Hi all,
I'd like to share a prototype and ask for some critique of the mapping design.
The task is the ushiro ukemi, the backward breakfall in judo. Done correctly it is one continuous roll that distributes impact: the hips arrive, the rounded back rolls on, and the palms and forearms slap the mat slightly before the back lands. That last interval is the whole technique. It is on the order of tens of milliseconds, and it is both hard to see from outside and hard to feel from inside while you are falling, which makes it an interesting candidate for auditory display.
The prototype lets you select a teaching progression, choose among sonification modes, and play the resulting sequence. It runs in the browser, with no install:
https://liapikou.github.io/ukemi_trainer_2.0/Four questions I would value other opinions on:
1. Event separation. At realistic fall speeds the slap and the back contact fall close together. Do the modes keep them distinct, and does one mode do it better than the others?
2. Interpretability without vision. If you listen cold, does anything about the movement come through? I am wary of having designed something that only makes sense to someone who already knows the answer.
3. Error salience. The interesting condition is the incorrect fall. Does a mistimed slap sound wrong in a way a learner could act on, or only in a way an expert can identify after the fact?
4. Literature. Most movement sonification I know maps continuous trajectory. Here the target is the interval between two discrete impacts. Pointers to work on that framing, from any domain, would be very welcome.
Background: this grew out of a study on real-time sonification for backward breakfalls (N=25) and a paper accepted at the 8th Research and Applications of Judo conference. I am interested in whether the approach generalises to other short, high-risk motor skills where the critical variable is timing between impacts.
The pedagogical side has already been through several judo coaches. What the work has not had is scrutiny from the audio side, which is what I am asking for here.
I will post a summary of any responses back to the list.
Thank you for your time.
Best regards,
Eirini Liapikou
Audio and Acoustics Engineer
Aalborg University, Copenhagen