) are coordinated in time between each other in an intermodal manner as well as between an infant and its caregiver (Stern, Beebe).In this study, we examine the timing of vocal and behavioral interactive configurations in three cases: a dyad where the mother has a diagnosis of Borderline Personality Disorder, a dyad where the mother is clinically depressed, and a healthy control dyad. The Face-to-Face Still Face Paradigm (Tronick) provides our framework for the study of timing in dynamic interaction. The Still Face procedure consists of two minutes of Free-Play followed by a sudden and prolonged interruption of the interaction by the mother which lasts two minutes and followed again by two minutes of Free Play (reunion). It allows us to compare the levels of organisation in the interaction before and after the interruption, and to examine infant behavior reaction to psychological stress. Previous work has shown that infants of healthy mothers as opposed to those of depressed mothers are more visibly distressed by the stress caused by the Still Face procedure. Also, depressed mothers were more likely to interact with less ' repair ' during reunion time than healthy control mothers, suggesting that the timing of their mutual regulation mechanisms is not well adjusted (Weinberg, Tronick).Other studies on depressed dyads (Field, Murray) tend to support the hypothesis that these interactions are characterized primarily by a lack of contingency between the partners' expressions, thus impeding the effective sharing of affect between them. We expect to find in this study that the type of temporal patterns displayed by the depressed dyad will lack contingency but that those displayed by the mother with Borderline Personality Disorder and her infant will lack coherence, will be somewhat chaotic. The three infants in this study were born full term and in good health, were female, and were two and a half months old. The three mothers were physically healthy. The diagnosis of Borderline Personality Disorder was made according to DSM4 Axis II criteria (SIDP4, BID) and that of postnatal depression was based on clinical interviews and diagnostic tools (EPDS, MADRS).Timing patterns were obtained from the inter-onset-intervals between the various affective configurations consisting of facial and vocal expression, posture, gaze, and distance between partners- as delineated by the Mother and Infant Regulatory Scoring Systems developed by Tronick and Weinberg.