I am going to present my latest results at SysMus21 (Systematic Musicology) between the 3rd and the 5th of November in Aarhus, Denmark 🌵
Accepted abstract (will be updated in September): “What makes musicians infer teaching intentions?”
We investigated which features of piano performance make musicians infer teaching intentions. Musicians listened to a number of piano recordings where a musical expressive technique of either articulation or dynamics was implemented. Musicians were asked to judge whether each performance was produced in order to teach the designated expressive technique. We quantified performances with regard to tempo, articulation and dynamics. Overall, slower tempo contributed to musicians’ teaching judgements. Moreover, performances with exaggeration for each technique (e.g., longer legato, shorter staccato, larger contrast between forte and piano) were more likely to be judged as teaching.
Keywords: teaching, skill transmission, musical expression
Social learning plays a key role in human skill transmission. It has been known that experts tend to change their behaviour specifically for teaching purposes. Our previous studies revealed that expert pianists successfully modulated their sounds so as to teach musical expressive techniques such as articulation and dynamics. For example, pianists tended to play slower and exaggerate each technique when they had an intention to teach. In order to achieve successful skill transmission, it is important to know whether such sound modulations are identified and used to infer teaching intentions by novices, who are going to acquire skills.
The aim of the current study was to investigate which features of piano performance make musicians (in the role of potential learners) infer teaching intentions, only by listening to recorded performances.
20 participants, who had on average 11.85 years of training in any musical instrument, were included for data analysis. Participants listened to a number of piano recordings and were asked to judge whether each recording was performed for teaching purposes (e.g., in a lesson / teaching) or as a part of a performance (e.g., in a concert / performing). They responded by pressing either a yes (teaching) or no (performing) button. One musical excerpt was selected as a stimulus, which was performed by different pianists with two different musical expressive techniques. Half of the recordings was performed with articulation, which consisted of legato and staccato whereas the other half was performed with dynamics, which consisted of forte and piano. We characterised performances with regard to tempo (interonset intervals), articulation (key overlap time), dynamics (key velocity) and dynamics contrast (key velocity difference).
We looked at correlations between performance features and participants’ teaching judgments. As for tempo, slower performances were more likely to be judged as teaching performances regardless of the techniques. For articulation recordings, performances with longer legato and shorter staccato were tended to be judged as teaching. For dynamics recordings, performances with louder forte were tended to be judged as teaching whereas there was no correlation between how softly the piece was played (i.e., piano) and teaching judgments. Moreover, performances with larger contrast between forte and piano were more likely to be judged as teaching.
Our findings suggest that musicians are able to infer teaching intentions by relying on specific performance features. In general, slower performances were likely to be identified as teaching. Also, performances with exaggeration (e.g., longer legato, shorter staccato, larger contrast between forte and piano) were tended to be judged as teaching. These performance features are consistent with expert pianists’ sound modulations for teaching. Therefore, musicians (in the role of potential learners) seem to be able to identify didactic sound modulations, only by listening to recorded performances.