Framing 'Collaboration': How Human-Human Principles Translate into Human-AI RealitiesThe recent advancement of AI has shifted terminology: humans "use" computers but "collaborate with" AI. This anthropomorphic framing shapes expectations of system capabilities. Despite the large body of research adopting "human-AI collaboration" as a term, there seems to be little consensus on a definition of the concept at a glance. To address this potential gap and to provide a comprehensive overview of existing related literature, we first conducted a thematic analysis on human-human collaboration literature (n=60) to extract definitional components and associated concepts. Second, we analyzed publications on human-AI collaboration (n=299) using OpenAI’s GPT4o mini and o3 mini models, mapping the identified concepts to the AI context to examine the extent to which these concepts of collaboration are represented there. Our findings provide a shared conceptual foundation to support interdisciplinary research and suggest future research directions. Additionally, they inform the design of human-AI interfaces and interaction processes, bridging theory and practice.2026KBKarin Breckner et al.University of Applied Sciences Upper AustriaHuman-LLM CollaborationAI-Assisted Decision-Making & AutomationParticipatory DesignCHI
The Socio-Technical Energy Triad: Uncovering Tensions between Service Providers, Users, and the Energy SystemThe digital transformation of the energy sector introduces HCI challenges that exceed traditional end-user design. While systemic perspectives are increasingly called for, empirical work on the tensions between provider constraints and domestic practices remains limited. We introduce the Socio-Technical Energy Triad, situating digital energy services at the intersection of providers, households, and the emerging energy system. Drawing on 21 provider interviews, a technology-probe deployment with 10 users, and a survey with 98 respondents, we identify a structural disconnect across the Triad. Providers struggle with fragmented data infrastructures and limited user trust, hindering data-driven optimization. For households, this system logic remains largely invisible: they focus on efficiency and resist additional cognitive labor in managing domestic energy. We argue that bridging this "Invisibility Gap" requires not just traditional visualization but the design of legible automation and the translation of system-level needs into clear, user-centered signals.2026FGFlorian Güldenpfennig et al.University of ViennaEnergy Conservation Behavior & InterfacesPrivacy by Design & User ControlExplainable AI (XAI)CHI
Cracking the Case Together: Role Perceptions in Human-AI Mystery Solving DialoguesLarge Language Models (LLMs) aim to mimic a natural form of human conversation, likely contributing to an anthropomorphic perception of AI in contrast to conventional human-computer interfaces. Our study explores human-AI conversations and humans’ perception of their counterpart in a collaborative mystery solving task with Anthropic’s Claude 3.5 Sonnet v2 model. We collected self-report data on participants’ perception of the interaction, measured task performance, and analyzed conversational dynamics using LLM-based emotion coding. We found that humans’ perception of AI, ranging from that of a teammate or colleague to a tool, did not necessarily impact performance in mystery solving, but correlated with aspects of the interaction itself. When participants perceived the AI as a teammate or colleague, they felt a stronger sense of team cohesion and their conversations were more collaborative, with more positive emotions. These findings may help practitioners design human-AI interfaces that foster positive interactions without endangering performance.2026KBKarin Breckner et al.University of Applied Sciences Upper AustriaHuman-LLM CollaborationEmpathy & Emotional DesignAffective Human-Computer DialogueCHI
Breaking Rotational Symmetry: Minimal Landmarks Stabilize Orientation in Screen-Based 3D GamesDisorientation in first-person, screen-based 3D games breaks flow, yet most evidence comes from head-mounted display Virtual Reality or abstract tasks. We ask what minimal in-world information stabilizes orientation in visually repetitive worlds. In a within-participant mixed-methods study (N=20), a symmetrical game environment orthogonally varied landmark geometry and featural polarity, plus a no-landmark baseline. In-situ pointing and orientation ratings during learning and subsequent relocation, together with trajectory analytics and interviews, show that coupling geometric structure with strong featural polarity is decisive: relative to baseline, localization error drops (approx. 79%) and pointing accuracy rises (approx. 64%), whereas geometry alone helps less and smooth, featureless forms least. Players anchor on discontinuities (edges/corners; black--white boundary) and triangulate with distances and angles; without an anchor they revert to ground/light patterns and report symmetry-induced confusion. We establish a minimal-cue testbed for desktop play and derive actionable guidance: break rotational symmetry in a persistent landmark to enable legible, low head-up display environments.2026CFChristian Feichtinger et al.Fachhochschule HagenbergGame UX & Player BehaviorGamification DesignCHI
Peer or Steer: A Pilot Study Exploring Human-AI Collaboration in Creative FieldsRecent years have brought immense advancements around development of Artificial Intelligence (AI) technology and its application. This has also led to a boost in interest on human-AI collaboration and co-creation. Specifically in the creative field, where it is usually not sufficient for an AI to solve given problems or produce deterministic output, this next-level interaction between humans and AI comes with huge potential but also major challenges, related to aspects such as role and power distribution, trust and reliance, or efficiency and effectiveness. In this paper, we present a pilot study on human-AI collaboration in three different creative fields (programming, marketing texting, and UI design), addressing User Experience, Technology Acceptance and, specifically, Perception of Collaboration. The study is based on a theoretical framework we derived from prior research through a focused, systematic literature review, and intended to raise research questions and identify related hypotheses informing future empirical work.2025DLDavid Lang et al.Human-AI (and Robot!) CollaborationCSCW
Pathways of Desire: Enhancing Navigation and Sense of Community Through Player-Generated Desire PathsNavigating is essential in many video games. However, previous work suggests that many games still suffer from navigational problems that decrease enjoyment. In this paper, we focus on "Desire Paths", informal trails collectively created by pedestrians representing the most convenient route. While they are known to be useful wayfinding aids, it is unclear how they affect navigation and experience in games. We therefore investigated diegetically visualized player trajectory data in a 2D game through virtual footprints that were persistently visible for all subsequent players. Through a mixed-methods study involving 50 participants, we found that virtual footprints improved navigation by guiding players to points of interest and reducing disorientation for early players. However, visual clutter from excessive footprints reduced their effectiveness in later stages. They also fostered a sense of community, especially for late-stage players and prompted exploration of yet undiscovered areas. We further discuss design implications and future research directions.2025MLMichael Lankes et al.University of Applied Sciences Upper Austria, Department of Digital MediaGamification DesignMultiplayer & Social GamesCHI
Investigating the Impact of Customized Avatars and the Proteus Effect during Physical Exercise in Virtual RealityVirtual reality (VR) allows to embody avatars. Coined the Proteus effect, an avatar's visual appearance can influence users' behavior and perception. Recent work suggests that athletic avatars decrease perceptual and physiological responses during VR exercise. However, such effects can fail to occur when users do not experience avatar ownership and identification. While customized avatars increase body ownership and identification, it is unclear whether they improve the Proteus effect. We conducted a study with 24 participants to determine the effects of athletic and non-athletic avatars that were either customized or randomly assigned. We developed a customization editor to allow creating customized avatars. We found that customized avatars reduced perceived exertion. We also found that athletic avatars decreased heart rate while holding weights, however, only when being customized. Results indicate that customized avatars can positively influence users during physical exertion. We discuss the utilization of avatar customization in VR exercise systems.2025MKMartin Kocur et al.University of Applied Sciences Upper AustriaIdentity & Avatars in XRFitness Tracking & Physical Activity MonitoringCHI
Hand Grips and Mobile Menus: Exploring Perceived Usability and User Preferences This paper investigates the relationship between menu design and hand positions in relation to the assessment of end users with main focus on usability, user preference, and potential adaptions to different hand positions. Sixteen (N = 16) participants first participated in a co-design workshop, in which they proposed menu designs for different hand grips. Based on the design proposals, a selection of menu designs were derived and implemented in a mobile app prototype, on which a menu selection study was conducted to investigate performance and perceived usability of the menus in one-handed and two-handed interaction. The results include user ratings and performance, which highlight the need for mobile menus to be adapted for different hand positions. Based on that, we derive design recommendations for more adaptive, user-centric and ergonomic mobile menu designs to match the natural interactions of users.2024TZTamara Zieher et al.Hand Gesture RecognitionPrototyping & User TestingMobileHCI
Changing Lanes Toward Open Science: Openness and Transparency in Automotive User ResearchWe review the state of open science and the perspectives on open data sharing within the automotive user research community. Openness and transparency are critical not only for judging the quality of empirical research, but also for accelerating scientific progress and promoting an inclusive scientific community. However, there is little documentation of these aspects within the automotive user research community. To address this, we report two studies that identify (1) community perspectives on motivators and barriers to data sharing, and (2) how openness and transparency have changed in papers published at AutomotiveUI over the past 5 years. We show that while open science is valued by the community and openness and transparency have improved, overall compliance is low. The most common barriers are legal constraints and confidentiality concerns. Although research published at AutomotiveUI relies more on quantitative methods than research published at CHI, openness and transparency are not as well established. Based on our findings, we provide suggestions for improving openness and transparency, arguing that the motivators for open science must outweigh the barriers. All supporting materials are freely available at: https://osf.io/zdpek/2024PEPatrick Ebel et al.Research Ethics & Open ScienceAutoUI
Development and Evaluation of Advanced Cyclist Assistance Systems on a Bicycle SimulatorResearch on cycling safety has recently gained the attention of the HCI community. While there have been multiple proposals for automated driving features on bikes, we are unaware of a project that systematically aims to translate and evaluate driver assistance systems from the automotive to the bike domain to promote cycling safety in traffic. Thus, we implemented an adaptive cruise control and a lane-keeping/centering system with hard- and software on a motion-based bicycle simulator and investigated their potential in a virtual reality experiment. Based on performance measurements and subjective ratings, results showed significant improvements in technology acceptance, subjective workload, and driving performance regarding the cruise control. In contrast, the lane-centering and lane-keeping features were rated significantly worse than the baseline without such assistance. The paper concludes with a critical reflection on automated driving features for bicycles and a list of recommendations for future projects in this field.2024YWYu Wang et al.External HMI (eHMI) — Communication with Pedestrians & CyclistsHead-Up Display (HUD) & Advanced Driver Assistance Systems (ADAS)Micromobility (E-bike, E-scooter) InteractionAutoUI
What Characterizes "Situations'' in Situation Awareness? Findings from a Human-centered InvestigationSituation Awareness (SA) is one of the core concepts describing drivers' interaction with vehicles, and the lack of SA has contributed to multiple incidents with automated systems. Despite existing definitions and measurements, little is known about what constitutes the concept of situations from users’ perspective, i.e., do they have a similar or different understanding of situation dynamics? Therefore, we conducted a video-based experiment where participants had to mark the onset of new situations from their perspective, provide a continuous criticality rating, and justify their decisions in a post-test interview. Our results indicate that the understanding of situations, their complexity, and their duration is quite diverse between people and independent of properties such as age, gender, or driving experience, while partly being influenced by the road type. Additionally, we found correlations between subjective situation durations, criticality ratings, and algorithm output, which can be exploited by future applications and experiments.2024PAPhilipp Michael Markus Peter Asteriou et al.Head-Up Display (HUD) & Advanced Driver Assistance Systems (ADAS)Eye Tracking & Gaze InteractionHuman Pose & Activity RecognitionAutoUI
Only Trust a Hidden Wizard: Investigating the Effects of Wizard Visibility in Automotive Wizard of Oz StudiesThe Wizard-of-Oz method has been widely used recently as it allows mimicking automated vehicles with relatively few resources. In some studies, it is challenging to ensure that the wizard remains fully hidden from participants, despite this being a crucial aspect of such experiments. To determine whether participants' awareness of the wizard influences the outcomes of these studies, we conducted an experiment investigating participants' crossing behavior and subjective perception of a remote-controlled automated vehicle. Participants were exposed to two conditions: in one, they solely focused on a simulated vehicle driving autonomously; in the other, they observed a wizard with a remote control and were instructed to imagine the car was automated. Results, based on scales for user experience, acceptance, and trust, as well as crossing behavior, indicate similar results. However, participants’ knowledge of the wizard necessitates careful interpretation when system errors are simulated. We conclude with recommendations for future Wizard-of-Oz experiments.2024HKHeike Christiane Kotsios et al.Teleoperated DrivingAutoUI
Voice Assistants' Accountability through Explanatory Dialogues As voice assistants (VAs) become more advanced leveraging Large Language Models (LLMs) and natural language processing, their potential for accountable behavior expands. Yet, the long-term situational effectiveness of VAs’ accounts when errors occur remains unclear. In our 19-month exploratory study with 19 households, we investigated the impact of an Alexa feature that allows users to inquire about the reasons behind its actions. Our findings indicate that Alexa’s accounts are often single, decontextualized responses that led to users’ alternative repair strategies over the long term, such as turning off the device, rather than initiating a dialogue about what went wrong. Through role-playing workshops, we demonstrate that VA interactions should facilitate explanatory dialogues as dynamic exchanges that consider a range of speech acts, recognizing users’ emotional states and the context of interaction. We conclude by discussing the implications of our findings for the design of accountable VAs.2024FAFatemeh Alizadeh et al.Intelligent Voice Assistants (Alexa, Siri, etc.)Multilingual & Cross-Cultural Voice InteractionExplainable AI (XAI)CUI
Threads of Traceability: Textile IDs in the Fabric of Sustainable FashionTextile fabrication, an ancient human technology, has evolved over millennia, transitioning from a focus on affordability and speed to a current emphasis on sustainability. With Textile ID, we envision a digital garment passport that seamlessly incorporates directly into textile surfaces as a design element to bridge the gap between sustainability and consumer engagement, transforming garments into interactive storytellers of their ecological journey. The visual surface of the garment can be scanned with a smartphone to access a unique identifier embedded within the fabric, which provides essential information about the product's lifecycle. This work discusses the design space of various visual and textile parameters, proposes design possibilities and insights for implementation. Finally, we showcase a set of sample garment designs and provide design recommendations for designers to use in their future work.2024MHMira Alida Haberfellner et al.Customizable & Personalized ObjectsEcological Design & Green ComputingDIS
Putting Things into Context: Generative AI-Enabled Context Personalization for Vocabulary Learning Improves Learning MotivationFostering students' interests in learning is considered to have many positive downstream effects. Large language models have opened up new horizons for generating content tuned to one's interests, yet it is unclear in what ways and to what extent this customization could have positive effects on learning. To explore this novel dimension, we conducted a between-subjects online study (n=272) featuring different variations of a generative AI vocabulary learning app that enables users to personalize their learning examples. Participants were randomly assigned to control (sentence sourced from pre-existing text) or experimental conditions (generated sentence or short story based on users’ text input). While we did not observe a difference in learning performance between the conditions, the analysis revealed that generative AI-driven context personalization positively affected learning motivation. We discuss how these results relate to previous findings and underscore their significance for the emerging field of using generative AI for personalized learning.2024JLJoanne Leong et al.MITGenerative AI (Text, Image, Music, Video)Human-LLM CollaborationOnline Learning & MOOC PlatformsCHI
Supporting Task Switching with Reinforcement LearningAttention management systems aim to mitigate the negative effects of multitasking. However, sophisticated real-time attention management is yet to be developed. We present a novel concept for attention management with reinforcement learning that automatically switches tasks. The system was trained with a user model based on principles of computational rationality. Due to this user model, the system derives a policy that schedules task switches by considering human constraints such as visual limitations and reaction times. We evaluated its capabilities in a challenging dual-task balancing game. Our results confirm our main hypothesis that an attention management system based on reinforcement learning can significantly improve human performance, compared to humans’ self-determined interruption strategy. The system raised the frequency and difficulty of task switches compared to the users while still yielding a lower subjective workload. We conclude by arguing that the concept can be applied to a great variety of multitasking settings.2024ALAlexander Lingler et al.University of Applied Sciences Upper AustriaPrivacy by Design & User ControlNotification & Interruption ManagementCHI
Loopsense: low-scale, unobtrusive, and minimally invasive knitted force sensors for multi-modal input, enabled by selective loop-meshingIntegrating sensors into knitted input devices traditionally comes with considerable constraints for textile and UI design freedom. In this work, we demonstrate a novel, minimally invasive method for fabricating knitted sensors that overcomes this limitation. We integrate copper wire with piezoresistive enamel directly into the fabric using weft knitting to establish strain and pressure sensing cells that consist only of single pairs of intermeshed loops. The result is unobtrusive and potentially invisible, which provides tremendous latitude for visual and haptic design. Furthermore, we present several variations of stitch compositions, resulting in loop meshes that feature distinct response with respect to direction of exerting force. Utilizing this property, we are able to infer actuation modalities and considerably expand the device's input space. In particular, we discern strain directions and surface pressure. Moreover, we provide an in-depth description of our fabrication method, and demonstrate our solution's versatility on three exemplary use cases.2024RARoland Aigner et al.University of Applied Sciences Upper AustriaShape-Changing Interfaces & Soft Robotic MaterialsOn-Skin Display & On-Skin InputCircuit Making & Hardware PrototypingCHI
From Real to Virtual: Exploring Replica-Enhanced Environment Transitions along the Reality-Virtuality ContinuumRecent Head-Mounted Displays enable users to perceive the real environment using a video-based see-through mode and the fully virtual environment within a single display. Leveraging these advancements, we present a generic concept to seamlessly transition between the real and virtual environment, with the goal of supporting users in engaging with and disengaging from any real environment into Virtual Reality. This transition process uses a digital replica of the real environment and incorporates various stages of Milgram’s Reality-Virtuality Continuum, along with visual transitions that facilitate gradual navigation between them. We implemented the overall transition concept and four object-based transition techniques. The overall transition concept and four techniques were evaluated in a qualitative user study, focusing on user experience, the use of the replica and visual coherence. The results of the user study show, that most participants stated that the replica facilitates the cognitive processing of the transition and supports spatial orientation.2024FPFabian Pointecker et al.University of Applied Sciences Upper AustriaMixed Reality WorkspacesImmersion & Presence ResearchCHI
Spot'Em: Interactive Data Labeling as a Means to Maintain Situation AwarenessAppropriate monitoring and successfully intervening when automation fails is one of the most critical issues in level 2 automated driving, since drivers suffer from low situation awareness when using such systems. To counter, we present a gamified in-vehicle interface based on ideas from previous work, where drivers have to support the vehicle by pointing at other traffic objects in the environment. We hypothesized that this system could help drivers in the monitoring task, maintain their situation awareness, and result in lower crash rates. We implemented a prototype of this system and evaluated it in a lab study with N=20 participants. The results indicate that participants were looking more intensively at lead vehicles and performed stronger braking actions. However, there was no measurable benefit on situation awareness and intervention performance in critical situations. We conclude by discussing differences to related experiments and present future ideas.2023PWPhilipp Wintersberger et al.Automated Driving Interface & Takeover DesignGamification DesignAutoUI
A Real Bottleneck Scenario with a Wizard of Oz Automated Vehicle - Role of eHMIsAutomated vehicles (AVs) are expected to encounter various ambiguous space-sharing conflicts in urban traffic. Bottleneck scenarios, where one of the parts needs to resolve the conflict by yielding priority to the other, could be utilized as a representative ambiguous scenario to understand human behavior in experimental settings. We conducted a controlled field experiment with a Wizard of Oz automated car in a bottleneck scenario. 24 participants attended the study by driving their own cars. They made yielding, or priority-taking decisions based on implicit and explicit locomotion cues on AV realized with an external display. Results indicate that acceleration and deceleration cues affected participants' driving choices and their perception regarding the social behavior of AV, which further serve as ecological validation of related simulation studies.2023HİHatice Şahin İppoliti et al.External HMI (eHMI) — Communication with Pedestrians & CyclistsAutoUI