Child/Computer interaction: observation in classroom setting.

Background in music education: The present study deals with an area still under-studied in the field of music education (Webster 2002), that of interactive musical systems, in an attempt to understand in what way it can affect the learning and the musical creativity of children. In particular, we chose to study young children, 3/5 years old, because at this age the problem of the interaction between child and machine takes on a fundamental role in the learning process. Imberty (2002), and Daniel Stern (1977), describe the development of young children as based on the vocal play between child and mother (lallation, baby-talk), characterised by the mechanism of repetition and variation. The point of interest is to verify what type of music development arises when this interaction takes place not between two human subjects, but rather between a child and a machine. Background in artificial intelligence: At the Sony CSL, a system was elaborated able to produce music in the same style as the person playing the keyboard, the Continuator (Pachet 2002). This system is based on the notion of Interactive Reflective systems. The core concept of this approach is to teach powerful – but complex – musical processes indirectly by putting the user in a situation where these processes are performed not by the user, nor by the machine, but by the actual interaction between the user and the system. The ability of the system to attract and hold the attention of children can be interpreted through the theory of Flow introduced by psychologist Csikszentmihalyi (1990). Aims: An experimental protocol has been established to observe systematically some interesting children's behaviours observed in preliminary experience. The general aim is to understand in what way the children relate to interactive musical systems, what kinds of musical behaviours are developed, and how interactive systems can be used in the educational field to stimulate creativity and the pleasure of playing. Method: The experiment was carried out with 27 children of 3/5 years, in an Italian kindergarten. Three sessions were held once a day for 3 consecutive days. In every session, the children were asked to play in 4 different ways: just with the keyboard, with the keyboard and the Continuator, with another child, and both with another child and the Continuator. All the sessions were recorded on video. The attention span of the children was measured for each task. Two case-studies were analysed. Successively, the most interesting conducts were selected to be tested also on the other children. Results: It was possible to observe a sort of life cycle of interaction, that move from surprise, to different phases of excitement, analytical behaviour, invention, readjustment and relaunch. The two tasks involving the system gave rise to the longest attention spans and show how most children reach a stable level of attention characterized by a strong intrinsic motivation, and by the socialization of the musical experience (join attention). The listening conducts were particularly varied and analytical. Conclusion: The results suggest that the Continuator, or other similar interactive reflective systems, is able to develop interesting child/computer interaction, thanks above all to its ability to replicate the musical style of the child that is playing