MIMIC EXPRESSION AND PIANO PERFORMANCE Roberto Caterina* , Luisa Bonfiglioli*, Mario Baroni ** and Anna Rita Addessi** *Department of Psychology – University of Bologna – Italy **Department of Music – University of Bologna – Italy ABSTRACT In piano players facial expression and other nonverbal cues may have an important role at three different levels: 1) in order to anticipate musical performance; 2) in order to modulate musical interpretation; 3) in order to express emotion according to the interpreter’s own listening experience. It is not always very easy to separate the previous aspects. In the work we are going to present we selected two pieces by Claude Debussy’s Preludes, played by Bruno Canino. We compared the musical score with the mimic movements. Facial expression (eyebrow raising or frowning, closing or opening eyes, closing or opening mouth, smiling, were the most common indicators), gaze direction, head position and posture were the main nonverbal signals taken into consideration. This study has to be considered as a first attempt in order to build a grid in which musical elements can be compared with nonverbal signals. 1. INTRODUCTION Piano players often use their body and their facial expression during their performances. Individual differences are remarkable: some piano players (such as Benedetti Michelangeli) are very controlled in their facial and body expression and only a very few expressions are visible; others (such as Bruno Canino), on the contrary, are much more expressive and their movements seem to accompany all their performances. Many other factors may be involved in different mimic expressive patterns concerning the ability of piano players – amateur piano players are often much more concentrated than experts on their hand movements and their expressions may be related to their fear to make mistakes – ,the nature and structure of the pieces (romantic repertoire, post tonal music, jazz, and so on), the presence-absence of public (concerts, trials, private auditions), performances by piano solo or with other instruments, performances where interpreters read the music score or where they play by hearth. Researches on interpreters’ expressiveness have mainly concerned musical, temporal and acoustic parameters (Gabrielsson and Juslin, 1996) and only occasionally non verbal clues of interpreters have been taken into consideration. Facial expressions and other body changes have been linked with memory strategies (Rubin-Rabson, 1939) and in the preparation of music performance (Miklaszewski, 1989). Individual differences among interpreters and difficulties in generalizing data may have hampered the interest of scientific research towards the relationship between performance and mimic expression. Furthermore in piano performance many teachers have stressed the importance of a correct posture suggesting how useful it can be for an interpreter to avoid unnecessary movements; movements control strategies have often been indicated as important tools to gain best interpretative results. Only in some special contexts, such as music therapy, body and facial expressions have been considered as important communicative elements. It seems worthwhile, despite all the difficulties, to see how communicative and non communicative aspects play a role in a piano interpretation; these aspects may be well related to mimic elements. Following a fruitful and long research tradition on nonverbal behaviour we could see how communicative or simply inferential are the nonverbal expressions in the context of a piano performance. Some studies on conductors’ face and gestures (Poggi, 2001) can be regarded as an useful reference in order to create a grid where one can describe the non verbal elements according to their communicative functions. The classification of facial movements according to Ekman and Friesen (1978) Facial Action Coding System (FACS) may be regarded as another important point of reference. FACS users may describe completely all the visible movements of face. Each single movement is called “action unit” (AU). Using FACS we may describe correctly single facial clues coming from different expressive contexts. The relationship between music elements and mimic expression finally is another topic which is interesting to study more in detail. A comparison between music analysis and mimic analysis can be done in order to get more information on the interpreter’s style and his/her communicative tools. Furthermore this comparison may be useful to get inferential information about the way an interpreter analyses the musical piece he/she is going to play. Formal analysis of music may be done following different criteria and theoretical models: not all the musical styles share the same formal rules. Furthermore analysis done on the musical score may be different by an analysis done listening to a performance: an interpreter may underline some points, unify or divide others and the analytic units may be different according to the performances. Perceptual criteria may be used in music analysis: change or discontinuity (Imberty, 1993), similarity and dissimilarity of musical material (Deliège, 2001; Deliège and El Ahmadi, 1990; Deliège and Melen, 1997) in a listening situation can be taken into consideration to pinpoint different units in a musical piece. An attempt to identify the reasons why a fragment is heard as a unit containing musical continuity on the basis of perceptual cues has been done in most perceptual analytical models. Tension and relaxation can be seen as predictive of segmentation (Addessi and Caterina, 2000).