Delays in direct teleoperation decouple operator input from robot feedback. We frame this not as a unitary problem but as three facets of operator uncertainty: (1) communication, when commands take effect, (2) trajectory, how inputs map to motion, and (3) environmental, how external factors alter outcomes. We externalized each facet through predictive visualizations: Network, Path, and Envelope. In a controlled study with 24 participants (novices in telerobotics) navigating a simulated robot under a fixed 2.56 s round-trip delay, we compared these visualizations against a delayed-video baseline. Path significantly shortened task time, lowered perceived cognitive load, and reduced reliance on reactive “move-and-wait” behavior. Envelope lowered cognitive load but did not significantly reduce reactive behavior or improve performance, while Network had no measurable effect. These results indicate that predictive support is effective only when trajectory uncertainty is externalized, enabling operators to move from reactive to more proactive control.

Virtual Reality (VR) environments challenge users with varied input devices, interaction methods, and interface designs, resulting in a steep learning curve. Feedforward "informs the user about what the result of his action will be", and it allows to ease the process of learning of the end user by providing ways to represent the required action to perform, using contextualised previews that show how to complete a given interaction. Yet, creating effective direct feedforward without specialised tools remains a tedious process. We present BeatriXR, a flexible and adaptive toolkit that simplifies the creation of direct-feedforward configurations in VR. It enables users to build, visualise, and customise direct feedforward using virtual avatars, offering both in-world representations and on-screen comparisons of interaction options. Mapped to the recognised design-space by Muresan et al. (including Triggering, Previewing, and Exiting phases), BeatriXR streamlines development and helps ensure effective feedforward integration in VR environments. As additional support for the feedforward design task, BeatriXR features an LLM-powered decision-support layer that suggests possible configuration options. This guidance helps designers select optimal settings during development for adapting the presentation to user needs and the expected configuration of the interaction context. We conducted an exploratory review with XR domain experts who rated the UI for modifying feedforward settings, and assessed four LLM models in the task of suggesting possible configuration options. The participants’ feedback was positive and provided valuable insights for improving both the user interface, which was generally perceived positively, and the quality of the LLM-generated responses.

Machine learning systems are often inspected through 2D projections of high-dimensional representations using techniques such as t-SNE or UMAP. While these visualizations provide useful overviews of clustering and similarity, they are inherently static: they display only the existing data points and do not allow users to interactively explore a model’s decision space. We present an interactive exploration system — LAPEX — that uses a Variational Autoencoder (VAE) as a generative proxy over a model’s training distribution, turning the latent space into a navigable workspace for model sensemaking. Unlike static embeddings, the proxy provides an explicit decoding path from latent coordinates to inputs, enabling interaction patterns such as continuous sampling, interpolation between anchors, and region probing. We operationalize these capabilities through a set of interactive probes that augment a familiar scatter-plot overview with generative overlays for comparing transitions between classes and examining sparsely populated regions. A within-subject formative study ( N = 16) comparing an interactive VAE-based method to a static t-SNE baseline shows that generative interaction substantially improves counterfactual reasoning and influences how users assess model behavior in sparse or uncertain regions, while static embeddings sometimes provide clearer boundary perception. From these findings, we derive concrete design guidelines and architectural considerations for engineering interactive AI model exploration systems using generative latent representations.

This paper introduces ECHO, an LLM-powered system framework to explore and interrogate the internals of AI models through tool-augmented language models. While traditional XAI methods typically offer a small and technical set of explanation types, ECHO advances the accessibility and usability of AI explanations through a conversational approach, combining LLMs with a collection of tools and a human-in-the-loop process. We identify various explanation types from the literature, for which we create a set of predefined tools for tabular data. Using a modular architecture, ECHO integrates these predefined tools with dynamically generated tools to interact with AI models, facilitating tailored explanations for a large variety of user queries. This paper details ECHO’s design, implementation, and use cases, demonstrating its capabilities in the context of a movie recommender, healthcare decision tree and neural network for educational classification.

In recent years, the integration of Artificial Intelligence (AI) has sparked revolutionary progress across diverse domains, with sports applications being no exception. At the same time, using real-world data sources, such as GPS, weather, and traffic data, offers opportunities to improve the overall user engagement and effectiveness of such applications. Despite the substantial advancements, including proven success in mobile applications, there remains an untapped potential in leveraging these technologies to boost motivation and enhance social group dynamics in Virtual Reality (VR) sports solutions. Our innovative approach focuses on harnessing the power of AI and real-world data to facilitate the design of such VR systems. To validate our methodology, we conducted an exploratory study involving 18 participants, evaluating our approach within the context of indoor VR cycling. By incorporating GPX files and omnidirectional video (real-world data), we recreated a lifelike cycling environment in which users can compete with simulated cyclists navigating a chosen (real-world) route. Considering the user’s performance and interactions with other cyclists, our system employs AI-driven natural language processing tools to generate encouraging and competitive messages automatically. The outcome of our study reveals a positive impact on motivation, competition dynamics, and the perceived sense of group dynamics when using real performance data alongside automatically generated motivational messages. This underscores the potential of AI-driven enhancements in user interfaces to not only optimize performance but also foster a more engaging and supportive sports environment.

Recent HCI research has shown significant interest in investigating digital working instructions for guiding novices to perform manual tasks. While performance enhancement has been a primary focus, it is increasingly recognized that technology’s impact extends beyond objective metrics. Trainee motivation and engagement plays a pivotal role in enhancing learning outcomes and effectiveness. This paper investigates the utilization of principles from Self Determination Theory–clear attainable goals, meaningful rationale, and perspective taking–in designing multimedia instructions to enhance novice users’ indicators of psychological well-being. We present findings from an experiment involving real-world woodworking, where novice users, in a between-subjects study, followed interactive, in-situ projection-based guidance. Results demonstrate that adhering to SDT postulates can positively influence perceived competence, intrinsic motivation and task execution quality. These findings offer valuable insights for designing digital instructions to guide and train novices, emphasizing the importance of psychological well-being alongside task performance.

Augmented Reality (AR) holds significant potential to facilitate users in executing manual tasks. For effective support, however, we need to understand how showing movement instructions in AR affects how well people can follow those movements in real life. In this paper, we examine the degree to which users can synchronize the speed of their movements with speed cues presented through an AR environment. Specifically, we investigate the effects of timing in AR visual guidance. We assess performance using a highly realistic Mixed Reality (MR) welding simulation. Welding is a task that requires very precise timing and control over hand and arm motion. Our results show that upfront visual guidance (before manual task execution) alone often fails to transfer the knowledge of intended speeds, especially at higher target speeds. Live guidance (during manual task execution) during the activity provides more accurate speed results but typically requires a higher overshoot at the start. Optimal outcomes occur when visual guidance appears upfront and continues during the activity for users to follow through.

Cleanroom cleaning is a surface coverage task where the pattern should be followed correctly, and the entire surface should be covered. We investigate the efficacy of augmented reality (AR) by implementing various pattern guidance designs to enhance a cleanroom cleaning task. We developed an AR guidance system for cleaning procedures and evaluated four distinct pattern guidance methods: (1) breadcrumbs, (2) examples, (3) middle lines, and (4) outlines. We vary the instructions on the entire surface or as a single step. To measure performance, accuracy, and user satisfaction associated with each guidance method, we conducted a large-scale (n=864) between-subjects study. Our findings indicate that single step instructions proved to be more intuitive and efficient than full instructions, especially for the breadcrumbs. We also discussed the implications of our results for the development of AR applications for surface coverage and pattern optimization.

Navigating the interaction landscape of Virtual Reality (VR) and Augmented Reality (AR) presents significant complexities due to the plethora of available input hardware and interaction modalities, compounded by spatially diverse visual interfaces. Such complexities elevate the likelihood of user errors, necessitating frequent backtracking. To address this, we introduce ViRgilites, a virtual guidance framework that delivers multi-level feedforward information covering the available interaction techniques as well as the future possibilities to interact with virtual objects, anticipating the interaction effects and how they fit with the overall user's goal. ViRgilites is engineered to facilitate task execution, empowering users to make informed decisions about action methodologies and alternative courses of action. This paper presents the architecture and functionality of ViRgilites and demonstrates its efficacy through evaluation with a formative user study.

In many jobs, workers execute precise line tracing tasks; welding, spray painting, or chiseling, for example. Training and support for such tasks can be done using VR and AR. However, to enable workers to achieve the required precision in movement and timing, the effect of visual guidance on continuous movement needs to be explored. In VR environments, we want to ensure people are trained so that the obtained skill is transferable to a real-world context, whereas, in AR, we want to ensure an ongoing task can be completed successfully when adding visual guidance. To simulate these various contexts, we employ a VR environment to investigate the effectiveness of different visualizations for motion-based guidance in a line tracing task. We tested five different visualizations, including faster and slower arrows on the pen, the same arrows on the line, a dynamic graph on the pen or line, and a ghost object to follow. Each visualization was tested with the same set of five lines of different target speeds (2cm/s to 10 cm/s in steps of 2 cm/s) with a training line of 5 cm/s. Our results show that the example ghost on the line turns out to be the most efficient visualization for allowing users to achieve a specific speed. Users also perceived this visualization as the most engaging and easy to use. These findings have significant implications for the development of AR-based guidance systems, specifically in the realm of speed control, across diverse domains such as industrial applications, training, and entertainment.

Operators’ well-being is a key factor for the success of industrial production processes. Even though research has studied the well-being aspects of the industry, such as support and improvement of ergonomics, there is still a long way to go to achieve a sustainable and healthy work context for manufacturing industry. We believe the Human-Computer Interaction community can contribute by developing research on worker well-being in real-life settings. This workshop intends to offer a venue for HCI researchers that focus on worker well-being for the manufacturing industry and other industry domains.

After a cardiac event, secondary prevention is recommended to foster recovery and reduce the risk of recurrent events. European guidelines and EAPC position statements on prevention of cardiovascular diseases recommend a holistic approach that actively engages patients by using shared decision making (SDM). It has been demonstrated that telerehabilitation can be a feasible and effective add-on or alternative compared to conventional in-hospital secondary prevention. However, till date, there is no eHealth solution that offers a holistic approach for secondary prevention that includes SDM. In this paper, we present the CoroPrevention-SDM approach, a technology-supported shared decision making approach for a comprehensive secondary prevention program for cardiac patients. The CoroPrevention Tool Suite consists of three applications that support patients and caregivers in following this approach: 1) a caregiver dashboard that includes decision support systems and supports SDM, 2) a patient mobile application that supports patients in making behaviour changes in their daily life, and 3) an extended ePRO application that collects patient reported outcomes and patient preferences. In a formative usability study, we assessed patients' and caregivers' opinion about our approach. The study indicated that both are willing to use our proposed approach to collaboratively set behavioural goals during SDM encounters.

Over-the-top (OTT) streaming services like YouTube and Netflix induce massive amounts of video data, hereby putting substantial pressure on network infrastructure. This paper describes a demonstration of the object-based video (OBV) methodology that allows for the quality-variant MPEG-DASH streaming of respectively the background and foreground object(s) of a video scene. The OBV methodology is inspired by research into human visual attention and foveated compression, in that it allows to adaptively and dynamically assign bitrate to those portions of the visual scene that have the highest utility in terms of perceptual quality. Using a content corpus of interview-like video footage, the described demonstration proves the OBV methodology's potential to downsize video bitrate requirements while incurring at most marginal perceptual impact (i.e., in terms of subjective video quality). Thanks to its standards-compliant Web implementation, the OBV methodology is directly and broadly deployable without requiring capital expenditure.

Over-the-top (OTT) streaming services like YouTube and Netflix induce massive amounts of video traffic. To combat the resulting network load, this article empirically explores the use of the object-based video (OBV) methodology that allows for the quality-variant HTTP Adaptive Streaming of respectively the background and foreground object(s) of a video scene. In particular, we study two alternative video object representation methods where the first meticulously follows the object contour, while the second uses axis-aligned bounding box enclosures. We subjectively compare both techniques to traditional, frame-based video compression in the context of live action content featuring talking persons. The resulting mixed methods data shows that (i) OBV-informed users tolerate substantial background quality degradations, and (ii) at an average bitrate reduction of 14 percent, perceptual differences between respectively contour-based OBV and traditional encoding are small or even non-existing for the non-movie content in our corpus. Although our evaluation focuses on interview-like footage, our qualitative data hints that the presented results might be extrapolatable to other video genres. As such, our findings inform content owners and network operators about video bitrate saving opportunities with marginal perceptual impact.

Keeping track of design processes is a cumbersome task due to the apparently unconstrained and unstructured nature of creative work. Traceability is fundamental to revisit and reflect on the design narratives that describe artefact evolution. In this paper, we aim to identify what characteristics are necessary to facilitate traceability of creative design processes. For this end, we use a functional prototype to connect artefacts, design rationale, and decisions in a shared workspace. We evaluated this prototype for 15 weeks with six pairs of students engaged in a user-centered design project. Our findings show that having a lean repository of artefacts annotated with design rationale can facilitate tracking progress in different phases of the process. We found that creating a record of the participants' design work is useful to reflect on and for team agreement, ensure consistency of evolving artefacts, and help in planning future steps in the design project.

In this paper, we set out to find what encompasses an appropriate visual language for information presented on near-eye out-of-focus displays. These displays are positioned in a user's peripheral view, very near to the user's eyes, for example on the inside of the temples of a pair of glasses. We explored the usable display area, the role of spatial and retinal variables, and the influence of motion and interaction for such a language. Our findings show that a usable visual language can be accomplished by limiting the possible shapes and by making clever use of orientation and meaningful motion. We found that especially motion is very important to improve perception and comprehension of what is being displayed on near-eye out-of-focus displays, and that perception is further improved if direct interaction with the content is allowed.

The bulk of contemporary online video traffic is encoded in a traditional manner, hereby neglecting most, if not all, of the semantics of the underlying visual scene. One essential piece of semantic information in the context of video streaming is awareness of the objects that jointly constitute the scene. A canonical example of a benefit associated with such object awareness is the ability to subdivide a video fragment in respectively a background and one or more foreground objects. This paper reports on a pragmatically designed video streaming approach that exploits object-related knowledge in order to improve the real-time adaptability of video streaming sessions (manifested in the form of increased granularity in terms of streaming quality control). The proposed approach is completely compliant with present-day video codecs and HTTP Adaptive Streaming schemes, most notably H.264 and MPEG-DASH. Findings from subjecting the proposed video streaming technique to a comparative subjective evaluation suggest that scenarios exist where the presented approach holds the capacity to improve on traditional streaming in terms of user-perceived video quality.

Omni-directional video (ODV) is a novel medium that offers viewers a 360º panoramic recording. This type of content will become more common within our living rooms in the near future, seeing that immersive displaying technologies such as 3D television are on the rise. However, little attention has been given to how to interact with ODV content. We present a gesture elicitation study in which we asked users to perform mid-air gestures that they consider to be appropriate for ODV interaction, both for individual as well as collocated settings. We are interested in the gesture variations and adaptations that come forth from individual and collocated usage. To this end, we gathered quantitative and qualitative data by means of observations, motion capture, questionnaires and interviews. This data resulted in a user-defined gesture set for ODV, alongside an in-depth analysis of the variation in gestures we observed during the study.