AI assistants are transforming creative and knowledge domains, holding similar promise for mechanical design via 3D CAD software. Yet, current AI assistance for CAD relies on geometry or command history, lacking rich design intent. We investigate think-aloud computing as a lightweight approach to capture designers’ spoken intent and inform how future AI assistance could leverage this to provide in-situ feedback. Through a three-part study with 10 designers and 10 experts, we (1) recorded designers’ think-aloud verbalizations during 3D modelling, (2) compared expert feedback with and without think-aloud recordings, and (3) interviewed the original designers to evaluate feedback quality. Findings show that verbalizations surface rationale, future actions, and challenges — insights absent from geometric and command data — that enable feedback attuned to designers’ goals. By harnessing think-aloud data, we uncover when to intervene, what to prompt, and characteristics of effective feedback, paving the way for context-aware AI assistance for CAD.

Design thrives on feedback. However, gathering constant feedback throughout the design process can be labor-intensive and disruptive. We explore how AI can bridge this gap by providing effortless, ambient feedback. We introduce FeedQUAC, a lightweight design companion that delivers real-time, read-aloud, AI-generated commentary from diverse personas based on live screenshots of the designer’s workspace. FeedQUAC is always available, context-aware, ambient, playful, and iteration-aware. In a design probe with eight 3D CAD designers, participants highlighted convenience, playfulness, confidence boosts, and inspiration. Our findings suggest that ambient interaction is a valuable consideration for both designing and evaluating future creativity support systems.

Generative AI tools are increasingly being integrated into various workflows. Many powerful AI systems, such as ChatGPT, Gemini, and Grok, are accessed through simple chat interfaces. While these language-based interactions may seem intuitive and futuristic, they also bear a resemblance to early computer command lines. With our prototype system, Curompt (combined cursor+prompt), we explore how to integrate conversational interaction powered by generative AI into familiar graphical interfaces with direct manipulation. We focused on the 3D environment, where relying on language-based interactions alone proves to be challenging.

Tutorial videos are a popular help source for learning feature-rich software. However, getting quick answers to questions about tutorial videos is difficult. We present an automated approach for responding to tutorial questions. By analyzing 633 questions found in 5,944 video comments, we identified different question types and observed that users frequently described parts of the video in questions. We then asked participants (N=24) to watch tutorial videos and ask questions while annotating the video with relevant visual anchors. Most visual anchors referred to UI elements and the application workspace. Based on these insights, we built AQuA, a pipeline that generates useful answers to questions with visual anchors. We demonstrate this for Fusion 360, showing that we can recognize UI elements in visual anchors and generate answers using GPT-4 augmented with that visual information and software documentation. An evaluation study (N=16) demonstrates that our approach provides better answers than baseline methods.

Text-to-image AI are capable of generating novel images for inspiration, but their applications for 3D design workflows and how designers can build 3D models using AI-provided inspiration have not yet been explored. To investigate this, we integrated DALL-E, GPT-3, and CLIP within a CAD software in 3DALL-E, a plugin that generates 2D image inspiration for 3D design. 3DALL-E allows users to construct text and image prompts based on what they are modeling. In a study with 13 designers, we found that designers saw great potential in 3DALL-E within their workflows and could use text-to-image AI to produce reference images, prevent design fixation, and inspire design considerations. We elaborate on prompting patterns observed across 3D modeling tasks and provide measures of prompt complexity observed across participants. From our findings, we discuss how 3DALL-E can merge with existing generative design workflows and propose prompt bibliographies as a form of human-AI design history.

We explore the design and utility of situated manual self-tracking visualizations on dedicated displays that integrate data tracking into existing practices and physical environments. Situating self-tracking tools in relevant locations is a promising approach to enable reflection on and awareness of data without needing to rely on sensorized tracking or personal devices. In both a long-term autobiographical design process and a co-design study with six participants, we rapidly prototyped and deployed 30 situated self-tracking applications over a ten month period. Grounded in the experience of designing and living with these trackers, we contribute findings on logging and data entry, the use of situated displays, and the visual design and customization of trackers. Our results demonstrate the potential of customizable dedicated self-tracking visualizations that are situated in relevant physical spaces, and suggest future research opportunities and new potential applications for situated visualizations.

Despite an abundance of carefully-crafted tutorials, trial-and-error remains many people’s preferred way to learn complex software. Yet, approaches to facilitate trial-and-error (such as tooltips) have evolved very little since the 1980s. While existing mechanisms work well for simple software, they scale poorly to large feature-rich applications. In this paper, we explore new techniques to support trial-and-error in complex applications. We identify key benefits and challenges of trial-and-error, and introduce a framework with a conceptual model and design space. Using this framework, we developed three techniques: ToolTrack to keep track of trial-and-error progress; ToolTrip to go beyond trial-and-error of single commands by highlighting related commands that are frequently used together; and ToolTaste to quickly and safely try commands. We demonstrate how these techniques facilitate trial-and-error, as illustrated through a proof-of-concept implementation in the CAD software Fusion 360. We conclude by discussing possible scenarios and outline directions for future research on trial-and-error.

User modeling is crucial to understanding user behavior and essential for improving user experience and personalized recommendations. When users interact with software, vast amounts of command sequences are generated through logging and analytics systems. These command sequences contain clues to the users’ goals and intents. However, these data modalities are highly unstructured and unlabeled, making it difficult for standard predictive systems to learn from. We propose SimCURL, a simple yet effective contrastive self-supervised deep learning framework that learns user representation from unlabeled command sequences. Our method introduces a user-session network architecture, as well as session dropout as a novel way of data augmentation. We train and evaluate our method on a real-world command sequence dataset of more than half a billion commands. Our method shows significant improvement over existing methods when the learned representation is transferred to downstream tasks such as experience and expertise classification.

Situated visualization is an emerging concept within visualization, in which data is visualized in situ, where it is relevant to people. The concept has gained interest from multiple research communities, including visualization, human-computer interaction (HCI) and augmented reality. This has led to a range of explorations and applications of the concept, however, this early work has focused on the operational aspect of situatedness leading to inconsistent adoption of the concept and terminology. First, we contribute a literature survey in which we analyze 44 papers that explicitly use the term "situated visualization" to provide an overview of the research area, how it defines situated visualization, common application areas and technology used, as well as type of data and type of visualizations. Our survey shows that research on situated visualization has focused on technology-centric approaches that foreground a spatial understanding of situatedness. Secondly, we contribute five perspectives on situatedness (space, time, place, activity, and community) that together expand on the prevalent notion of situatedness in the corpus. We draw from six case studies and prior theoretical developments in HCI. Each perspective develops a generative way of looking at and working with situatedness in design and research. We outline future directions, including considering technology, material and aesthetics, leveraging the perspectives for design, and methods for stronger engagement with target audiences. We conclude with opportunities to consolidate situated visualization research.

People’s physical movement and body language implicitly convey what they think and feel, are doing or are about to do. In contrast, current smart speakers miss out on this richness of body language, primarily relying on voice commands only. We present QUBI, a dynamic smart speaker that leverages expressive physical motion – stretching, nodding, turning, shrugging, wiggling, pointing and leaning forwards/backwards – to convey cues about its underlying behaviour and activities. We conducted a qualitative Wizard of Oz lab study, in which 12 participants interacted with QUBI in four scripted scenarios. From our study, we distilled six themes: (1) mirroring and mimicking motions; (2) body language to supplement voice instructions; (3) anthropomorphism and personality; (4) audio can trump motion; (5) reaffirming uncertain interpretations to support mutual understanding; and (6) emotional reactions to QUBI’s behaviour. From this, we discuss design implications for future smart speakers.

Cross-device interactions enable ad hoc sharing of content and control in co-located collaboration. Cross-device research often draws from proxemics theory for designing interactions based on detection of spatial relations such as distance and orientation between people and devices. However, detection of human-human or human-device proximity also constrains flexibility in co-located social interaction. We suggest a proxemics-based approach to designing flexible cross-device interactions. From observations in a field study, we articulate how co-located sharing practices are shaped by the interplay between everyday mobile devices and the physical environment. Based on these insights, we present three cross-device prototypes as proofs-of-concept, demonstrating three design sensitivities for considering proxemics beyond proximity; incorporating features in the environment, enabling flexibility in interpersonal distance and orientation, and providing multiple alternative action possibilities. Drawing from characteristics of our prototypes, we discuss concrete proposals for designing cross-device interactions to enable flexible social interaction.

We report on findings from seven design workshops that used ideation and sketching activities to prototype new situated visualizations - representations of data that are displayed in proximity to the physical referents (such as people, objects, and locations) to which the data is related. Designing situated visualizations requires a fine-grained understanding of the context in which the visualizations are placed, as well as an exploration of different options for placement and form factors, which existing methods for visualization design do not account for. Focusing on small displays as a target platform, we reflect on our experiences of using a diverse range of sketching activities, materials, and prompts. Based on these observations, we identify challenges and opportunities for sketching and ideating situated visualizations. We also outline the space of design activities for situated visualization and highlight promising methods for both designers and researchers.

Astral is a prototyping tool for authoring mobile and smart object interactive behaviours. It mirrors selected display contents of desktop applications onto mobile devices (smartphones and smartwatches), and streams/remaps mobile sensor data to desktop input events (mouse or keyboard) to manipulate selected desktop contents. This allows designers to use familiar desktop applications (e.g. PowerPoint, AfterEffects) to prototype rich interactive behaviours. Astral combines and integrates display mirroring, sensor streaming and input remapping, where designers can exploit familiar desktop applications to prototype, explore and fine-tune dynamic interactive behaviours. With Astral, designers can visually author rules to test real-time behaviours while interactions take place, as well as after the interaction has occurred. We demonstrate Astral's applicability, workflow and expressiveness within the interaction design process through both new examples and replication of prior approaches that illustrate how various familiar desktop applications are leveraged and repurposed.

We investigate how people discover the functionality of an interactive visualization that was designed for the general public. While interactive visualizations are increasingly available for public use, we still know little about how the general public discovers what they can do with these visualizations and what interactions are available. Developing a better understanding of this discovery process can help inform the design of visualizations for the general public, which in turn can help make data more accessible. To unpack this problem, we conducted a lab study in which participants were free to use their own methods to discover the functionality of a connected set of interactive visualizations of public energy data. We collected eye movement data and interaction logs as well as video and audio recordings. By analyzing this combined data, we extract exploration strategies that the participants employed to discover the functionality in these interactive visualizations. These exploration strategies illuminate possible design directions for improving the discoverability of a visualization's functionality.

Because gesture design for augmented reality (AR) remains idiosyncratic, people cannot necessarily use gestures learned in one AR application in another. To design discoverable gestures, we need to understand what gestures people expect to use. We explore how the scale of AR affects the gestures people expect to use to interact with 3D holograms. Using an elicitation study, we asked participants to generate gestures in response to holographic task referents, where we varied the scale of holograms from desktop-scale to room-scale objects. We found that the scale of objects and scenes in the AR experience moderates the generated gestures. Most gestures were informed by physical interaction, and when people interacted from a distance, they sought a good perspective on the target object before and during the interaction. These results suggest that gesture designers need to account for scale, and should not simply reuse gestures across different hologram sizes.

Advances in artificial intelligence, sensors and big data management have far-reaching societal impacts. As these systems augment our everyday lives, it becomes increasing-ly important for people to understand them and remain in control. We investigate how HCI researchers can help to develop accountable systems by performing a literature analysis of 289 core papers on explanations and explaina-ble systems, as well as 12,412 citing papers. Using topic modeling, co-occurrence and network analysis, we mapped the research space from diverse domains, such as algorith-mic accountability, interpretable machine learning, context-awareness, cognitive psychology, and software learnability. We reveal fading and burgeoning trends in explainable systems, and identify domains that are closely connected or mostly isolated. The time is ripe for the HCI community to ensure that the powerful new autonomous systems have intelligible interfaces built-in. From our results, we propose several implications and directions for future research to-wards this goal.

Toolkit research plays an important role in the field of HCI, as it can heavily influence both the design and implementation of interactive systems. For publication, the HCI community typically expects toolkit research to include an evaluation component. The problem is that toolkit evaluation is challenging, as it is often unclear what 'evaluating' a toolkit means and what methods are appropriate. To address this problem, we analyzed 68 published toolkit papers. From our analysis, we provide an overview of, reflection on, and discussion of evaluation methods for toolkit contributions. We identify and discuss the value of four toolkit evaluation strategies, including the associated techniques that each employs. We offer a categorization of evaluation strategies for toolkit researchers, along with a discussion of the value, potential limitations, and trade-offs associated with each strategy.

One potential method of improving the efficiency of human-computer interaction is to display information subliminally. Such information cannot be recalled consciously, but has some impact on the perceiver. However, it is not yet clear whether people can extract meaning from subliminal presentation of information in mobile contexts. We therefore explored subliminal semantic priming on smartphones. This builds on mixed evidence for subliminal priming across HCI in general, and mixed evidence for the effect of subliminal affective priming on smartphones. Our semi-controlled experiment (n=103) investigated subliminal processing of numerical information on smartphones. We found evidence that concealed transfer of information is possible to a very limited extent, but little evidence of a semantic effect. Overall, the impact is effectively negligible for practical applications. We discuss the implications of our results for real-world deployments and outline future research themes as HCI moves beyond mobile.

Digital behaviour change interventions, particularly those using pervasive computing technology, hold great promise in supporting users to change their behaviour. However, most interventions fail to take habitual behaviour into account, limiting their potential impact. This failure is partly driven by a plethora of overlapping behaviour change theories and related strategies that do not consider the role of habits. We critically review the main theories and models used in the research to analyse their application to designing effective habitual behaviour change interventions. We highlight the potential for Dual Process Theory, modern habit theory, and Goal Setting Theory, which together model how users form and break habits, to drive effective digital interventions. We synthesise these theories into an explanatory framework, the Habit Alteration Model, and use it to outline the state of the art. We identify the opportunities and challenges of habit-focused interventions.

We extend the popular brushing and linking technique by incorporating personal agency in the interaction. We map existing research related to brushing and linking into a design space that deconstructs the interaction technique into three components: source (what is being brushed), link (the expression of relationship between source and target), and target (what is revealed as related to the source). Using this design space, we created MyBrush, a unified interface that offers personal agency over brushing and linking by giving people the flexibility to configure the source, link, and target of multiple brushes. The results of three focus groups demonstrate that people with different backgrounds leveraged personal agency in different ways, including performing complex tasks and showing links explicitly. We reflect on these results, paving the way for future research on the role of personal agency in information visualization.

Subliminal priming has the potential to influence people's attitudes and behaviour, making them prefer certain choices over others. Yet little research has explored its feasibility on smartphones, even though the global popularity and increasing use of smartphones has spurred interest in mobile behaviour change interventions. This paper addresses technical, ethical and design issues in delivering mobile subliminal priming. We present three explorations of the technique: a technical feasibility study, and two participant studies. A pilot study (n=34) explored subliminal goal priming in-the-wild over 1 week, while a semi-controlled study (n=101) explored the immediate effect of subliminal priming on 3 different types of stimuli. We found that although subliminal priming is technically possible on smartphones, there is limited evidence of impact on changes in how much stimuli are preferred by users, with inconsistent effects across stimuli types. We discuss the implications of our results and directions for future research.

An increasing share of our daily interactions with others is mediated through mobile communication technologies. People communicate via text, emoticons, emojis and rich media such as video. We explore the design of Heartefacts, short video clips composed of highlights determined by heart rate changes while watching videos. Our survey investigated video sharing behaviour, and our feasibility study examined the possibility of detecting highlights in videos by monitoring people's heart rates measured with off-the-shelf wrist-worn sensors. Our results show that people do indeed have measurable responses with respect to their heartbeat patterns to six different emotions elicited by video clips. We compare video highlights verbally identified by our participants to physiological highlights as indicated by their heart rate data and also discuss and compare the automatically generated Heartefacts with video highlights created by an expert in video art. We close with design considerations for Heartefacts in mobile technology.

Notifications are extremely beneficial to users, but they often demand their attention at inappropriate moments. In this paper we present an in-situ study of mobile interruptibility focusing on the effect of cognitive and physical factors on the response time and the disruption perceived from a notification. Through a mixed method of automated smartphone logging and experience sampling we collected 10372 in-the-wild notifications and 474 questionnaire responses on notification perception from 20 users. We found that the response time and the perceived disruption from a notification can be influenced by its presentation, alert type, sender-recipient relationship as well as the type, completion level and complexity of the task in which the user is engaged. We found that even a notification that contains important or useful content can cause disruption. Finally, we observe the substantial role of the psychological traits of the individuals on the response time and the disruption perceived from a notification.

In this workshop, we will review and discuss open issues, technical challenges and conceptual models for multi-device spatial or proxemic interaction. We aim to bring together researchers, students and practitioners working on technical infrastructures, studies and designs of spatial in-terfaces, or domain specific multi-device applications that use space as a unit of analysis. We focus specifically on analysing how such interfaces, tools and tracking technolo-gy can be deployed "in the wild". The workshop will facili-tate knowledge exchange about the current state of spatial and proxemic interactive systems, identify application do-mains and enabling technologies for cross-surface interac-tions in the wild, and establish a research community to develop effective strategies for successful design of cross-device interactions.

Proxemics theory explains peoples' use of interpersonal distances to mediate their social interactions with others. Within Ubicomp, proxemic interaction researchers argue that people have a similar social understanding of their spatial relations with nearby digital devices, which can be exploited to better facilitate seamless and natural interactions. To do so, both people and devices are tracked to determine their spatial relationships. While interest in proxemic interactions has increased over the last few years, it also has a dark side: knowledge of proxemics may (and likely will) be easily exploited to the detriment of the user. In this paper, we offer a critical perspective on proxemic interactions in the form of dark patterns: ways proxemic interactions can be misused. We discuss a series of these patterns and describe how they apply to these types of interactions. In addition, we identify several root problems that underlie these patterns and discuss potential solutions that could lower their harmfulness.

The misalignment between the timeframe of systems and that of their users can cause problems, especially when the system relies on implicit interaction. It makes it hard for users to understand what is happening and leaves them little chance to intervene. This paper introduces the design concept of slow-motion feedback, which can help to address this issue. A definition is provided, together with an overview of existing applications of this technique.

A considerable amount of research has been carried out towards making long-standing smart home visions technically feasible. The technologically augmented homes made possible by this work are starting to become reality, but thus far living in and interacting with such homes has introduced significant complexity while offering limited benefit. As these technologies are increasingly adopted, the knowledge we gain from their use suggests a need to revisit the opportunities and challenges they pose. Synthesizing a broad body of research on smart homes with observations of industry and experiences from our own empirical work, we provide a discussion of ongoing and emerging challenges, namely challenges for meaningful technologies, complex domestic spaces, and human-home collaboration. Within each of these three challenges we discuss our visions for future smart homes and identify promising directions for the field.

Modern knowledge work consists of both individual and highly collaborative activities that are typically composed of a number of configuration, coordination and articulation processes. The desktop interface today, however, provides very little support for these processes and rather forces knowledge workers to adapt to the technology. We introduce co-Activity Manager, an activity-centric desktop system that (i) provides tools for ad hoc dynamic configuration of a desktop working context, (ii) supports both explicit and implicit articulation of ongoing work through a built-in collaboration manager and (iii) provides the means to coordinate and share working context with other users and devices. In this paper, we discuss the activity theory informed design of co-Activity Manager and report on a 14 day field deployment in a multi-disciplinary software development team. The study showed that the activity-centric workspace supports different individual and collaborative work configuration practices and that activity-centric collaboration is a two-phase process consisting of an activity sharing and per-activity coordination phase.

Feedback and affordances are two of the most well-known principles in interaction design. Unfortunately, the related and equally important notion of feedforward has not been given as much consideration. Nevertheless, feedforward is a powerful design principle for bridging Norman's Gulf of Execution. We reframe feedforward by disambiguating it from related design principles such as feedback and perceived affordances, and identify new classes of feedforward. In addition, we present a reference framework that provides a means for designers to explore and recognize different opportunities for feedforward.

Activity-Based Computing (ABC) has been proposed as an organisational structure for local desktop management and knowledge work. Knowledge work, however, typically occurs in partially overlapping subgroups and involves the use of multiple devices. We introduce co-Activity Manager, an ABC approach that (i) supports activity sharing for multiple collaborative contexts, (ii) includes collaborative tools into the activity abstraction and (iii) supports multiple devices by seamlessly integrated cloud support for documents and activity storage. Our 14 day field deployment in a multidisciplinary software development team showed that activity sharing is used as a starting point for long-term collaboration while integrated communication tools and cloud support are used extensively during the collaborative activities. The study also showed that activities are used in different ways ranging from project descriptions to to-do lists, thereby confirming that a document-driven activity roaming model seems to be a good match for collaborative knowledge work

Informing caregivers by providing them with contextual medical information can significantly improve the quality of patient care activities. However, information flow in hospitals is still tied to traditional manual or digitised lengthy patient record files that are often not accessible while caregivers are attending to patients. Leveraging the proliferation of pervasive awareness technologies (sensors, actuators and mobile displays), recent studies have explored this information presentation aspect borrowing theories from context-aware computing, i.e., presenting subtle information contextually to support the activity at hand. However, the understanding of the information space (i.e., what information should be presented) is still fairly abstruse, which inhibits the deployment of such real-time activity support systems. To this end, this paper first presents situated glyphs, a graphical entity to encode situation specific information, and then presents our findings from an in-situ qualitative study addressing the information space tailored to such glyphs. Applying technology probes using situated glyphs and different glyph display form factors, the study aimed at uncovering the information space pertained to both primary and secondary medical care. Our analysis has resulted in a large set of information types as well as given us deeper insight on the principles for designing future situated glyphs. We report our findings in this paper that we expect would provide a solid foundation for designing future assistive systems to support patient care activities.

Users often become frustrated when they are unable to understand and control a pervasive computing environment. Previous studies have shown that allowing users to pose why and why not questions about context-aware applications resulted in better understanding and stronger feelings of trust. Although why and why not questions have been used before to aid in debugging and to clarify graphical user interfaces, it is currently not clear how they can be integrated into pervasive computing systems. We explain in detail how we have extended an existing pervasive computing framework with support for why and why not questions. This resulted in PervasiveCrystal, a system for asking and answering why and why not questions in pervasive computing environments.

Designers still often create a specific user interface for every target platform they wish to support, which is time-consuming and error-prone. The need for a multi-platform user interface design approach that designers feel comfortable with increases as people expect their applications and data to go where they go. We present Gummy, a multi-platform graphical user interface builder that can generate an initial design for a new platform by adapting and combining features of existing user interfaces created for the same application. Our approach makes it easy to target new platforms and keep all user interfaces consistent without requiring designers to considerably change their work practice.