Enhancing the experience of virtual reality (VR) through haptic feedback could benefit applications from leisure to rehabilitation and training. Devices which provide more realistic kinesthetic (force) feedback appear to hold more promise than their simpler vibrotactile counterparts. However, our understanding of kinesthetic feedback on virtual embodiment is still limited due to the novelty of appropriate kinesthetic devices. To contribute to the line of this research, we constructed a wearable system with state-of-the-art kinesthetic gloves for avatar full-body control, and conducted a between-subjects study involving an avatar self-touch task. We found that providing a kinesthetic sense of touch substantially strengthened the embodiment illusion in VR. We further explored the ability of these kinesthetic gloves to present virtual objects haptically. The gloves were found to provide useful haptic cues about the basic 3D structure and stiffness of objects for a discrimination task. This is one of the first studies to explore virtual embodiment by employing state-of-the-art kinesthetic gloves in full-body VR.
Kinesthetic interaction between the user and the computer mainly utilizes the hand-based input with force-feedback devices. There are two major shortcomings in hand-based kinesthetic interaction: physical fatigue associated with continuous hand movements and the limited workspace of current force-feedback devices for accurately exploring a large environment. To address these shortcomings, we developed two interaction techniques that use eye gaze as an additional input modality: HandGazeTouch and GazeTouch. HandGazeTouch combines eye gaze and hand motion as the input for kinesthetic interaction, i.e. it uses eye gaze to point and hand motion to touch. GazeTouch replaces all hand motions in touch behavior with eye gaze, i.e. it uses eye gaze to point and gaze dwell time to trigger the touch. In both interaction techniques, the user feels the haptic feedback through the force-feedback device. The gaze-based techniques were evaluated in a softness discrimination experiment by comparing them to the traditional kinesthetic interface, HandTouch, which only uses the hand-based input. The results indicate that the HandGazeTouch technique is not only as accurate, natural, and pleasant as the traditional interface but also more efficient.
Kinesthetic interaction typically employs force-feedback devices for providing the kinesthetic input and feedback. However, the length of the mechanical arm limits the space that users can interact with. To overcome this challenge, a large control-display (CD) gain (>1) is often used to transfer a small movement of the arm to a large movement of the onscreen interaction point. Although a large gain is commonly used, its effects on task performance (e.g., task completion time and accuracy) and user experience in kinesthetic interaction remain unclear. In this study, we compared a large CD gain with the unit CD gain as the baseline in a task involving kinesthetic search. Our results showed that the large gain reduced task completion time at the cost of task accuracy. Two gains did not differ in their effects on perceived hand fatigue, naturalness, and pleasantness, but the large gain negatively influenced user confidence of successful task completion.
8 December 2025 in Human-Intelligent Systems Integration
The application of Large Language Models (LLMs) in academic research faces unique challenges of privacy and workflow integration. This paper introduces TAUCHI-GPT, a novel, open-source AI assistant whose evolution informs our analysis. We detail its two versions: a cloud-based V1 using GPT-4 and reflection cycles, and a local, privacy-preserving V2 with RAG architecture. Based on empirical findings from two user studies, we present a critical Human-System Integration (HSI) analysis of the security vulnerabilities and alignment challenges inherent in local LLM deployments. We examine how recent development trends—such as model distillation and reward-model learning—and the complexities of internal model mechanisms exacerbate risks like prompt injection, RAG data failures, and unfaithful explanations that impact user trust. Drawing from HCI principles and mechanistic interpretability insights, we propose and discuss a multi-layered mitigation strategy. This work contributes significantly to HSI and AI by presenting an evaluated system, a rigorous analysis of local deployment risks from a sociotechnical perspective, and actionable, stakeholder-specific guidelines for the secure and responsible utilization of LLMs in academia.
Kinesthetic interaction is an important interaction method for virtual reality. Current kinesthetic interaction using a grounded force-feedback device, however, is still considered difficult and time-consuming because of the interaction difficulty in a three-dimensional space. Velocity-oriented dynamic control–display (CD) gain has been used to improve user task performance with pointing devices, such as the mouse. In this study, we extended the application of this technique to kinesthetic interaction and examined its effects on interaction speed, positioning accuracy and touch perception. The results showed that using this technique could improve interaction speed without affecting positioning accuracy in kinesthetic interaction. Velocity-oriented dynamic CD gain could negatively affect touch perception in softness while using large gains. However, it is promising and particularly suitable for kinesthetic tasks that do not require high accuracy in touch perception.