In virtual reality (VR) experiences, mismatches between reality and virtuality are usually undesirable, as they can disrupt immersion and induce cybersickness. However, when carefully controlled, they may expand the design space of VR. This research investigates perceptual detection thresholds for mismatches between real and virtual inclinations during cycling in VR. Using a custom simulation, N = 30 participants cycled through a virtual city while physical and visual inclinations were independently manipulated. Real inclinations were implemented with a tilting indoor bike, providing vestibular and proprioceptive feedback, while virtual inclinations within the simulated environment were presented visually. A multiple staircase procedure derived estimates for perceptual thresholds that approximate which mismatches in visual and physical inclination were still perceived as congruent. These thresholds reveal a window of perceived congruence before mismatches become noticeable to users. These findings advance understanding of sensory integration in VR cycling and inform applications in immersive training, exergames, and rehabilitation systems.

Micromobility vehicles, such as e-scooters, Segways, skateboards, and unicycles, are increasingly adopted for short-distance travel due to their low weight and low emissions. Despite their growing popularity, we lack controlled, low-risk environments to study rider experiences and performance. While virtual reality (VR) simulators offer a promising approach by reducing safety risks and providing immersive experiences, micromobility simulators remain largely underexplored. We introduce MicroVRide, a modular 4-in-1 VR micromobility simulator that supports e-scooters, Segways, electric unicycles, and one-wheeled skateboards on a single platform. The simulator preserves vehicle-specific physical constraints and control metaphors, enabling the study of diverse riding behaviors with minimal hardware reconfiguration. We contribute the simulator design and report a preliminary within-subject study (N = 12) that demonstrates feasibility and reveals distinct experiential profiles across vehicles.

Recent advancements in augmented reality (AR) technologies have brought us closer to the vision of everyday ubiquitous computing and pervasive AR use. State-of-the-art AR glasses now enable mobile, on-the-move experiences that extend beyond laboratory settings. This promise has already been explored across diverse mobile contexts, including transportation, entertainment, shopping, and tourism. This workshop focuses on AR on-the-move, examining how AR can support interaction in dynamic settings where environmental and social contexts shift rapidly and users are in locomotion. We will discuss both the unique challenges and the rich opportunities such scenarios present for meaningful augmentation and new forms of interaction. By uniting established scholars and emerging researchers from across HCI, this workshop seeks to map out critical directions for future inquiry. Through hands-on activities and reflection, participants will collectively identify key opportunities, challenges, and next steps for advancing research on AR on-the-move.

Cultural differences influence how cyclists and drivers interact, affecting global autonomous vehicle (AV) adoption. AV-cyclist interfaces are needed to clarify AV intentions and resolve ambiguities when no human driver is present. These must adapt across cultures and road infrastructure. We conducted the first cross-cultural AV-cyclist user study across Stockholm (high segregation of cyclists from drivers), Glasgow (some segregation), and Muscat (no segregation). Cyclists used an AR simulator to cycle in physical space and experienced three holistic AV-cyclist interfaces. These integrated multiple interfaces into a larger ecosystem, e.g., a smartwatch synchronised with on-vehicle eHMI. Interfaces communicated AV location, intentions, or both. Riders from all cities preferred combined AV location and intention information but used it differently. Stockholm cyclists focused on location, validating intentions with driving behaviour. Glasgow riders valued both cues equally. Muscat cyclists trusted interfaces, prioritising intentions without relying on driving behaviour. These insights are key for global AV adoption.

Virtual Reality (VR) locomotion methods are mainly ground-based, room-scale, or discrete, making them ill-suited for flying experiences. Although leaning- and controller-based techniques are promising for flying in VR, we lack empirical evidence of their advantages. We compared combinations of leaning- and controller-based methods for steering and velocity in a user study (N = 24) using a broom metaphor to integrate these methods into an understandable locomotion reference. The steering methods were: 1) controller-pointing (CP) and 2) headset-leaning (HL); and for velocity control: 1) controller linear displacement (CLD) and 2) headset linear displacement (HLD). Results indicate that HL increase presence compared to CP. However, combining HL with CLD worsens coin collection rate, completion time, mental load, control factor ratings, and enjoyment. In contrast, HLD worked well when paired with either steering method. CP-CLD led to the highest coin collection rate and lowest mental load. All methods had comparable feelings of flying.

Animations produced by generative models are often evaluated using objective quantitative metrics that do not fully capture perceptual effects in immersive virtual environments. To address this gap, we present a preliminary perceptual evaluation of generative models for animation synthesis, conducted via a VR-based user study (N = 48). Our investigation specifically focuses on animation congruency—ensuring that generated facial expressions and body gestures are both congruent with and synchronized to driving speech. We evaluated two state-of-the-art methods: a speech-driven full-body animation model and a video-driven full-body reconstruction model, assessing their capability to produce congruent facial expressions and body gestures. Our results demonstrate a strong user preference for combined facial and body animations, highlighting that congruent multimodal animations significantly enhance perceived realism compared to animations featuring only a single modality. By incorporating VR-based perceptual feedback into training pipelines, our approach provides a foundation for developing more engaging and responsive virtual characters.

Introduction: Social interactions incorporate various nonverbal signals to convey emotions alongside speech, including facial expressions and body gestures. Generative models have demonstrated promising results in creating full-body nonverbal animations synchronized with speech; however, evaluations using statistical metrics in 2D settings fail to fully capture user-perceived emotions, limiting our understanding of the effectiveness of these models. Methods: To address this, we evaluate emotional 3D animation generative models within an immersive Virtual Reality (VR) environment, emphasizing user—centric metrics-emotional arousal realism, naturalness, enjoyment, diversity, and interaction quality—in a real-time human-agent interaction scenario. Through a user study (N = 48), we systematically examine perceived emotional quality for three state-of-the-art speech-driven 3D animation methods across two specific emotions: happiness (high arousal) and neutral (mid arousal). Additionally, we compare these generative models against real human expressions obtained via a reconstruction-based method to assess both their strengths and limitations and how closely they replicate real human facial and body expressions. Results: Our results demonstrate that methods explicitly modeling emotions lead to higher recognition accuracy compared to those focusing solely on speech-driven synchrony. Users rated the realism and naturalness of happy animations significantly higher than those of neutral animations, highlighting the limitations of current generative models in handling subtle emotional states. Discussion: Generative models underperformed compared to reconstruction-based methods in facial expression quality, and all methods received relatively low ratings for animation enjoyment and interaction quality, emphasizing the importance of incorporating user-centric evaluations into generative model development. Finally, participants positively recognized animation diversity across all generative models.

Generative Artificial Intelligence (GenAI) chatbots start changing experiences with art for museum visitors by making them more interactive and engaging. However, it remains underexplored how GenAI chatbots influence visitors' in-field experience and interaction at art museums regarding finding information, engagement, and enjoyment compared to existing museum tour-guide applications. In this paper, we contribute the design and implementation of a smartphone-based chatbot that detects artwork, generates textual and auditory information, and interactively answers visitors' questions. To explore visitors' experience with it, we conducted a field experiment (N=30) at the National Art Museum, comparing it to the existing museum application. Our results indicate that visitors showed higher artwork engagement with the chatbot than the museum application. Moreover, they enjoyed an interactive experience using the chatbot to learn about the art collection and have equally preferred textual and auditory information representation.

As Virtual Reality enters professional domains, efficient text entry remains challenging due to limited tactile feedback and ergonomics. We present a novel Virtual Reality keyboard that integrates a smartphone-like layout, hover effects, dynamic colors, haptic feedback (SenseGlove Nova 2), auditory cues, and a key-weighting mechanism. In a study (N = 11), our design was compared with a curved MRTK keyboard. Although the MRTK keyboard yielded faster speeds (8.99 vs. 7.47 WPM) and higher accuracy (92.14% vs. 86.85%), our custom design offered a more engaging and ergonomic experience with improved comfort over prolonged use. These findings underscore the potential of multimodal feedback for advancing VR text entry.

Precise manipulation, fast interaction, and lasting comfort are hard to reconcile in VR. Modern controllers can track at sub-millimetre resolution in ideal conditions, yet mid-air reach, tremor, and arm elevation still impair accuracy and induce fatigue. We present Eyes on Target, Pen on Table, combining gaze-driven target selection with desk-supported, pressure-adaptive pen input. In a within-subjects study (N=12), it was compared with a direct mid-air pen baseline. In 2D tasks, mean placement error dropped by 40% with no significant time cost, while in 3D tasks error fell by 58% at a 52% time cost. With control-display gain with axis constraint, error reductions reached 77% in 2D and 81% in 3D relative to the baseline. Across conditions, the approach also reduced arm fatigue and NASA-TLX workload. Our findings position the Gaze + Pen on Table technique as a balanced, sustainable alternative for production-level VR design, with practical guidelines for implementation.