Embedded vs. Situated: An Evaluation of AR Facial Training Feedback

2026-02-01

AI Summary

This study investigates effective approaches to delivering real-time augmented reality (AR) feedback for facial muscle training, with a focus on balancing training accuracy, cognitive load, and user experience. We designed and systematically compared three AR feedback modalities that differ in spatial relationship to the user: screen-situated, proxy-embedded (on a dummy face), and fully embedded (directly on the user’s own face). Through a within-subjects experiment, we evaluated these modalities across multiple dimensions, including movement precision, cognitive load, and user preference. Results indicate that embedded feedback significantly reduces cognitive load and is strongly preferred by users, whereas situated feedback enables higher accuracy in movement correction. This work reveals, for the first time, a design tension in AR-based facial feedback between enhanced user experience and the challenges of self-awareness and interpretability, and proposes design guidelines for applications in rehabilitation and performance training.

Technology Category

• Cognitive Modeling & Cognitive Systems: Affective Computing
• Humans and AI: User Experience and Usability
• Computer Vision: Multi-modal Vision

Application Category

• User Modeling, Personalization and Recommendation: Accountability, Transparency, and Ethics for personalization
• Responsible Web: Machine-in-the-loop, human agency and autonomy
• Search and Retrieval-Augmented AI: Web evaluation methodologies and metrics

Abstract

While augmented reality (AR) research demonstrates benefits of embedded visualizations for gross motor training, its applicability to facial exercises remains under-explored. Providing effective real-time feedback for facial muscle training presents unique design challenges, given the complexity of facial musculature. We developed three AR feedback approaches varying in spatial relationship to the user: situated (screen-fixed), proxy-embedded (on a mannequin), and fully embedded (overlaid on the user's face). In a within-subjects study (N=24), we measured exercise accuracy, cognitive load, and user preference during facial training tasks. The embedded feedback reduced cognitive load and received higher preference ratings, while the situated feedback enabled more precise corrections and higher accuracy. Qualitative analysis revealed a key design tension: embedded feedback improved experience but created self-consciousness and interpretive difficulty. We distill these insights into design considerations addressing the trade-offs for facial training systems, with implications for rehabilitation, performance training, and motor skill acquisition.

Problem

• augmented reality
• facial training
• real-time feedback
• motor skill acquisition
• user interface design

Innovation

• augmented reality
• facial training
• embedded feedback
• situated feedback
• motor skill acquisition