Personalized Recommendations in Mixed Reality Enhance Explanation Satisfaction and Hedonic User Experience in Board Game LearningBoard games often involve strategic decision making and procedural planning tasks. Such tasks require learners to make decisions based on dynamically evolving game state and changing information that is situated in a physical environment. Recommender systems can filter available information and provide learners with personalized and actionable suggestions that simplify their decision making while playing board games. Such recommendations can further be spatially aligned with relevant physical elements through Mixed Reality (MR). We present an MR system called GLAMRec for an engine-building strategy board game. GLAMRec provides personalized, transparent recommendations by integrating user data, real-time game state tracking, and ontology-based reasoning during a complex board game, which we use as a proxy environment for procedural learning tasks. We interviewed six board game designers to improve the GLAMRec and conducted a within-subjects design user study (N=32) to investigate how personalized explanations affect explanation satisfaction, user experience, and trust. We found that personalized recommendations significantly improve explanation satisfaction and hedonic user experience without affecting trust ratings, recommendation compliance, and game performance. These findings suggest that personalization primarily shaped perception of enjoyment rather than measurable learning outcomes or trust.2026SDSandra Dojcinovic et al.University of St. GallenGame UX & Player BehaviorMixed Reality WorkspacesAI-Assisted Decision-Making & AutomationIUI
Understanding How Mobile Interactions Shape Grasp and Contact Patterns Beyond the TouchscreenThe way users hold a smartphone depends on the interaction task, yet little is known about the fingers' engagement with the device's surfaces beyond the touchscreen. Such an understanding not only opens up opportunities for novel on- and off-screen interactions, but also the device’s possible physical affordances. We present a study (N=23) that examines the hands' physical engagement with the smartphone beyond the touchscreen across nine mobile interactions. Grasps were annotated from photographs, and contact regions were captured using residual heat traces from grasping the device. Our findings show that fingers and palms adopt a variety of support roles and postures when engaging with the smartphone's back and side edges. The hand-contact maps reveal distinct patterns, differing in contact frequency and placement. This work contributes an empirical characterisation of hands' back and edge engagement, highlighting design opportunities for future smartphone usage extending beyond the touchscreen.2026CSCarolin Stellmacher et al.University of BremenOne-Handed Operation & Mobile GesturesTouch Target Selection & PointingCHI
Adaptive Tutoring Modalities for Supporting Learners’ Reflective Writing PracticesReflection is essential for fostering metacognitive development. However, many learners struggle to engage in meaningful, structured reflection without further support. To support learners in reflective practices, we developed MindBuddy, a learner-centered tutor that guides students individually through reflective writing tasks and provides adaptive feedback. After an iterative user-centered development process (two pilot studies, n=81), we conducted a longitudinal field-experimental classroom study with n=34 undergraduates over a six-week period to compare two different tutoring modalities in MindBuddy (1) interactive conversational tutoring (TG1) with (2) constructive feedback-only (TG2). No significant differences in perceived skills were found, suggesting that the conversational interactions may enhance students’ confidence in their reflective abilities, similarly to adaptive feedback interaction. While differences were found for formal reflective structure, our findings suggest that conversational tutoring has the potential to increase learners’ engagement with reflective writing. Future research on whether such engagement translates into measurable performance gains is necessary.2026LWLéane Wettstein et al.Bern University of Applied SciencesIntelligent Tutoring Systems & Learning AnalyticsCollaborative Learning & Peer TeachingAI-Assisted Writing & Text GenerationCHI
Leveraging Learner Errors in Digital Argumentation Learning: How ALure Helps Students Learn from their Mistakes and Write Better ArgumentsProviding argumentation feedback is considered helpful for students preparing to work in collaborative environments, helping them with writing higher-quality argumentative texts. Domain-independent natural language processing (NLP) methods, such as generative models, can utilize learner errors and fallacies in argumentation learning to help students write better argumentative texts. To test this, we collect design requirements, and then design and implement two different versions of our system called ALure to improve the students’ argumentation skills. We test how ALure helps students learn argumentation in a university lecture with 305 students and compare the learning gains of the two versions of ALure with a control group using video tutoring. We find and discuss the differences of learning gains in argument structure and fallacies in both groups after using ALure, as well as the control group. Our results shed light on the applicability of computer-supported systems using recent advances in NLP to help students in learning argumentation as a necessary skill for collaborative working settings.2025SNSeyed Parsa Neshaei et al.Fighting Misinformation, Building BelievabilityCSCW
Emotionally Aware Moderation: The Potential of Emotion Monitoring in Shaping Healthier Social Media ConversationsSocial media platforms increasingly employ proactive moderation techniques, such as detecting and curbing toxic and uncivil comments, to prevent the spread of harmful content. Despite these efforts, such approaches are often criticized for creating a climate of censorship and failing to address the underlying causes of uncivil behavior. Our work makes both theoretical and practical contributions by proposing and evaluating two types of emotion monitoring dashboards to users' emotional awareness and mitigate hate speech. In a study involving 211 participants, we evaluate the effects of the two mechanisms on user commenting behavior and emotional experiences. The results reveal that these interventions effectively increase users' awareness of their emotional states and reduce hate speech. However, our findings also indicate potential unintended effects, including increased expression of negative emotions (Angry, Fear, and Sad) when discussing sensitive issues. These insights provide a basis for further research on integrating proactive emotion regulation tools into social media platforms to foster healthier digital interactions.2025XSXiaotian Su et al.Toxic and Anti-Social BehaviorCSCW
Vocalizing User Feedback: The Impact of Input Modality on Self-DisclosureThis study explores how input modality — voice versus text — affects self-disclosure in user feedback, leveraging a novel approach that uses transformer-based models to detect self-disclosure per token embedded in context. In an online experiment with 122 participants, results indicate that participants using voice input engaged significantly less in self-disclosure than those using text, a finding associated with reduced perceived anonymity in voice interactions. This effect persisted after accounting for response length, suggesting that the influence of voice input on self-disclosure is not merely due to brevity but also reflects unique psychological responses to voice-based communication. These findings contribute to a deeper theoretical understanding of input modality’s role in shaping disclosure behavior in user feedback contexts. Practical implications offer design guidance for voice-based feedback systems to encourage more open and authentic feedback in sensitive settings.2025MGMarc Christopher Grau et al.Voice TechnologyCSCW
To Cuddle, Mingle, Venture, or Guide: How Architectural Affordances Influence the Experience of Social VR PlacesSocial virtual reality (VR) encompasses a growing network of three-dimensional virtual worlds where users interact in a shared, embodied way. While research has focused on the social interactions between the users themselves, less is known about how the design of virtual spaces influences these interactions. Our study combines interviews with 15 social VR users logging over 1,000 hours and a 20-hour spatial protocol of a purposeful sampling of VR worlds. We analysed how spatial characteristics (including proportion, sightlines, materiality, atmosphere, and navigation) influence meaningful user interaction to turn space into place. We synthesised four place types for a new social VR typology: Cuddle worlds that encourage cosy conversations; Mingle worlds that facilitate new encounters; Venture worlds that promote exploration; and Guided worlds that elicit a sense of belonging with the online community. By relating architectural affordances to social patterns, we contribute insights towards the purposeful design of social VR places.2025JHJihae Han et al.Social & Collaborative VRImmersion & Presence ResearchVisualization Perception & CognitionDIS
Towards Societally Beneficial Personalized Realities: A Conceptual Foundation for Responsible Ubiquitous Personalization SystemsPersonalization of online realities is today ubiquitous to support decision making or reduce information overload. Recently, through the expanding capabilities and pervasiveness of Mixed Reality and Ubiquitous Computing technologies, we observe increasing personalization also of physical reality. This might yield more convenient, efficient and inclusive everyday interactions. However, it may readily lead to serious societal consequences such as the loss of shared worlds and the emergence of perceptual filter bubbles. To mitigate such harms while retaining the benefits of personalization, it is important to understand how ubiquitous personalization systems may operate responsibly. Responding to this need, we propose a conceptual model that overcomes the limitations of established personalization models and expands their applicable scope to physical, virtual, and hybrid environments. We validated our model in relation to existing literature and show how it provides a conceptual foundation for the analysis and study of responsible personalization systems that create individually and societally beneficial Personalized Realities.2025JSJannis Strecker-Bischoff et al.AI-Assisted Decision-Making & AutomationAI Ethics, Fairness & AccountabilityUbiquitous ComputingDIS
Experiencing the World through Imperfect Lenses: An Autoethnography of Living in Mixed RealityMixed reality (MR) technologies are evolving to become more portable, incorporating video see-through capabilities, which enable a shift from stationary to mobile use. This development allows MR headsets to be used in various everyday contexts, including eating, travelling, and exercising. Before MR technologies reshape how we live and seamlessly integrate into our daily activities, we must understand the lived experiences of using MR in our personal lives and their influences and implications on our day-to-day activities. This paper presents an autoethnographic study that adopts an exploratory first-person perspective to uncover challenges and opportunities within this intimate context. We present the experiences and challenges of living in mixed reality, including on-the-go scenarios and social interactions. Our findings reveal issues such as social and ethical concerns and offer lessons learned to inform the design of future interactive systems for mobile mixed reality.2025YSYu Sun et al.Mixed Reality WorkspacesImmersion & Presence ResearchContext-Aware ComputingDIS
UrbAI: Exploring the Possibilities of Generative AI Image Processing to Promote Citizen ParticipationGiving citizens a voice in urban development processes is crucial for enabling socially sustainable cities and communities. However, citizens' opportunities to express ideas are often limited to communication channels that offer poor incentives for participation. In this paper, we conducted an in-the-wild technology probe study (N=16) using a generative AI (GenAI) tool to allow citizens to visualise and submit urban development ideas by taking pictures and manipulating them with GenAI. The results highlight the potential of GenAI to empower, engage, and inspire citizens‘ creativity. We then conducted additional expert interviews (N=6) with city representatives and community associates. They voiced GenAI's value in early-stage citizen participation but raised concerns about excluding senior citizens. Building on these insights, we present the design and evaluation (N=10) of UrbAI, a co-creative system tailored to urban development participation and conclude with lessons learned to inform how GenAI could be embedded in future citizen participation processes.2025APAdrian Preussner et al.University of St. GallenGenerative AI (Text, Image, Music, Video)Community Engagement & Civic TechnologyCHI
LifeInsight: Design and Evaluation of an AI-Powered Assistive Wearable for Blind and Low Vision People Across Multiple Everyday Life ScenariosAssistive technologies (ATs) have the potential to empower blind and low vision (BLV) people. Yet, they often remain underutilised due to their immobility and limited applicability across scenarios. This paper presents LifeInsight, an AI-powered assistive wearable for BLV people that uses a wearable camera, microphone and single-click interface for goal-oriented visual querying. To inform the design of LifeInsight, we first collected a corpus of BLV people’s daily experiences using video probes and interviews. Ten BLV people recorded their daily experiences over one week using GoPro cameras, providing empirical insights. Based on these, we report on LifeInsight and its evaluation with 13 BLV people across six scenarios. LifeInsight effectively responded to visual queries, such as distinguishing between jars or identifying the status of a candle. Drawing on our work, we conclude with key lessons and practical recommendations to guide future research and advance the development and evaluation of AI-powered assistive wearables.2025FMFlorian Mathis et al.University of St. Gallen; University of Applied Sciences of the GrisonsHaptic WearablesVisual Impairment Technologies (Screen Readers, Tactile Graphics, Braille)CHI
Ad-Blocked Reality: Evaluating User Perceptions of Content Blocking Concepts Using Extended RealityInspired by the concepts of diminishing reality and ad-blocking in browsers, this study investigates the perceived benefits and concerns of blocking physical, real-world content, particularly ads, through Extended Reality (XR). To understand how users perceive this concept, we first conducted a user study (N=18) with an ad-blocking prototype to gather initial insights. The results revealed a mixed willingness to adopt XR blockers, with participants appreciating aspects such as customizability, convenience, and privacy. Expected benefits included enhanced focus and reduced stress, while concerns centered on missing important information and increased feelings of isolation. Hence, we investigated the user acceptance of different ad-blocking visualizations through a follow-up online survey (N=120), comparing six concepts based on related work. The results indicated that the XR ad-blocker visualizations play a significant role in how and for what kinds of advertisements such a concept might be used, paving the path for future feedback-driven prototyping.2025CKChristopher Katins et al.HU BerlinPrivacy by Design & User ControlSocial Platform Design & User BehaviorCHI
Moving Beyond the Simulator: Interaction-Based Drunk Driving Detection in a Real Vehicle Using Driver Monitoring Cameras and Real-Time Vehicle DataAlcohol consumption poses a significant public health challenge, presenting serious risks to individual health and contributing to over 700 daily road fatalities worldwide. Digital interventions can play a crucial role in reducing these risks. However, reliable drunk driving detection systems are vital to effectively deliver these interventions. To develop and evaluate such a system, we conducted an interventional study on a test track to collect real vehicle data from 54 participants. Our system reliably identifies non-sober driving with an area under the receiver operating characteristic curve (AUROC) of 0.84 ± 0.11 and driving above the WHO-recommended blood alcohol concentration limit of 0.05 g/dL with an AUROC of 0.80 ± 0.10. Our models rely on well-known physiological drunk driving patterns. To the best of our knowledge, we are the first to (1) rigorously evaluate the potential of (2) driver monitoring cameras and real-time vehicle data for detecting drunk driving in a (3) real vehicle.2025RDRobin Deuber et al.ETH ZürichTeleoperated DrivingHuman Pose & Activity RecognitionCHI
Describing Explored Places through OpenStreetMap DataMobile navigation applications are good at providing efficient navigation instructions. However, they currently lack the capability to facilitate free exploration. Therefore, users are limited to encountering only places close to the shortest paths, neglecting places that could diversify navigation and foster spatial learning. To better understand what characteristics places have that users like to explore we collected a dataset with a mobile application that encourages free exploration using gamification (n = 39, t = 455 days, 106.50 km2). Using OpenStreetMap data, we found highly frequented freely explored places comprising office, educational, retail, touristic and commercial places. When comparing the characteristics of the freely explored places to those along the shortest path, those categories were different. Based on our findings, we propose that implementing more diverse routing algorithms can enhance navigation diversity, improve spatial learning, and optimise the utilisation of urban spaces for travel.2025ESEve Schade et al.University of St. GallenGeospatial & Map VisualizationPublic Transit & Trip PlanningCHI
InterFACE: Establishing a Facial Action Unit Input Vocabulary for Hands-Free Extended Reality Interactions, From VR Gaming to AR Web BrowsingExtended Reality (XR) interactions often rely on spatial hand or controller inputs - necessitating dexterous wrist, hand and finger movements including pressing virtual buttons, pinching to select, and performing hand gestures. However, there are scenarios where such dependencies may render XR devices and apps inaccessible to users - from situational/temporary impairments such as encumbrance, to physical motor impairments. In this paper, we contribute to a growing literature considering facial input as an alternative. In a user study (N=20) we systematically evaluate the usability of 53 Facial Action Units in VR, deriving a set of optimal (comfort, effort, performance) FAUs for interaction. We then use these facial inputs to drive and evaluate (N=10) two demonstrator apps: VR locomotion, and AR web browsing, showcasing how close facial interaction can get to existing baselines, and demonstrating that FAUs offer a viable, generalizable input modality for XR devices.2025GWGraham Wilson et al.University of Glasgow, School of Computing ScienceHand Gesture RecognitionFull-Body Interaction & Embodied InputEye Tracking & Gaze InteractionCHI
Real-Time Adaptive Industrial Robots: Improving Safety And Comfort In Human-Robot CollaborationIndustrial robots become increasingly prevalent, resulting in a growing need for intuitive, comforting human-robot collaboration. We present a user-aware robotic system that adapts to operator behavior in real time while non-intrusively monitoring physiological signals to create a more responsive and empathetic environment. Our prototype dynamically adjusts robot speed and movement patterns to proxemics while measuring operator pupil dilation. Our user study compares this adaptive system to a non-adaptive counterpart, and demonstrates that the adaptive system significantly reduces both perceived and physiologically measured cognitive load while enhancing usability. Participants reported increased feelings of comfort, safety, trust, and a stronger sense of collaboration when working with the adaptive robot. This highlights the potential of integrating real-time physiological data into human-robot interaction paradigms. This novel approach creates more intuitive and collaborative industrial environments where robots effectively ’read’ and respond to human cognitive states, and we feature all data and code for future use.2025DHDamian Hostettler et al.University of St. Gallen, ICS-HSGBiosensors & Physiological MonitoringHuman-Robot Collaboration (HRC)CHI
Efficient Management of LLM-Based Coaching Agents' Reasoning While Maintaining Interaction Quality and SpeedLLM-based agents improve upon standalone LLMs, which are optimized for immediate intent-satisfaction, by allowing the pursuit of more extended objectives, such as helping users over the long term. To do so, LLM-based agents need to reason before responding. For complex tasks like personalized coaching, this reasoning can be informed by adding relevant information at key moments, shifting it in the desired direction. However, the pursuit of objectives beyond interaction quality may compromise this very quality. Moreover, as the depth and informativeness of reasoning increase, so do the number of tokens required, leading to higher latency and cost. This study investigates how an LLM-based coaching agent can adjust its reasoning depth using a discrepancy mechanism that signals how much reasoning effort to allocate based on how well the objective is being met. Our discrepancy-based mechanism constrains reasoning to better align with alternative objectives, reducing cost roughly tenfold while minimally impacting interaction quality.2025AGAndreas Göldi et al.University of St.GallenHuman-LLM CollaborationAI-Assisted Decision-Making & AutomationCHI
A Digital Companion Architecture for Ambient IntelligenceGarcia等人提出面向环境智能的数字伴侣架构，整合多模态感知与智能交互，为用户提供个性化的持续辅助服务。2024KGKimberly Garcia et al.Context-Aware ComputingSocial Robot InteractionUbiComp
MetaFormer: Domain-Adaptive WiFi Sensing with Only One Labelled Target SampleSheng 等人提出 MetaFormer 框架，通过元学习仅使用一个标注目标样本实现 WiFi 感知的跨域迁移，有效降低标注成本。2024BSBiyun Sheng et al.Context-Aware ComputingUbiquitous ComputingUbiComp
DIY Digital Interventions: Behaviour Change with Trigger-Action ProgrammingWhether it is sleep, diet, or procrastination, changing behaviours can be challenging. Individuals could design and build their own personalised digital interventions to help them reach their goals, but little is known about this process. Building upon previous research we propose the Behaviour Change with Trigger-Action Programming (BC-TAP) model which describes how individuals could bridge the gap between their current and desired behaviour through the creation of `Do-It-Yourself' (DIY) digital interventions. We conducted a two-day participatory workshop based on the BC-TAP model with 28 participants. Participants articulated plans to change a behaviour of their choice and represented these plans in mobile device automations. After using their interventions for up to three weeks, participants reflected on their experience. Our findings report opportunities and challenges at each stage of the process. While formulating a digital proxy for certain behaviours was challenging, both failures and successes facilitated participants’ awareness of their behaviour, and their ability to change it.2024ASAva Elizabeth Scott et al.Creative Collaboration & Feedback SystemsKnowledge Worker Tools & WorkflowsMobileHCI