Virtual Selves, Real Emotions: The Impact of Photorealistic Avatar Personalization on Emotion in Virtual RealityEmbodying photorealistic personalized avatars that closely resemble their users has recently become technologically achievable in virtual reality (VR). While previous work highlighted several benefits of personalization, its impact on emotions, a central VR element, remains underexplored. Given their high production cost, assessing the added value of photorealistic personalized avatars is essential before their adoption as a design feature. To address this, we designed and validated (n=302) four virtual environments (VEs) that induce emotions in different quadrants of Russell's Circumplex Model. Using these VEs, we systematically investigated the impact of avatar type (no avatar, generic avatar, personalized avatar) on perceived emotions and physiological responses (n=51). Generic avatars enhanced valence, arousal, embodiment, presence and selected physiological measures, while personalized avatars further amplified these effects and additionally increased heart rate variability, a marker of self-regulation. Our study demonstrates that photorealistic personalized avatars markedly enhance VR’s capacity to foster emotional engagement.2026ASAnca Salagean et al.University of BathImmersion & Presence ResearchIdentity & Avatars in XREmotion Recognition & DetectionCHI
Sociotechnical Challenge Modeling: A Design Method for Responsible AI in Healthcare and Social WelfareWe present Sociotechnical Challenge Modeling (STCM), a workshop-based design method to help healthcare and social welfare practitioners identify and address sociotechnical challenges in machine learning (ML) deployments. We evaluated STCM in a field experiment with two UK organizations, involving 26 practitioners including managers, data scientists, and frontline care professionals. The evaluation found that STCM cultivated a sociotechnical perspective by revealing interdependencies between ML tools and organizational practices. The physical cards stimulated exchange and experimentation, while the workshop fostered collaboration across disciplines. However, participants found predefined countermeasures too prescriptive, which prompted revisions to support more open-ended ideation. Our contributions are a novel design method for anticipating and mitigating sociotechnical challenges of ML in care settings, and an empirical evaluation of its perceived value and limitations. To support adoption and further research, all STCM materials, including editable card templates and worksheets, are available at: https://bit.ly/4plXkfi.2026TRTyler Reinmund et al.University of OxfordAI-Assisted Decision-Making & AutomationAI Ethics, Fairness & AccountabilityResearch Ethics & Open ScienceCHI
HydroHaptics: High-Fidelity Force-Feedback on Soft Deformable Interfaces using Hydrostatic TransmissionSoft deformable interfaces offer unique interaction potential through input flexibility and diverse forms. However, force feedback on these devices remains limited, with pneumatic approaches lacking responsiveness and precision, while microhydraulic solutions are constrained to small form factors with limited input. We present HydroHaptics, a novel platform that enables high-fidelity force feedback on deformable interfaces via hydrostatic transmission. Surpassing current state-of-the-art methods, our approach allows fine-grained force feedback on soft interfaces, achieving a 10 N force change in < 100 ms and accurate 1 N, 10 Hz oscillation rendering. We detail the system's design and implementation, highlighting its ability to maintain the inherent interaction benefits of soft interfaces. A user study (N = 18)evaluates the system's performance, showing high accuracy in rendering distinct haptic effects (82.6% accuracy) and classifying input gestures (89.1% accuracy). To showcase the platform’s versatility, we present four applications illustrating HydroHaptics' potential to enhance interaction with deformable devices and unlock novel user experiences.2025JNJames David Nash et al.Force Feedback & Pseudo-Haptic WeightShape-Changing Interfaces & Soft Robotic MaterialsUIST
It Sounds Squishy: Understanding Cross-Modal Correspondences of Deformable Shapes and Sounds Computing interfaces are becoming increasingly sophisticated, with systems that engage multiple sensory channels simultaneously. Deformable and shape-changing interfaces offer rich tactile experiences, but there is limited understanding of how they can be combined with other modes of sensory feedback. We systematically explored the audio, visual and tactile cross-modal correspondences of deformable shapes with a particular focus on auditory feedback. 50 participants were asked to associate deformable tactile stimuli, varying in stiffness and shape, with the sound qualities pitch, brightness, fade-in time and fade-out time, under visuo-tactile and tactile-only conditions. Our findings provide the first insights on how (1) shape, both its form and visibility, play a significant role in associations for pitch and brightness; (2) stiffness plays a dominant role in associations over a sound’s fade-in and fade-out times. These findings are distilled into the first design guidelines for integrating auditory feedback into physical interfaces.2025MPMaisie Palmer et al.In-Vehicle Haptic, Audio & Multimodal FeedbackMid-Air Haptics (Ultrasonic)Shape-Changing Interfaces & Soft Robotic MaterialsDIS
Transformers and Human-robot Interaction for Delirium DetectionAn estimated 20% of patients admitted to hospital wards are affected by delirium. Early detection is recommended to treat underlying causes of delirium, however workforce strain in general wards often causes it to remain undetected. This work proposes a robotic implementation of the Confusion Assessment Method for the Intensive Care Unit (CAM-ICU) to aid early detection of delirium.Interactive features of the assessment are performed by Human-robot Interaction while a Transformer-based deep learning model predicts the Richmond Agitation Sedation Scale (RASS) level of the patient from image sequences; thermal imaging is used to maintain patient anonymity.A user study involving 18 participants role-playing each of alert, agitated, and sedated levels of the RASS is performed to test the HRI components and collect a dataset for deep learning. The HRI system achieved accuracies of 1.0 and 0.833 for the inattention and disorganised thinking features of the CAM-ICU, respectively, while the trained action recognition model achieved a mean accuracy of 0.852 on the classification of RASS levels during cross-validation.The three features represent a complete set of capabilities for automated delirium detection using the CAM-ICU, and the results demonstrate the feasibility of real-world deployment in hospital general wards.2023JJJoe Jeffcock et al.Human Pose & Activity RecognitionBiosensors & Physiological MonitoringRobots in Education & HealthcareHRI
Studying How Digital Luthiers Choose Their ToolsDigital lutherie is a sub-domain of digital craft focused on creating digital musical instruments: high-performance devices for musical expression. It represents a nuanced and challenging area of human-computer interaction that is well established and mature, offering the opportunity to observe designers' work on highly demanding human-computer interfaces. This paper explores how and why digital luthiers choose their tools and how these tools relate to the challenges they face. Findings from 27 standardised open-ended interviews with prominent digital luthiers from commercial, research, independent and artistic backgrounds are analysed through reflexive thematic analysis. Our discussion explores their perspectives, finding that a process of pragmatic rationalisation and environmental influences play a significant role in tool selection. We also present how challenges faced by digital luthiers relate to social creativity and meta-design. These findings build upon the existing literature that examines the designer-tool relationship.2022NRNathan Renney et al.University of West of EnglandMusic Composition & Sound Design ToolsCreative Coding & Computational ArtCustomizable & Personalized ObjectsCHI
“Are We Now Post-COVID?”: Exploring Post-COVID Futures Through a Gamified Story Completion MethodCOVID-19 has heavily impacted our lives. To date, the ongoing pandemic continues to cause dramatic societal changes and raises shared sentiments of uncertainty for our future. As such, however, COVID-19 provides opportunities to explore futures through speculative research. Here, we gamify the story completion method (SCM) to explore futures post-COVID and ask 37 participants to play a day in the life of Sal in a post-COVID future. The game asks participants to describe what Sal sees, hears, or does throughout a day based on multiple story stems. Our analysis reveals narratives of post-COVID futures as business as usual, back to basics, or everyday chaos. Notably, these narratives raise concerns about privacy loss and increased militarization, but also envision futures post-COVID that reclaim stronger bond with nature and family. We discuss the lessons learned from gamifying the SCM and the temporal implications of performing speculative research during evolving dramatic events.2021GTGiovanni Maria Troiano et al.Technology Ethics & Critical HCIDesign FictionDIS
Examining Profiles for Robotic Risk Assessment: Does a Robot's Approach to Risk Affect User Trust?As autonomous robots move towards ubiquity, the need for robots to make decisions under risk that are trustworthy becomes increasingly significant; both to aid acceptance and to fully utilise their autonomous capabilities. We propose that incorporating a human approach to risk assessment into a robot’s decision making process will increase user trust. This work investigates four robotic approaches to risk and, through a user study, explores the levels of trust placed in each. These approaches are: risk averse, risk seeking, risk neutral and a human approach to risk. Risk is artificially stimulated through performance-based compensation, in line with previous studies. The study was conducted in a virtual nuclear environment created using the Unity games engine. Forty participants were asked to complete a robot supervision task, in which they observed a robot making risk based decisions and were able to question the robot, question the robot further and ultimately accept or alter the robot’s decision. It is shown that a robot that is risk seeking is significantly less trusted than a risk averse robot, a risk neutral robot and a robot utilising human approach to risk. There was found to be no significant difference between the levels of trust placed in the risk averse, risk neutral and human approach to risk. It is also found that the level to which participants question a robot’s decisions does not form an accurate measure of trust. The results suggest that when designing a robot that must make risk based decisions during teleoperation in a hazardous environment, an engineer should avoid a risk seeking robot. However, that same engineer may choose whichever of the remaining risk profiles best suits the implementation, with knowledge that the trust in their system is unlikely to be significantly affected.2020TBTom J F Bridgwater et al.Human-Robot Collaboration (HRC)HRI
Behind the Curtain of the "Ultimate Empathy Machine": On the Composition of Virtual Reality Nonfiction ExperiencesVirtual Reality nonfiction (VRNF) is an emerging form of immersive media experience created for consumption using panoramic "Virtual Reality" headsets. VRNF promises nonfiction content producers the potential to create new ways for audiences to experience "the real"; allowing viewers to transition from passive spectators to active participants. Our current project is exploring VRNF through a series of ethnographic and experimental studies. In order to document the content available, we embarked on an analysis of VR documentaries produced to date. In this paper, we present an analysis of a representative sample of 150 VRNF titles released between 2012-2018. We identify and quantify 64 characteristics of the medium over this period, discuss how producers are exploiting the affordances of VR, and shed light on new audience roles. Our findings provide insight into the current state of the art in VRNF and provide a digital resource for other researchers in this area.2019CBChris Bevan et al.University of BristolImmersion & Presence Research360° Video & Panoramic ContentInteractive Narrative & Immersive StorytellingCHI