Previous research has shown that rigidity in interpersonal coordination (too muchalignment) is negatively associated with creativity as measured by ratings of aesthetic appeal. In this study, we examine the signatures of music improvisation following a targeted manipulation involving a mirroring task. In 18 pairs, participants completed a series of tasks: a solo music improvisation performance using a percussion instrument (cajón), a mirroring task with a partner, and joint music improvisation. Across pairs, we manipulated three different types of mirroring to examine its effects on coordination during joint music improvisation. A third of the pairs engaged in Hierarchical mirroring, with partner A leading the movement and partner B following. Another third of the pairs engaged in Turn-Taking during mirroring, with partner A leading the first half of this phase and partner B leading the latter half. The final third of the pairs engaged in Egalitarian mirroring, with partners co-creating spontaneous movement together. Partners were video and audio recorded during this phase. From these recordings, we have extracted signatures of interpersonal coordination in terms of acoustic properties of the performance, using cross-correlation. We predicted that the mirroring conditions involving more rigidity (i.e., more asymmetrical roles) would be associated with more rigidity (i.e., more alignment) during music improvisation. Specifically, We predicted that the opportunity to take the lead during motor mirroring would impact coordination during improvisation as follows: pairs in which both partners had the opportunity to lead (i.e., in Turn-Taking) or co-lead (i.e., in Egalitarian mirroring) would exhibit higher behavioral complementarity (less alignment) compared to those pairs in which one partner had solely taken the lead (i.e., the Hierarchical mirroring). Measures assessing individual differences in music sophistication, personality, and prosocial orientation were also collected. Regression models with cross-correlation metrics as outcome measures displayed marginally significant differences between the Turn-Taking and Egalitarian mirroring conditions with the Turn-Taking condition displaying lower levels of cross-correlation. Regression models were also built with individual differences measures as covariates suggesting that musical ability may influence music improvisation. However, a study with a large sample size will allow for a more accurate assessment of the impact of partners’ movement patterns in a subsequent performance, thus providing theoretical insights about improvisation and collaboration.