Collecting user data is ubiquitous in the modern web, because it enables different kinds of services like personalised recommendations. Specific data shared by the users can be analysed further to infer additional information - e.g., gender, age, appearance, or preferences. Such inferences are challenging to predict for users when deciding on whether or not to share data. In this paper, we explore privacy avatars as a way to visualise collected and inferred data to users by a picture. For this, we conducted semi-structured interviews (N=20) where participants were shown privacy avatars based on products they bought in a fictitious online-shopping scenario and additional privacy avatars with a varied level of detail and information. Our results show that participants were shocked by avatars that accurately depicted parts of their visual appearance and perceived inaccurate avatars as privacy-preserving. However, participants preferred information they considered sensitive (e.g., gender) to be accurate creating a tension between privacy and accuracy of shared or inferred information. We conclude with a discussion on how such privacy avatars could be generated and used in practice to offer intuitive and informative privacy interfaces.
Driving is a cognitively demanding activity increasingly shaped by interactions with in-vehicle systems. However, current voice assistants lack contextual understanding and often struggle to interpret ambiguous user intent that requires disambiguation. This research explores how large language model (LLM)-based in-car assistants can interpret and act on user intent in a safe, context-sensitive, and human-like manner. It is guided by two main research questions: (1) how users express intent through multimodal cues such as speech, gaze, and touch to provide contextual grounding for the LLM, and (2) how the assistant should adapt its timing, output modality, and level of proactivity to the driving situation and social context.
Mid-air haptic technology enables touchless interaction through focused ultrasound and holds promise for inclusive design in mobile and ubiquitous multimedia systems. However, accessibility aspects remain largely unexplored. This poster introduces an accessibility-centered framework for mid-air haptic interaction, building on prior reviews of user experience and technical design. We systematically analyzed 42 studies published between 2014 and 2024 and conducted twelve expert interviews, including accessibility specialists and a visually impaired participant. The analysis highlights three main challenges: low stimulus clarity, limited multimodal feedback, and insufficient personalization. Our proposed framework addresses these gaps through four design dimensions-perceptual optimization, multimodal integration, spatial configuration, and individual calibration. By integrating literature and expert insights, this work advances a structured foundation for accessible mid-air haptic systems and identifies future research directions toward inclusive, touchless interaction in mobile, public, and mixed-reality contexts.
Traditional methods for teaching electromagnetism often fall short, constrained by complex math and static 2D diagrams that fail to convey the dynamic, 3D nature of Electromagnetic fields. Even modern interactive tools frequently offer only scripted experiences, limiting exploration. This work presents a novel VR application designed to overcome these challenges by providing an interactive, real-time simulation of fundamental electromagnetic phenomena. Our core innovation is a compute shader GPU engine optimized for commodity hardware. It integrates Finite-Difference Time-Domain for field evolution with Particle-in-Cell for dynamic charges. This architecture allows for the stereoscopic visualization of complex 3D vector fields and provides low-latency user interaction. Students can directly manipulate charges and magnetic dipoles, observing the changes in the electromagnetic field. This seamless cause-and-effect loop is engineered to rapidly build physical intuition. Future work will focus on expanding the simulation features, as well as conducting empirical studies to validate the application's impact on learning.
As living and working spaces become scarce and costly, interiors transitioning between living, working, and sleeping configurations while enabling customized setups are in demand. Traditional furniture consumes space and is cumbersome to rearrange. Shape-changing furniture could solve this, yet existing options lack resolution, stability, or adaptability. We present AirClick, which facilitates on-demand room transformation using modular interactive inflatables. Our fabrication process supports personally fabricated and retrofitted retail inflatables. The touch-actuated modules connect to a floor-based air connector grid, facilitating interaction while integrating with traditional furniture. In a lab study (N=20) across four scenarios (office, meeting room, apartment room, multipurpose hall), participants rapidly transformed rooms and perceived AirClick as significantly more usable with higher intention to use in the everyday scenarios than in the hall, indicating suitability for routine activities with low to medium requirements for robustness. User feedback highlights AirClick's usefulness and scalability in diverse settings, hence showing AirClick's space-saving and customizable design can enhance the functionality and adaptability of living and working spaces.
In manufacturing environments, operators are expected to perform structured troubleshooting to ensure process stability and minimize downtime. However, the adoption of formal problem-solving protocols remains low in practice. Through a formative study in a pharmaceutical manufacturing setting, we identify key challenges in current troubleshooting practices: high cognitive load, fragmented documentation, and loss of tacit operational knowledge. We propose the exploration of a conversational AI (CAI) assistant to reframe these structured protocols as natural dialogues that support reasoning and learning rather than administrative compliance.
Delivery robots are an emerging technology that promise benefits (reducing reliance on human-driven delivery vehicles and minimizing the impact of local traffic conditions). One limitation of such robots is their restricted range. A potential solution is to integrate these robots into existing public transport (PT) systems. One underexplored direction is extending their mobility through PT like buses. This scenario introduces distinct human-robot interaction (HRI) challenges: close physical proximity, shifting attention, implicit social norms, and multiple stakeholders occupying the same confined space. With this poster we aim to stimulate a discussion on key HCI questions around intent communication, coexistence, and social legibility when robots travel alongside passengers. We outline concrete interaction scenarios (boarding, moving through aisles negotiating space and signaling actions) and use them to provoke a discussion on (the need of) eHMI design in dynamic, shared environments. Rather than proposing a finalized solution, the goal is to map open challenges and stimulate community engagement with this emerging field of HRI.
Gaze input for electric wheelchairs offers an alternative to joysticks, but conventional dwell-time methods suffer from user fatigue and the "Midas Touch" problem. Furthermore, their reliance on infrared (IR) cameras leads to high costs and poor robustness outdoors due to sunlight. This study proposes an IR-free system combining the Eye-Glance gesture method with a low-cost, visible-light (RGB) camera. While our prior work introduced this concept, its simple algorithm was limited to indoor use. We overcome this by introducing a new pupil-center detection algorithm based on image saturation, shape constraints, and blink handling, specifically designed to tolerate ambient light changes. Evaluation with 8 participants performing four gestures under indoor and outdoor conditions achieved average first-attempt success rates of 95.9% (indoors) and 94.4% (outdoors). This work demonstrates that our visible-light pipeline provides a low-cost, robust gaze interface suitable for electric wheelchairs in varied environments.
Gait is a distinctive behavioral trait, yet its vulnerability against imitation remains underexplored in immersive environments. We present a study investigating how real-time visual feedback in virtual reality (VR) influences a person's ability to mimic another's gait. Through two experiments, we first identify the most usable feedback design (N=8), then evaluate its impact on imitation performance compared to a baseline without feedback (N=18). We analyze positional and rotational similarity between participants and target avatars, examining the influence of avatar-user gender matching and repeated practice. Our findings reveal that visual feedback significantly improves rotational alignment and that practice leads to measurable improvements in mimicry accuracy. We discuss implications for avatar embodiment, personalization in VR applications, and potential considerations for behavioral biometric systems. We also contribute a publicly available dataset of gait mimicry in VR, supporting further research on motion learning and imitation.
Social media platforms have transformed the way individuals document and revisit personal experiences. We conducted a study of Instagram, one of the most popular mobile social media platforms (about 95 million images are uploaded daily) to investigates the impact of reviewing Instagram posts on episodic memory-the ability to recall specific events and details. Fifty Instagram users participated in a questionnaire-based study, where they were prompted to recall memories associated with three of their posted images across six prompts: story (the story of the event), location, time, people, before/after events, and order. Results revealed significant variations in recall accuracy: participants were better at recalling the story of the events and associated people compared to the temporal details (the timing) of the event (H = 157.04, p < 0.000). A weak but statistically significant correlation (r = 0.66, p < 0.001) was observed between the number of likes a post received and participants' ability to recall event details. Higher social engagement thus appears to enhance recall accuracy, highlighting the role of social reinforcement in memory processes. To our knowledge, this study is the first to examine the effect of the non-ephemeral platform "Instagram" use on episodic memory. While the study offers promising evidence of Instagram's cognitive benefits, further research is needed to explore long-term implications and the interplay between social engagement and episodic memory.
Embodiment in mixed reality describes the sensation of experiencing a virtual representation as an extension of one's own body. While research has extensively examined embodiment in virtual reality (VR) and head-mounted augmented reality (AR), its impact on smartphones remains underexplored. This study examines how smartphone-based AR embodiment affects user engagement and cognitive performance in a comprehension task. A study involving 24 participants explored whether using a smartphone AR face-filter to embody a virtual audience member influenced the recall of a historical speech. Findings show that participants in the AR condition scored higher on a factual quiz than those in the control group. At the same time, stronger perceived embodiment, especially self-location, was negatively associated with quiz performance, consistent with Cognitive Load Theory. These results should be interpreted cautiously: our comparison contrasted a static image (no AR) with AR that included facial embodiment, so we did not include an "AR without embodiment" condition to fully separate AR novelty from embodiment. Stimuli were also restricted to a single speech and a single historical scene presented as a static image, limiting generalizability to other content and to dynamic or interactive AR. Finally, the sample was modest (N=24), so estimates are preliminary and warrant replication. We discuss implications for designing smartphone AR that balances engagement with cognitive efficiency.
The growing accessibility of virtual reality (VR) technology presents new possibilities for individuals with physical disabilities to engage in adaptive sports training. This paper reports on a qualitative study exploring the perceptions, experiences, and feedback of athletes with physical disabilities after interacting with bespoke VR games for two adaptive sports, namely Volt Hockey and Power Soccer. Results indicated that participants view VR as a valuable tool for overcoming physical and logistical barriers to practice, enhancing skill development, and promoting sustained engagement with their sport. At the same time, participants identified key challenges related to equipment accessibility, physical comfort, input responsiveness, and the need for multiplayer functionality. This study highlights the potential of bespoke VR games to complement traditional adaptive sports training and underscores the importance of user-centered design practices in designing and developing accessible and meaningful VR experiences for people with disabilities. Our findings offer practical insights for developers, designers, and researchers aiming to support equal access to adaptive sports training through VR technology.
Large Language Models (LLMs) such as ChatGPT are increasingly used in education, but their general-purpose design limits support for conceptual learning. We present a study of a custom GPT, scaffolded with Bloom's Taxonomy and framed as an educator, to examine how structured dialogue shapes learning. Using deceptive patterns as a test domain, 38 students compared the custom GPT with ChatGPT-4 and an expert-authored website. The custom GPT led to higher satisfaction, trust, and perceived learning than ChatGPT-4, though the website remained most trusted overall. Students valued structured guidance for deeper reflection but also found it demanding, and still relied on expert sources for credibility. Our contribution is threefold: we provide evidence on how these tools differ in shaping learning experiences; show how structured prompting and role framing act as pedagogical scaffolds; and propose design directions that combine scaffolding with source cues, adapt to learner needs, and complement teaching practices.
Multimodal interfaces increasingly blend voice, touch, and other inputs to enable natural interaction. However, transitions between modalities often introduce hidden cognitive costs not captured by traditional usability metrics. This Work-in-Progress paper presents the Modality-Switch Cognitive Cost (MSCC), a validated metric for quantifying the performance and workload penalties of switching between modalities. Building on an empirical study with 65 participants performing smart assistant tasks, we observed asymmetric switching costs, with Touch -> Voice transitions producing higher cognitive load than Voice -> Touch. We discuss MSCC as a foundation for adaptive multimodal design and outline ongoing work extending the metric to AR/VR and cross-device contexts.
Virtual museum tours are increasingly used to expand access to cultural heritage, yet their experiential equivalence to physical visits remains unclear. This study compared user experience in physical and virtual tours of the same museum exhibition. Thirty-two participants explored either the physical exhibition or an photo-based 360 degrees virtual tour and completed standardized and custom questionnaires. No significant differences were observed on experiential scales between conditions. However, results revealed distinct spatial usability effects: the physical tour yielded higher ratings for navigation ease, orientation, and backtracking. These results indicate that high-quality virtual tours may replicate some experiential aspects of physical visits but still present challenges in spatial clarity and intuitive wayfinding. The findings contribute insight into where current virtual museum technologies show similarities and differences in navigation, satisfaction, and accessibility in digital heritage experiences.
Touch-sensitive surfaces offer an intuitive and flexible form of interaction, for example through gesture input. Despite being the primary input modality for mobile devices, they hardly find application in desktop settings. At the same time, the computer mouse is still the most efficient and accurate input device for pointing. Consequently, keeping the unmatched functionality of a mouse but extending it with new input options via a touch-sensitive surface, is a promising approach. While research prototypes and niche products for multi-touch mice exist, the concept has not yet become established. In this work, we follow a user-centered approach towards touch interaction on computer mice. In a user study (n=12), we identified which areas on the mouse are suitable for touch input. Further, we explored potential usage scenarios in a diary study (n=11) and intuitive gestures in an elicitation study (n=10). We compile our findings into a gesture set which future research can build upon to implement touch interaction on computer mice.
The rapid turnover of mobile devices, driven by social pressure and frequent market releases, leads to large numbers of smartphones becoming outdated yet still technically capable, retaining considerable computational power, memory, and sensing capabilities. We propose the Internet of Legacy Devices (I-oLD), an open framework designed to abstract the heterogeneity of old hardware, aiming to recycle and repurpose these devices, leveraging their residual computing power, sensors, and displays for distributed computing, modular interfaces, and pervasive sensing. This approach reduces electronic waste, promoting sustainable "green" computing, and provides low-cost hardware solutions for researchers, developers, and users worldwide.
The increasing integration of artificial intelligence (AI) in complex domains like semiconductor design requires a deeper understanding of human factors. Typically, chip design tools rely on manual parameter tuning, which demands extensive domain expertise and is susceptible to robustness and interpretability issues. This paper presents the results of a user-centered design process applied to re-designing the user interface (UI) of AIMS, a simulation management tool used to identify failure conditions in integrated circuits design. An iterative approach, including expert review, interviews, and an evaluation workshop, was adopted. Results show improvements in usability, usefulness, and perceived reliability, particularly concerning the integration of AI-powered anomaly detection features. The evaluation also highlighted areas for further refinement, stressing the need for continuous user-centered design efforts.
Thermal imaging plays a vital role in search and rescue, yet little is known about how color palette selection affects human recognition performance. While AI-based detection dominates current research, manual identification remains crucial in unpredictable or ethically sensitive conditions. This study evaluates three commonly used thermal palettes, White Hot, Ironbow, and Rainbow, using a recently available dataset of challenging aerial images. In an online user study with 75 participants, task time, hit accuracy, and not found responses were measured. Results show that White Hot enabled faster and more accurate performance, while Rainbow significantly hindered detection. These findings offer practical guidance for designing thermal interfaces and highlight the importance of perceptual optimization in augmented vision systems.
As cities explore intelligent technologies to support urban living, youth perspectives remain underrepresented in design processes despite their role as future residents. This study explores how young people (aged 13-17) envision intelligent technologies that support neighbourhood communality in the emerging Nordic Superblock urban model. Through four co-design workshops, we investigated youth perspectives on shared spaces, social interaction, and community participation, applying a multi-level framework of micro, macro-, and meta-level considerations. The findings highlight youth's expectations for intelligent technologies that organise and optimise shared space use, facilitate social interaction, and support collective decision-making, such as planning of shared spaces. The study identifies concerns regarding privacy, inclusivity, and over-reliance on AI. The study presents four initial technology concepts for advancing communality, ranging from AI-facilitated shared dining to organising events and space use. This paper contributes to ubiquitous technology design understanding by foregrounding youth perspectives to community-centric intelligent neighbourhood technologies that can support residents' wellbeing.
