Preserving Agency During Electrical Muscle Stimulation Training Speeds up Reaction Time Directly After Removing EMS
Shunichi Kasahara, Kazuma Takada, Jun Nishida, Kazuhisa Shibata, Shinsuke Shimojo, and Pedro Lopes
ACM CHI 2021
Collaborative projects with UChicago Human Computer Integration Lab (Pedro Lopes, Jun Nishida) , Kazuhisa Shibata (RIKEN CBS) and Shinsuke Shimojo (Caltech)
Paper (ACM Digital Library) · PDF · Source code
Abstract
Force feedback devices, such as motor-based exoskeletons or wearables based on electrical muscle stimulation (EMS), have the unique potential to accelerate users’ own reaction time (RT). However, this speedup has only been explored while the de- vice is attached to the user. In fact, very little is known regarding whether this faster reaction time still occurs after the user removes the device from their bodies–this is precisely what we investigated by means of a simple reaction time (RT) experiment, in which participants were asked to tap as soon as they saw an LED flashing. Participants experienced this in three EMS conditions: (1) fast-EMS, the electrical impulses were synced with the LED; (2) agency-EMS, the electrical impulse was delivered 40ms faster than the participant’s own RT, which prior work has shown to preserve one’s sense of agency over this movement; and, (3) late-EMS: the impulse was delivered after the participant’s own RT. Our results revealed that the participants’ RT was significantly reduced by approximately 8ms (up to 20ms) only after training with the agency-EMS condition. This finding suggests that the prioritizing agency during EMS training is key to motor-adaptation, i.e., it enables a faster motor response even after the user has removed the EMS device from their body.
Does a speedup last after the device comes off?
Wearable force feedback devices, such as exoskeletons and electrical muscle stimulation (EMS), can move a user’s body, and they have been used to help people learn physical skills. EMS can even move a finger faster than the user’s own reaction, and when it fires within a certain time window before the user’s own movement, the user still feels that they made the movement (Preemptive Action). All of this, however, has been studied while the device is on the body. We asked whether people who were accelerated by EMS during training are still faster after they remove it.
A movement driven by EMS differs from voluntary practice in one respect: the user is moved rather than moving, so the sense of agency over the movement is weakened or lost. Our hypothesis was that agency decides whether the speedup carries over: without feeling that we performed an action ourselves, we are unlikely to get better at it.
Finding the timing that keeps agency
Participants rested a hand in a holder and tapped an aluminum bar with the ring finger as soon as an LED lit up. A capacitive sensor measured the reaction time with less than 1 ms of latency, and electrodes on the muscle that flexes the ring finger let EMS make the same tap.
In a first study, 18 participants did this task while the EMS fired at random times, from 100 ms before to 300 ms after the LED, and after each tap they rated whether they did it, from 1 (“I did not do it”) to 7 (“I did it”). Their own reaction time averaged about 204 ms. We fitted each participant’s ratings against the preemptive gain, how much earlier than their own reaction time the EMS made them tap. With 40 ms of preemptive gain, 15 of the 18 participants kept a sense of agency above the midpoint of the scale, so 40 ms became the agency-preserving timing for the main study.
Three ways to train with EMS
In the main study, 17 of the same participants trained under three EMS conditions:
- fast-EMS: the EMS moved the finger the moment the LED lit up, giving very fast taps with little or no sense of agency.
- agency-EMS: the EMS moved the finger 40 ms before the participant’s own reaction time, a speedup that preserves some agency.
- late-EMS: the EMS came only after the participant’s own reaction time, so participants tapped by themselves with full agency but no speedup.
Comparing the three separates two things a lasting speedup could come from: having been moved fast (fast-EMS) and having felt the movement as one’s own (late-EMS). Only agency-EMS combines both. Each condition took place on a different day, in counterbalanced order with three days in between. Reaction time was measured without EMS before training and after each of two EMS training blocks.
Only agency-preserving training left people faster
Participants were faster after the EMS was removed only when the training had preserved their sense of agency: after agency-EMS their reaction time without EMS was about 8 ms shorter (up to 20 ms), while training that sped them up without agency (fast-EMS) or kept agency without a speedup (late-EMS) left no significant gain.
The gain after agency-EMS (8.1 ms on average) differed significantly from zero, and the gains after fast-EMS (3.9 ms) and late-EMS (−0.9 ms) did not. Across the three conditions the gains differed significantly, with agency-EMS larger than late-EMS.
Eight milliseconds came from only 30 or 60 trials of EMS training; in earlier studies, three weeks of practice shortened simple reaction time by about 20 ms. In a fast tennis serve, the ball travels about 0.5 m in 8 ms.
Why agency matters for adaptation
The interviews show how differently the conditions felt. After fast-EMS, almost all participants said they felt the EMS much earlier than their own movement (“I felt that my finger was completely actuated by EMS”), and three described losing motivation (“I lose my motivation to move by myself because I know EMS will exceed my speed”). After agency-EMS, participants described the EMS as arriving together with their own movement, although it actually came earlier: “I felt that I could touch faster even with my own touch, it was positive for me.”
These comments indicate that too much assistance is not beneficial, as research on motor learning has also found. The results fit our hypothesis: when the faster movement still felt like their own, participants adapted to it, and the speedup remained after the EMS was gone. The sense of agency appears to play a crucial role in adapting to a faster reaction.
From training to applications
The findings fit interfaces where a short-term speedup helps and the system does not need to predict complex decisions. One example is a video game “power-up” that, instead of making the player’s character shoot faster, makes the player’s own finger faster with preemptive EMS; collecting power-ups then works as a training session. Another is sprint training with EMS on the legs to keep a faster start. Open questions are whether the effect extends to exoskeletons and other force feedback devices, whether longer training consolidates it, and how agency plays out when users have to make decisions while being accelerated.
Talk at CHI 2021
Reference
Shunichi Kasahara, Kazuma Takada, Jun Nishida, Kazuhisa Shibata, Shinsuke Shimojo, and Pedro Lopes. 2021. Preserving Agency During Electrical Muscle Stimulation Training Speeds up Reaction Time Directly After Removing EMS. In Proceedings of the 2021 CHI Conference on Human Factors in Computing Systems (CHI ’21). Association for Computing Machinery, New York, NY, USA, 1–9. https://doi.org/10.1145/3411764.3445147
BibTeX
@inproceedings{10.1145/3411764.3445147,
author = {Kasahara, Shunichi and Takada, Kazuma and Nishida, Jun and Shibata, Kazuhisa and Shimojo, Shinsuke and Lopes, Pedro},
title = {Preserving Agency During Electrical Muscle Stimulation Training Speeds up Reaction Time Directly After Removing EMS},
booktitle = {Proceedings of the 2021 CHI Conference on Human Factors in Computing Systems},
series = {CHI '21},
year = {2021},
pages = {1--9},
publisher = {Association for Computing Machinery},
address = {New York, NY, USA},
doi = {10.1145/3411764.3445147}
}Figures from the paper (Kasahara et al., CHI ’21). © 2021 Copyright held by the owner/author(s). Publication rights licensed to ACM.