Difficulty spillover and suboptimal help-seeking challenge the sequential, knowledge-intensive nature of digital tasks. In online surveys, tough questions can drain mental energy and hurt performance on later questions, while users often fail to recognize when they need assistance or may satisfy, lacking motivation to seek help. We developed a proactive, adaptive system using electrodermal activity and mouse movement to predict when respondents need support. Personalized classifiers with a rule-based threshold adaptation trigger timely LLM-based clarifications and explanations. In a within-subjects study (N=32), aligned-adaptive timing was compared to misaligned-adaptive and random-adaptive controls. Aligned-adaptive assistance improved response accuracy by 21%, reduced false negative rates from 50.9% to 22.9%, and improved perceived efficiency, dependability, and benevolence. Properly timed interventions prevent cascades of degraded responses, showing that aligning support with cognitive states improves both the outcomes and the user experience. This enables more effective, personalized large language model (LLM) support in survey-based research.

In light of inherent trade-offs regarding fairness, privacy, interpretability and performance, as well as normative questions, the machine learning (ML) pipeline needs to be made accessible for public input, critical reflection and engagement of diverse stakeholders.In this work, we introduce a participatory approach to gather input from the general public on the design of an ML pipeline. We show how people’s input can be used to navigate and constrain the multiverse of decisions during both model development and evaluation. We highlight that central design decisions should be democratized rather than “optimized” to acknowledge their critical impact on the system’s output downstream. We describe the iterative development of our approach and its exemplary implementation on a citizen science platform. Our results demonstrate how public participation can inform critical design decisions along the model-building pipeline and combat widespread lazy data practices.

Large language models (LLMs) like ChatGPT have been widely adopted in work contexts. We explore the impact of ChatGPT on young professionals’ perception of productivity and sense of accomplishment. We collected LLMs’ main use cases in knowledge work through a preliminary study, which served as the basis for a two-week diary study with 21 young professionals reflecting on their ChatGPT use. Findings indicate that ChatGPT enhanced some participants’ perceptions of productivity and accomplishment by enabling greater creative output and satisfaction from efficient tool utilization. Others experienced decreased perceived productivity and accomplishment, driven by a diminished sense of ownership, perceived lack of challenge, and mediocre results. We found that the suitability of task delegation to ChatGPT varies strongly depending on the task nature. It’s especially suitable for comprehending broad subject domains, generating creative solutions, and uncovering new information. It’s less suitable for research tasks due to hallucinations, which necessitate extensive validation.

Captions are a valuable scaffold for language learners, aiding comprehension and vocabulary acquisition.Past work has proposed enhancements such as keyword highlights for increased learning gains.However, little is known about learners' experience with enhanced captions, although this is critical for adoption in everyday life.We conducted a survey and focus group to elicit learner preferences and requirements and implemented a processing pipeline CCS Concepts• Applied

Human-AI interaction in text production increases complexity in authorship. In two empirical studies (n1 = 30 & n2 = 96), we investigate authorship and ownership in human-AI collaboration for personalized language generation. We show an AI Ghostwriter Effect : Users do not consider themselves the owners and authors of AI-generated text but refrain from publicly declaring AI authorship. Personalization of AI-generated texts did not impact the AI Ghostwriter Effect , and higher levels of participants’ influence on texts increased their sense of ownership. Participants were more likely to attribute ownership to supposedly human ghostwriters than AI ghostwriters, resulting in a higher ownership-authorship discrepancy for human ghostwriters. Rationalizations for authorship in AI ghostwriters and human ghostwriters were similar. We discuss how our findings relate to psychological ownership and human-AI interaction to lay the foundations for adapting authorship frameworks and user interfaces in AI in text-generation tasks.

Artificial intelligence has enabled scalable auto-creation of context-aware personalized learning materials. However, it remains unclear how content personalization shapes the learners’ experience. We developed one personalized and two non-personalized, crowdsourced versions of a mobile language learning app: (1) with personalized auto-generated photo flashcards, (2) the same flashcards provided through crowdsourcing, and (3) manually generated flashcards based on the same photos. A two-week in-situ study (n = 64) showed that learners assessed the quality of the non-personalized auto-generated material to be on par with manually generated material, which means that auto-generation is viable. However, when the auto-generation was personalized, the learners’ quality rating was significantly lower. Further analyses suggest that aspects such as prior expectations and required efforts must be addressed before learners can actually benefit from context-aware personalization with auto-generated material. We discuss design implications and provide an outlook on the role of content personalization in AI-supported learning.

Research on Extended Reality (XR) and Artificial Intelligence (AI) is booming, which has led to an emerging body of literature in their intersection. However, the main topics in this intersection are unclear, as are the benefits of combining XR and AI. This paper presents a scoping review that highlights how XR is applied in AI research and vice versa. We screened 2619 publications from 203 international venues published between 2017 and 2021, followed by an in-depth review of 311 papers. Based on our review, we identify five main topics at the intersection of XR and AI, showing how research at the intersection can benefit each other. Furthermore, we present a list of commonly used datasets, software, libraries, and models to help researchers interested in this intersection. Finally, we present 13 research opportunities and recommendations for future work in XR and AI research.

Attention management systems seek to minimize disruption by intelligently timing interruptions and helping users navigate multiple tasks and activities. While there is a solid theoretical basis and rich history in HCI research for attention management, little progress has been made regarding their practical implementation and deployment. Building sophisticated attention management systems requires a great variety of sensors, task- and user models, and multiple devices while considering the complexity of user context and human behavior. Novel AI technologies, such as generative systems, reinforcement learning, and large language models, open new possibilities to create intelligent, practical, and user-centered attention management systems. This proposed workshop aims to bring together researchers and practitioners from diverse backgrounds to discuss and formulate a research agenda to advance attention management systems using novel AI tools to manage and mitigate interruptions from computing systems effectively.

COVID-19 caused an abrupt switch from face-to-face to online teaching. This led to unknown challenges and consequences for students and lecturers. In the first semester after its outbreak, we developed a messenger-based chatbot to perform an experience sampling study to evaluate students’ well-being and experiences (n = 31) with the radical changes in higher education. Finding a decrease in students’ perceived motivation but an increase in productivity, we conducted a follow-up survey to compare the development a year later (n = 41). Our results revealed two main student profiles, one feeling severely impacted by the persisting social distance in their study performance and the other appreciating the flexibility and expended free time due to the changes in the teaching formats. Based on our findings, we introduce implications for the overall design of higher education and show the benefits and challenges of combining chatbot-enabled experience sampling with traditional surveys.

The ubiquity of mobile devices has fueled the popularity of microlearning, namely informal self-directed learning during brief personal downtime. However, learner engagement is challenging to maintain, and microlearning habits are hard to establish. Scheduled reminders are ineffective as they do not match the users’ variable schedules and their intention or capacity to engage. In this paper, we propose a schedule-based and an activity-based trigger for microlearning. The first trigger is sensitive to the learners’ agenda and device status and includes a snooze mechanism. A four-week study (n=10) showed slightly lower response times when compared to triggers scheduled at a fixed time but did not improve learner engagement. The second trigger initiates audio-based microlearning when plugging in headphones. Thus, we minimize the access to personal data and capture a moment where learners engage with their device for a listening activity. In an exploratory user study (n=10), the plugin trigger achieved high compliance rates and was less likely to induce annoyance in users than lock screen notifications. We conclude that intelligent reminders with simple interaction options can contribute to learner engagement.

Books can be a valuable resource for language learners, providing entertainment and showcasing authentic language usage. To further support learners, e-book platforms already include interactive vocabulary aids. Similarly, research has started to investigate the feasibility and usability of grammar aids. However, grammar aids can hinder reading flow. In this paper, we design an interactive e-reading interface with different levels of grammar support. We perform a within-subject user study (n = 24) where we assess the relationship between the reading flow, usability, and usefulness with these designs in a fiction reading scenario. Our findings show that more detailed designs are considered more useful, but also more disruptive and less usable than simpler designs. Hence, grammar support interfaces need to balance the focus on learning and the reading experience.

Augmented Reality (AR) provides a unique opportunity to situate learning content in one's environment. In this work, we investigated how AR could be developed to provide an interactive context-based language learning experience. Specifically, we developed a novel handheld-AR app for learning case grammar by dynamically creating quizzes, based on real-life objects in the learner's surroundings. We compared this to the experience of learning with a non-contextual app that presented the same quizzes with static photographic images. Participants found AR suitable for use in their everyday lives and enjoyed the interactive experience of exploring grammatical relationships in their surroundings. Nonetheless, Bayesian tests provide substantial evidence that the interactive and context-embedded AR app did not improve case grammar skills, vocabulary retention, and usability over the experience with equivalent static images. Based on this, we propose how language learning apps could be designed to combine the benefits of contextual AR and traditional approaches.

Distractions and interruptions often disrupt mobile learners. Luckily, task resumption (memory) cues can support users in resuming a learning task. These cues can have multiple forms and designs, but their effectiveness depends heavily on their adaptation to the specific learning use case. This work explores the causes of interruptions during mobile learning and outlines designs for task resumption support. We report findings from two focus groups with HCI experts (N = 4) and users of mobile learning applications (N = 3). Finally, we discuss these findings by drawing on literature, and we derive a research agenda of currently unexplored concepts. We state limitations and open questions in the domain of task resumption support for mobile learning.

Reading foreign language texts is a frequently used strategy for language learning. Visual text augmentation methods further support the learning experience, e.g., by annotating vocabulary or grammar. Common approaches are integrated dictionaries or static grammar highlights. This work investigates how we can further support grammar learning with the dynamic visualization and interaction opportunities offered by digital reading devices. In collaboration with teachers and potential learners, we identify difficulties learners experience with English grammar and gather ideas for suitable interactive text augmentations. Based on this, we design four different concepts that augment adjectives and adverbs in English-language texts using typographic cues and interactive information displays. The concepts are evaluated in a within-subject study (N = 16). Results show that participants preferred concepts that presented case-specific support, did not distract too much from the text, and gave details on demand. We conclude with design recommendations for designing text augmentation for language learning.