This paper presents an experiment to demonstrate that social stimulus (changes in facial expressions) can generate haptic sensations in a human-agent interactions setting. We designed and developed a system in which an human-like agent on the display behaved as if it pulled the mouse cursor against a human participant control. The results of the experiment showed that participants felt significantly larger force when the agent exhibited the negative face during the pulling motion than that when the agents kept the positive face. Besides, the force sensations quantified with control-display ratio negatively correlated with some of sub-scales of the Autism-spectrum Quotient score. This indicates that the social stimulus is as important as physical stimulus when designers and researchers design human-agent haptic interactions.

It is still unknown how the viewer perceives the self-body in virtual space. In this study, we investigated the differences in performance of a simple motor task in a virtual space when the visible type of the 3D self-body model was changed to one of three types: “the whole body visible,” “the body partially visible,” or “the body invisible.” The quantitative data of the motor were used to analyze the durations and accelerations of the motor task. The results showed that the duration of the motor task varied depending on the visibility of the body models. The results also suggested that visual body information, particularly the self-body, continuously provides a potential interference to the smoothness and quickness of human motor control.

Training with distorted vision has been shown to change posture information in a body representation. We investigated an interaction in a change of symmetric body parts in the body representation, both upper limbs. Two cues that made clear the positional relationship between both hands were tested to whether these facilitated the representation change: (1) whether the participants move both hands simultaneously, and (2) whether the participants move both hands to the same point. The results of our experiments showed that the representation of both hands was changed with neither cue but not facilitated by either of the cues.

Episodic memory is known to be environmental context-dependent. The most famous effect is the reinstatement effect: retrieval works better in the same context (SC) as encoding than in a different context (DC). While various environmental context manipulation methods have been examined, we propose a new method using 360-degree video and self-avatar. Comparing the number of words recalled between the SC condition and the DC condition in a within-participant design, there was no reinstatement effect. The result was consistent with the previous works using immersive virtual environments as context. In addition, we found a negative correlation between the presence and the number of words recalled, but it didn’t reach statistical significance. Also, we revealed a time-compression effect of VR only in the condition using self-avatars relevant to virtual environments, not in the condition using a neutral avatar. These findings would be important for the various VR applications.

Having a self-body recognition for avatars with human augmentation is important to improve the user experience. In this study, we examine the self-body recognition for a four-armed avatar operated by two people. The arm directly operated by the subject is called the main arm, and the arm operated by another subject (the experimenter) is called the subarm. Questionnaires and instantaneous heart rate were used as indices of self-body recognition. The heart rate results showed that the subject’s self-body recognition for the subarm emerged when the subarm moved to meet the subject’s goal. This suggests that it is important for the extended body part of a human augmentation avatar to move in the manner that shares the intention with the subject during operation to recognize the avatar’s body as their own.

Bouldering is a sport where competitors climb an artificial wall using holds as support. The climbing route from start to end is called a problem and each problem has its difficulty grade. As the grade of a problem plays an essential role in measuring climber’s ability, a unified standard is desirable. However, measuring the grade still currently relies on personal experience of the routesetters, which results in huge individual difference. The purpose of this study is to evaluate the grade of a bouldering problem in a uniformed manner. Specifically, we propose a two-step machine learning model that takes into account the difficulty and arrangement of each single hold. Experimental results using 11906 problems show that our model has achieved up to 65.5% of accuracy for classifying 13 classes of problem grades.

We aimed to investigate whether illusory embodiment occurs in arms that stretch farther than usual arms in a virtual environment and how embodiment is acquired in different stretching methods. We realized three types of arm stretching methods: the hand extended ten times the actual movement (constant stretching), the hand extended in an accelerated manner (accelerated stretching), and the hand immediately extended to the target when the user fixed the eyes on the target and moved the hand towards the target (instant stretching). Then, we investigated sense of body ownership, sense of agency and task performance for those. We found that the illusory body ownership and sense of agency were higher in the order of constant stretching, accelerated stretching, and instant stretching. The learning effects on embodiment depended on the stretching method. Task performance was better in the order of constant, instant, and accelerated stretching. These results suggest that the illusory embodiment of the stretching arm and its learning depend on the type of stretching.

Visual-tactile synchronization or visual-motor synchronization makes a non-native body feel like one’s own body (illusory body ownership). Although previous studies have shown that body ownership can be induced in multiple bodies, it is unclear whether body ownership switches between bodies or is generated in multiple bodies simultaneously. We hypothesized that if body ownership is generated in multiple bodies simultaneously, the perception of affordance for aperture pass-ability will change corresponding to the number of bodies. This study investigated whether two avatars moving synchronously with the participant’s movements could induce body ownership in multiple avatars and whether this could change the affordance perception. The results showed that two avatars induced weaker body ownership than one avatar. However, the affordance perception did not change regardless of the number of bodies. One possibility is that the body ownership switched between the two avatars. Thus, even if the body ownership for one body switches between bodies, all the multiple bodies are may still be experienced as one’s own body.

Falls have become a worldwide problem for the health of the elderly. In recent years, equipment developed to prevent falls has focused primarily on increasing muscle strength. The purpose of this study is to establish a method to prevent falls by supplementing sensory function. Applying subthreshold vibrations to somatosensory mechanoreceptors can improve tactile sensitivity. In this study, we apply this method and propose a method to enhance knee joint coordination by presenting weak vibrations to the thigh. Experimental results showed that the proposed method significantly improved the perceived sensitivity of skin stretching for the group with low perceived sensitivity to the original skin stretching. The proposed method also significantly improved the accuracy of knee joint position sense for the group with low original knee joint position sense.

This paper proposed a new method to enhance the sensation of flight or floating up and down by visually presenting images of flying through an HMD and simultaneously applying vibrations to the soles. We varied the frequency and waveform of the vibration pattern applied to the foot soles and verified which vibration conditions enhanced the sensation of flight. The results showed that the frequency of 80 Hz and the waveform of asymmetric oscillation or white-noise vibration were more effective in inducing the sense of flight. This method can expand the range of VR content and reduce the system cost and space requirements.

This paper proposes a humanoid cybernetic avatar (CA) that can replicate and extend human presence inexpensively and create social interactions that are difficult to simulate among humans. Standard life-size humanoid robots are expensive and have the disadvantage of being difficult to customize and use multiple units. Therefore, we constructed a humanoid robot with an inflatable structure, omitting work and walking abilities and focusing only on appearance and gestures. Humanoid inflatable CAs are lightweight, contact-safe, foldable when deflated, and can be customized in shape and texture at low cost. The social interactions enabled by such robots are discussed and categorized.

The full-body illusion is known as the illusion of self-body perception for avatars in virtual reality (VR). In this study, experiments were conducted under three conditions, in which the subject experienced the full-body illusion to a doctor avatar, they experienced the full-body illusion to an ordinary person avatar, and the full-body illusion to a doctor avatar was disrupted by delayed visual feedback. The change of executive function scores and personality traits of the subject before and after the experiment were measured, and we confirmed that the executive function was improved when the subject had the full-body illusion to the doctor avatar than in the other conditions, and that the openness to experience of the personality traits were increased. These results suggest that the full-body illusion facilitates back projection of characteristics of the intelligent avatar onto the subject's own cognitive abilities and personality traits.

In sports, it is essential to accurately predict the results of others’ actions based on the movements’ characteristics and respond appropriately to them. Previous studies have proposed systems that predict motions using machine learning and present these results to users. However, it remains unclear whether the judgment basis constructed by machine learning is useful for humans to predict movements accurately. This study developed a Neural Network that predicted the kicking direction of soccer penalty kicks and extracted the basis. Also, we tested whether the extracted basis was useful for participants in their predictions. The results showed that participants perceived the extracted basis as valid, even though they had not used it as a strategy at first.

Many previous studies have reported on the transformation of perception/behavior, body image, and body schema due to the use of avatars in VR space. However, there are few previous studies on those transformations associated with the wearing of avatars with non-innate physical characteristics such as animal avatars, leaving many unexplored areas. In this study, we focused on the octopus as an example of an animal with non-native physical characteristics such as softness and structural consistency from arm to fingertip. Therefore, we developed an avatar that externally mimics the arm of an octopus, a box-shaped avatar, and a human arm-shaped avatar, and measured the movement of the right upper limb during manipulation and the sensory and cognitive changes after manipulation. The results showed that subjects felt their bodies became softer and their body image changed after manipulating the avatar that imitated an octopus arm. The possibility of behavioral change was also confirmed with the octopus avater.

When seeing and touching objects, we recognize their material categories and properties. The material category and property information may be combined across vision and touch. However, visuo-tactile material perception is poorly understood. In this study, we asked 24 participants to rate stimuli in 13 material categories with 12 material properties under vision-only, tactile-only and visuo-tactile conditions. The results indicated interactions between vision and touch in material category perception when participants felt natural for visuo-tactile combination. On the other hand, visuo-tactile material property perception can be explained by the weighted average of visual information and tactile information.

This paper describes the techniques to reducing VR sickness for wheelchair simulator. The aim of this research is to reduce the VR sickness in the wheelchair simulator, which is caused by the yaw rotation. To reduce the VR sickness, we propose two methods to reducing VR sickness, which are a reference grid frame and a translucent black outer frame. The two frames are displayed fixed to the movement of the head. The comparative experiment was performed to verify the effectiveness and interaction of the proposed methods. The experimental results are statistically analyzed, the analysis finds interaction for the two proposed methods.

Collaborative action enables individuals to exceed the limits of their abilities, and encourages new creativity. These are promoted by VR, computer-assisted systems, sensory sharing, etc., but the body is independent. In this paper, we propose a Collaborative Avatar, in which two people are integrated into a single robotic avatar, and investigate the possibilities for embodied collaboration through the avatar with others who have different skills and experiences. This system integrates the movements of two people in any ratio, and sharing each other’s senses, to embody them into a single robotic arm. Evaluation test showed that this system induced the equivalent sense of agency, body ownership, and usability as when the robot was operated by alone. We also found that the task performance was improved when the robot was operated by two people, and that the two people’s movements were synchronized. Interestingly, we found a correlation between the sense of agency and usability when the robot was operated by two people, but no correlation when operated by one person. It is expected that the Collaborative Avatar will improve our abilities and induce various embodied coordination between people, which will become a new way of collaboration in the cyber-physical society.