Fluifiber: Interweaving Dancing Bodies over Time and Space via Programmable Tensegrity

An embodied approach to movement archiving, learning and transfer

Forms of movement such as dance, music making, calligraphy involves dexterous movement techniques that combine performance with expression and aesthetics. Nevertheless, little is understood about complex hand dexterity of a virtuoso, or how danseur expertise is acquired, due to the versatility of movement combinations and learning techniques available to complete any given task.

Watching someone else perform is a common approach to observational learning. One of the reasons that observational learning has been found incomplete by a long history of research is that there are important dimensions of the movement that are unavailable to the subject's view; such as pressures, muscular tensions, and external features of the movement that cannot be seen. However, observational learning may not be destined to be incomplete forever as ingenious ways might be devised to make the unobservable observable, the absent present.

In a recent study, Rivière et al. investigated dance movement acquisition by relying on the learners' (the dancers) experiences to better understand different techniques used by dancers during their training. As a result, they identified several concepts as techniques of learning that are observation, repetition, imitation, marking, segmentation, mental simulation and personal adaptation. The latter category is learning progression, consisting of phases such as analysis, integration, fluidity, personalization and implicit variations.  These techniques, highlighted by interviewed dancers, were interpreted by the authors as a set of tools that operate as variations on the initial movement.  Previous work in motor skill learning has shown the benefit of variation on retention and skill transfer. (Adams, 1987) 

Fluifiber: A synchronized tangible representation of the dexterous body

R. Buckminster Fuller, the father of tensegrity, defines it as “Islands of compression inside an ocean of tension.” Tensegrity has been used as a solution in various fields such as molecular biology for modeling virus structures to understanding some motion mechanism depending on the substrate the living cell moves on. (Ingber, 2003) In recent years efforts have aimed to establish links between tensegrity systems, muscular-tendons-ligaments apparatus in vertebrates and mechanisms of locomotion to better comprehend living systems. Previously, researchers at McGill University have developed a wearable spine tensegrity instrument that creates sounds in tune with the movements of the body.

A tensegrity system, due to their resemblance to the biological body in components and force distribution, can be utilized as a dynamic tangible representation of the moving body. Such model may enable the novice or an expert dancer to externalize and perceive their own performance and expressions, or that of another dancer's.  In this work, we explore  whether such externalized physical model of movement can be used as a learning support technique, that helps bring the embodied memory of the performance back to the motor neurons and musculoskeletal system. 

Intrigued by tensegrity systems' inherent ability to contain within the structural framework needed for stability with little weight and thin parts while allowing richness in movement; we explore a programmable tensegrity framework consisting of sensorized fluid-actuated fiber muscles "fluifibers" as tendons, and static struts that together capture the movements of the dancer and play back at desired speed and sequences for observation, imitation, segmentation of the movement. For example, a dancer may want to deconstruct or refine an integrated movement by iterating between analysis and integration steps utilizing the feedback from the closed-loop tensegrity as a perceptual trace.

An important aspect derived from research on motor skill acquisition is that there is a wide variety of techniques used by humans, for learning a new movement or skill and that it mainly relies on practice involving explicit and implicit variations of the same movement. We believe that this can be afforded by such an interactive system like ours that allows for appropriation. (Dix, 2007)

 We might not be able to design for the unexpected but we can design to allow the unexpected.

"Presence of Absence" and Fluifiber

We hypothesize that the nervous system, given abstract information in the form of a tangible dancing tensegrity model, will try to compensate for the missing (absent) information in an endeavor to understand it.  If our hypothesis holds true, our interface may be utilized for kinesthetic learning activities in remote scenarios where other telecommunication modalities remain insufficient for transfer of bodily skills. We envision expanding our work to different form factors that afford wearability such as a structural tensegrity skin, as well as supporting different forms of movement than dance, such as the transfer and retention of the  tacit knowledge of a luthier (e.g. violin maker) across generations(time) and distances(space).

Contributors

Ozgun Kilic Afsar and Hiroshi Ishii