Michiteru Kitazaki

Providing convincing walking experiences in VR is important for embodied interaction, yet walking is often performed with others and the role of social presence remains unclear. We investigated whether virtual co-walkers facilitate or inhibit seated multisensory virtual walking induced by optic flow and synchronized foot vibrations. Twenty-four participants experienced a first-person walking self-avatar under four within-subject conditions: synchronous-walking others, asynchronous-walking others, non-walking others (forward translation without gait), and no others. After each trial, they rated walking-related sensations, telepresence, and motivation. Synchronous-walking others enhanced self-motion sensation, leg-action sensation and telepresence compared with no others, whereas non-walking others reduced all walking-related ratings and increased pressure/tension. Synchrony between the participant and others had no significant effect. These results suggest that co-walking avatars can augment virtual walking experiences, but behaviorally implausible motion may undermine them in multi-agent VR locomotion.

Extending virtual arms enables beyond-reach interaction, but the kinematics of extension may critically shape embodiment. We investigated whether a human-like extension profile based on the minimum-jerk model enhances body ownership and sense of agency compared with commonly used profiles. Twenty-eight participants performed a reaching task in VR using a self-extending virtual arm under a 3 x 2 within-subject design: extension profile (constant velocity, accelerated, minimum-jerk) and target outcome (full-success vs. partial-success at 50%). After each condition, participants rated ownership, agency, and usability (SUS). Repeated-measures ANOVAs revealed significant interactions for ownership and agency: under full success, minimum-jerk extension produced higher ownership and agency than accelerated extension. SUS showed a strong main effect of success rate. These findings highlight that human-like motion design improves embodiment when action outcomes are reliable.

Robotic massage systems can reproduce treatment forces consistently, but most still rely on predefined trajectories and fixed force distributions that ignore recipients’ subjective sensations. We propose a conversational robotic massage system that updates a polynomial force map for latissimus dorsi acupressure from recipient speech and applies it through force feedback control. In a study with 13 participants, simple commands such as "little weaker" and "little stronger" changed the applied force in the intended direction; questionnaires indicated high force adaptation, satisfaction, and consistency, with low pain, while perceived safety varied widely, suggesting a trade-off between interactive adaptation and predictable contact. These findings indicate that speech is a promising modality for tailoring robotic massage and highlight the need for careful gain tuning and feedback about force changes. Although the system currently uses only four predefined speech patterns, it represents an initial step toward conversational control of spatially varying acupressure forces.

The human sense of self is plastic and can be modulated by multisensory and sensorimotor coupling in immersive virtual reality. In particular, synchronous visuo-tactile stimulation and visuo-motor coupling via first-person control are widely used to induce embodiment. However, when these two streams promote different candidate self-locations, their relative contributions to the perceived self-location remain unclear. In this study, we examined perceived self-location using a robot-mediated self-touch paradigm that creates a cue conflict: synchronous visuo-tactile stimulation in this paradigm tends to bias self-location toward the participant’s physical body, whereas visuo-motor coupling via first-person control tends to bias self-location toward a remote robot viewpoint. We compared this conflict condition with control conditions in which either the visuo-tactile or visuo-motor stimulation was attenuated. Perceived self-location was assessed using questionnaire ratings and a behavioral estimate, namely, the Self-Location EstimatEd Pointing (SLEEP), which estimates participants’ perceived self-location using a finger-pointing task in a 360 degree image. Overall, the results are consistent with a stronger contribution of synchronous visuo-tactile stimulation than visuo-motor coupling to perceived self-location in this whole-body embodiment setting.

Autonomous robotic prostheses can aid individuals with limb loss regain functionality and near-normal appearance. However, to psychologically integrate such limbs into one’s body image, it is necessary to understand how movement characteristics, such as movement speed, affect the sense of embodiment. Using a VR simulation, we investigated how the speed of autonomous lower-arm movements affects embodiment and user-perception. Nineteen healthy participants embodied an amputated virtual avatar with a prosthetic lower left-arm moving autonomously at six different speeds following minimum-jerk trajectories. Participants rated sense of body ownership, sense of agency, usability, competence, warmth, and discomfort after performing a reaching task at each speed. Ownership, sense of agency, and usability were highest for moderate speeds (autonomous movements lasting 1 s), and were significantly lower for both faster (125 ms) and slower (4 s) movements ( p < 0.05). Competence was significantly higher at moderate and moderately fast speeds compared to slower speeds. Discomfort was significantly higher at the fastest speed compared to moderate and slower speeds. Overall, the results show a tendency of moderate movement speeds being favorable for user perception of autonomous limbs and hint at the existence of an optimal speed or a speed range for enhancing embodiment and usability.

This paper presents a millimeter-scale piezoelectric effect-based force sensor that uses the change in its resonant frequency as the detection principle for high sensitivity and a wide measurement range. Such characteristics are suited for robot hand applications that can not only detect a small force but also handle large payloads. In this paper, we develop a methodology to estimate the relationship between the applied force and the resonant frequency shift by combining classic contact theory and the finite element method (FEM) analysis. Although this relationship is non-linear, the designability of the sensitivity and measurement range is demonstrated by the simulation. The simulation results based on the method are verified, showing good agreement with experimental results. The static characteristics, including sensitivity, standard deviation, and resolution, are evaluated using the prototype sensors with characteristic lengths ranging from 1mmto 4 mm. The 4-mm model has a measurement range of 77 mN to 300 N, and the smallest model, which is one of the smallest force sensors suitable for practical implementation, has a measurement range of 9 mN to 20 N.

In human–robot interaction (HRI) involving physical contact, establishing user trust is critical—especially in applications such as robotic massage, where direct tactile stimulation can provoke anxiety. This study investigates how the visual appearance of virtual avatars influences users’ psychological and physiological responses, with the goal of identifying design strategies for enhancing trust. We developed a Virtual Reality (VR) massage system that synchronizes a robotic arm with a virtual therapist avatar and examined six avatar conditions varying in human-likeness, body size, and gender. Subjective ratings and physiological signals were collected under each condition. The results suggest three observations. First, trust is not determined solely by the avatar’s appearance but is significantly enhanced when the tactile experience is consistent with the expectations induced by that appearance. Second, trust comprises two distinct components—social and functional—which must be explicitly defined in design. Third, the avatar’s appearance affected both subjective ratings and autonomic responses, including CCV-TP-based heart rate variability, indicating that visual expectations can modulate users’ perception of physical sensations. These findings contribute to both academic understanding and practical design of HRI systems involving physical interaction, such as those used in remote healthcare and assistive robotics.

This study investigates how both the body part used to control a VR avatar and the avatar's appearance affect redirection detection thresholds. We conducted experiments comparing hand and foot manipulation of two types of avatars: a hand-shaped avatar and an abstract spherical avatar. Our results show that, irrespective of the body part used, the redirection detection threshold increased by 21% when using the hand avatar compared to the abstract avatar. Additionally, when the avatar's position was redirected toward the body midline, the detection threshold increased by 49% compared to redirection away from the midline. No significant differences in detection thresholds were observed between the hand and foot manipulations. These findings suggest that avatar appearance and redirection direction significantly influence user perception in VR environments, offering valuable insights for the design of full-body VR interactions and human augmentation systems.