UCF researchers are partnering with Meta Platforms Inc. to study how people learn to control digital systems using muscle signals, work that could improve human-computer interaction in virtual and augmented environments.
Supported by a gift from Meta, the two-year project uses electromyographic (EMG)-based human-machine interface technology as a platform for investigating motor learning through gamified training systems. While EMG systems are often studied in the context of prosthetic limb control, the broader goal of the project is to understand how adaptive interfaces can become more intuitive and embodied over time.
“This Meta support will enable my lab to work on real-world problems that can have an immediate impact on neurotechnologies.” — Mohsen Rakhshan, assistant professor
UCF was selected through Meta’s competitive funding initiative, in part because of its interdisciplinary approach pairing engineering with philosophy and ethics.
Mohsen Rakhshan, an assistant professor in UCF’s Department of Electrical and Computer Engineering and the Disability, Aging and Technology (DAT) faculty cluster initiative, and Jonathan Beever, a professor of philosophy and director of the UCF Center for Ethics, will lead the project.
“This Meta support will enable my lab to work on real-world problems that can have an immediate impact on neurotechnologies,” Rakhshan says. “The impact ranges from individuals using augmented and virtual reality for entertainment to individuals with amputation or paralysis seeking to improve their quality of life. It also gives my engineering students the opportunity to integrate ethics research into their technical work.”
Advancing Motor Learning Through EMG
EMG-based interfaces translate electrical signals generated by muscle activity into digital commands, allowing users to control devices through subtle physical gestures. In immersive environments, these systems can enable more natural interaction with virtual objects. In rehabilitation settings, they can assist in training neural prostheses.
The UCF team is using this technology to examine how people learn new motor skills in digital environments, particularly through gamified interaction tasks designed to strengthen human-computer coordination. By training both the participant and the signal-processing algorithm (often called a “decoder”) simultaneously, through a process known as co-adaptation, researchers aim to create systems that improve alongside the user.
“A significant challenge for most of these systems is that they require constant retraining or calibration of the decoder,” Rakhshan says. “Retraining after each use can discourage individuals from using these devices long term. The human nervous system is plastic — it can adapt and improve performance over time. But if the decoder is constantly reset or kept static, it may prevent the nervous system from leveraging that plasticity. We aim to develop a co-adaptive loop between the human and the device.”
Rather than focusing solely on stable decoding, the project investigates how adaptive systems can enhance motor learning, improve user confidence and promote a stronger sense of embodiment in human-machine interaction.
If successful, the research could inform next-generation EMG systems used in immersive computing, rehabilitation technologies and assistive devices.
Embedding Ethics Into Engineering
A defining feature of the project is the integration of ethics alongside engineering from the outset.
“Interdisciplinary collaboration between ethics and technical experts is the best path forward for responsible innovation.” — Jonathan Beever, professor
Longitudinal EMG studies can reveal subtle motor signatures that uniquely identify individuals, raising questions about privacy and data protection. Adaptive systems may also influence a user’s sense of agency, whether individuals feel genuinely in control of the interface. For example, if an EMG system begins adjusting its interpretation of muscle signals automatically, users may feel the device is responding to them intuitively or, in some cases, acting unpredictably. Researchers want to better understand how these dynamics affect trust, confidence, and long-term use.
To address these questions, Beever will be embedded within the UCF Laboratory for Interaction of Machine and Brain (LIMB), contributing directly to experimental design and evaluation. The team will conduct structured assessments of agency and embodiment while examining potential privacy leakage from EMG signal data.
“Interdisciplinary collaboration between ethics and technical experts is the best path forward for responsible innovation,” Beever says. “Technological advancement must be guided toward good ends. Our work emphasizes not only ethical research practices but also deeper questions about autonomy and agency in human-machine interfaces.”
A Three-Phase Study
The longitudinal study will involve 30 participants completing 10 sessions over two months, allowing researchers to measure both short-term and long-term motor learning outcomes.
The project will occur in three phases:
Phase 1: Standardizing muscle signal data so artificial intelligence systems can more accurately interpret user intent.
Phase 2: Training both participants and machine learning models simultaneously — a co-adaptive process designed to improve human-computer interaction through gamified tasks.
Phase 3: Conducting structured evaluation of agency, embodiment and privacy risks while developing a publishable ethics framework for adaptive EMG-based systems.
“There has been a significant increase in industry interest in using biological signals such as EMG, from muscles, and EEG, from the brain, to interact with virtual and augmented reality, consumer electronics, prostheses for individuals with amputation and robotic systems for individuals with paralysis,” Rakhshan says.
This research is supported by a gift from Meta. The project is conducted by faculty, staff and students in UCF’s Department of Electrical and Computer Engineering, the Disability, Aging and Technology research cluster and the UCF Center for Ethics.
