In the current media landscape, various ideas and narratives gain traction, influenced by the dynamics of selective exposure and a decline in trust in traditional information sources. This trend holds the potential to cultivate polarisation of perspectives, as individuals actively seek information that resonates with their existing attitudes. Hence, diversifying information that is available online can encourage users to engage with multiple perspectives, especially when provided by a trustworthy source. This paper presents findings from two studies which compared individuals with a higher belief in generic conspiracy theories (Study 1; n = 84) and specific conspiracy beliefs on climate change (Study 2; n = 23) to those with lower conspiracy beliefs (nstudy 1 = 93; nstudy 2 = 35) on perceived trustworthiness and usefulness of the so called ‘balanced news chatbots’. These chatbots present a selection of opposing alternative and mainstream perspectives on topics of societal divide like climate change. We found that participants from both groups responded positively to the balanced news chatbot. Trust and perceived usefulness were identified to be key indicators of a positive attitude towards and high intentions of using such a chatbot, corroborating the acceptance of balanced news chatbots as a potential tool to reduce polarisation and conflict, piercing existing information bubbles. In both studies we also found that participants with higher conspiratorial beliefs responded even more positively to the balanced news chatbot than individuals with lower conspiratorial beliefs. We conclude that balanced chatbots are promising as a trusted source of diversified information for individuals with varying levels of conspiracy beliefs.

Background: Problematic social media use has been linked to reduced well-being and impulse control difficulties. While digital self-control apps show potential for reducing general app usage, they often lack customization, leading to limited effectiveness and increased user resistance. Their impact on problematic social media use remains uncertain. Objective: This study evaluates the effectiveness of the Wellspent app, a customizable mobile intervention app designed to promote self-regulated social media use by targeting user-defined problematic app use and offering tailored behavioral nudges. Methods: In a 3-week randomized controlled trial, 70 iPhone users (mean age 26.2, SD 5.6 years; 47/70, 67% female), regularly using at least 1 social media app, were randomly assigned to an intervention (n=35) or control group (n=35). The intervention group received personalized full-screen reminders with the option to quit or continue social media app use whenever an app session exceeded a self-defined time limit. Participants completed weekly online surveys measuring problematic social media use, problematic smartphone use, self-efficacy, and daily screen time on their most problematic app. Linear mixed models tested intervention effects. Results: While no significant reduction in problematic social media use or increase in self-efficacy was observed, the intervention group showed a significant reduction in daily screen time on their most problematic app by approximately 29 minutes (estimate=-29.35, SE 6.84, 95% CI -42.79 to -15.99; P<.001), and a significant decrease in perceived problematic smartphone use (estimate=-0.46, SE 0.18, 95% CI -0.80 to -0.11; P=.01). Conclusions: The Wellspent app demonstrated short-term efficacy in reducing problematic smartphone use. By allowing users to tailor interventions to their personal goals, the app shows promise as a self-directed tool to support healthier digital habits. Further research should explore long-term effects and feature-specific impacts.

Individuals tend to apply preferences and beliefs as heuristics to effectively sift through the sheer amount of information available online. Such tendencies, however, often result in cognitive biases, which can skew judgment and open doors for manipulation. In this work, we investigate how individual and contextual factors lead to instances of confirmation bias when seeking, evaluating, and recalling polarising information. We conducted a lab study, in which we exposed participants to opinions on controversial issues through a Twitter-like news feed. We found that low-effortful thinking, strong political beliefs, and content conveying a strong issue amplify the occurrences of confirmation bias, leading to skewed information processing and recall. We discuss how the adverse effects of confirmation bias can be mitigated by taking bias-susceptibility into account. Specifically, social media platforms could aim to reduce strong expressions and integrate media literacy-building mechanisms, as low-effortful thinking styles and strong political beliefs render individuals especially susceptible to cognitive biases.

Text summaries, images, and mind maps are well-known methods for priming readers to better engage with content. Previously, these “primes” needed to be hand-crafted, limiting their use. The advent of generative technologies makes the automatic creation of custom primes for any passage a realistic possibility. Here, we evaluate the efficacy of primes generated using AI on reading comprehension, reading speed, and re-engagement during mobile reading, which is notorious for its frequent interruptions. We used a mobile platform to present a reading task with an interruption to 44 readers (21 with English as a first language). We found that AI primes increased reading speed by an average of 7% for all readers in the initial reading task with no loss of comprehension and that visual primes had a significant interruption recovery effect for people whose first language was not English. Across all readers, text primes had both the initial reading speed increase and were overall most preferred.

Where you read matters—so what if you could read literally inside your book? Reading in Virtual Reality (VR) has been shown to support deep immersion and narrative engagement. We argue that Generative Artificial Intelligence (GenAI) can significantly expand what is currently possible in VR reading by dynamically producing story-driven environments from text. In this vision paper, we present a pipeline that combines recent advances in language, audio, and 3D scene generation to allow automatically augmenting fiction reading with environmental backdrops. As the reader progresses, these AI-generated environments transition in real time, acting as cognitive props for visual imagery. We also describe a user evaluation, discuss limitations, and address implications and open questions raised by the proposed approach.

The growing sophistication of Large Language Models allows conversational agents (CAs) to engage users in increasingly personalized and targeted conversations. While users may vary in their receptiveness to CA persuasion, stylistic elements and agent personalities can be adjusted on the fly. Combined with image generation models that create context-specific realistic visuals, CAs have the potential to influence user behavior and decision making. We investigate the effects of linguistic and visual elements used by CAs on user perception and decision making in a charitable donation context with an online experiment (n=344). We find that while CA attitude influenced trust, it did not affect donation behavior. Visual primes played no role in shaping trust, though their absence resulted in higher donations and situational empathy. Perceptions of competence and situational empathy were potential predictors of donation amounts. We discuss the complex interplay of user and CA characteristics and the fine line between benign behavior signaling and manipulation.

Background
Interventions targeting social media use show mixed results in improving well-being outcomes, particularly for persons with problematic forms of smartphone use. This study assesses the effectiveness of an intervention app in enhancing well-being outcomes and the moderating role of persons' perceptions about problematic smartphone use (PSU).

Methods
In a randomized controlled trial, N = 70 participants, allocated to the intervention (n = 35) or control condition (n = 35), completed weekly online surveys at baseline, post-intervention, and follow-up. Participants from the intervention condition received personalized full-screen nudges to reduce their social media app use. This secondary analysis focuses on the repeatedly assessed outcomes well-being, positive affect, negative affect, and perceived stress. Linear mixed models were computed.

Results
No significant time x group effects were found, but intervention condition participants reported higher well-being and lower negative affect and stress levels at follow-up. Only one significant moderation was found, indicating that participants reporting higher PSU levels benefited more from the intervention in reducing stress.

Conclusions
The intervention was partly effective and particularly beneficial in reducing stress among smartphone users with higher PSU, highlighting the need to tailor interventions. Present findings need to be replicated by future research using a larger sample size.

Mental Workload (MWL) is a construct widely used in HCI to assess the cognitive demand users must exert to perform a task. Research in human factors, however, has suggested several issues regarding its definitions, scales, and applications. This paper, first, introduces debates surrounding the MWL concept and its most popular measure, the NASA-TLX. We present a systematic review of CHI papers involving MWL and highlight severe issues in its application. Finally, through a validation experiment, we assess the convergent validity and sensitivity of two MWL instruments—NASA-TLX and MRQ. Our findings reveal disagreements in the definitions of MWL and severe drawbacks in NASA-TLX and its applications. Our validation study also presents evidence for a lack of convergent validity and sensitivity of MWL subjective scales in HCI tasks. Our findings recommend caution when employing NASA-TLX in user studies and highlight the need for an MWL definition that is agreed upon within the HCI community.

Background:
Well-designed conversational agents can improve health care capacity to meet the dynamic and complex needs of people self-managing cardiometabolic diseases (CMD). However, a lack of empirical evidence on conversational agent–enabled intervention design features and their impact on engagement make it challenging to comprehensively evaluate effectiveness. This review synthesizes evidence on conversational agent–enabled intervention design features and how they impact on engagement to inform the development of more engaging conversational agent–enabled interventions that effectively help people with CMD to self-manage their condition.

Objective:
The aim of the study is to synthesize evidence pertaining to conversational agent–enabled intervention design features and their impact on engagement of people self-managing CMD.

Methods:
Searches were conducted in Ovid (MEDLINE), Web of Science, and Scopus databases. Inclusion criteria were primary research studies reporting on conversational agent–enabled interventions that included measures of engagement and included adults with CMD. Data extraction captured perspectives of people with CMD on various design features of conversational agent–enabled interventions.

Results:
Of 1366 studies identified for screening, 20 were included in the review. In total, 18 of these were qualitative or quasi-experimental evaluations of conversational agent–enabled intervention prototypes. Five domains of design features that impact user engagement with conversational agent–enabled interventions emerged: communication style, functionality, accessibility, visual appearance, and personality.

Conclusions:
Across all 5 domains, integrating redundancy and anthropomorphism were identified as effective strategies for improving engagement by increasing user autonomy and investment. Future research should adopt design strategies that are inclusive and adaptive to the diverse needs of users and aligned with the unique considerations relevant to conversational agent–enabled interventions.

Designers often engage with video to gain rich, temporal insights about the context of users, collaboratively analyzing it to gather ideas, challenge assumptions, and foster empathy. To capture the full visual context of users and their situations, designers are adopting 360° video, providing richer, more multi-layered insights. Unfortunately, the spherical nature of 360° video means designers cannot create tangible video artifacts such as storyboards for collaborative analysis. To overcome this limitation, we created Tangi, a web-based tool that converts 360° images into tangible 360° video artifacts, that enable designers to embody and share their insights. Our evaluation with nine experienced designers demonstrates that the artifacts Tangi creates enable tangible interactions found in collaborative workshops and introduce two new capabilities: spatial orientation within 360° environments and linking specific details to the broader 360° context. Since Tangi is an open-source tool, designers can immediately leverage 360° video in collaborative workshops.

We explore the metaphorical "daily memory pill"concept - a brief pictorial lifelog recap aimed at reviving and preserving memories. Leveraging psychological strategies, we explore the potential of such summaries to boost autobiographical memory. We developed an automated lifelogging memory prosthesis and a research protocol (Automated Memory Validation "AMV") for conducting privacy-aware, in-situ evaluations. We conducted a real-world lifelogging experiment for a month (n=11). We also designed a browser "Pixel Memories"for browsing one-week worth of lifelogs. The results suggest that daily timelapse summaries, while not yielding significant memory augmentation effects, also do not lead to memory degradation. Participants' confidence in recalled content remains unaltered, but the study highlights the challenge of users' overestimation of memory accuracy. Our core contributions, the AMV protocol and "Pixel Memories"browser, advance our understanding of memory augmentations and offer a privacy-preserving method for evaluating future ubicomp systems.

Computing systems are increasingly designed to adapt to users' cognitive states and mental models. Yet, cognitive biases affect how humans form such models and, therefore, they can impact their interactions with computers. To better understand this interplay, we conducted a scoping review to chart how Human-Computer Interaction (HCI) researchers study cognitive biases. Our findings show that computing systems not only have the potential to induce and amplify cognitive biases but also can be designed to steer users' behaviour and decision-making by capitalising on biases. We describe how HCI researchers develop algorithms and sensing methods to detect and quantify the effects of cognitive biases and discuss how we can use their understanding to inform system design. In this paper, we outline a research agenda for more theory-grounded research and highlight ethical issues when researching and designing computing systems with cognitive biases in mind as they affect real-world behaviour.

Despite its well-documented benefits, reading literature in a second language remains challenging. Graphic novels, by integrating text and images, effectively support English as a Second language (ESL) readers; however, adapting literary works into this format is resource-intensive and lacks scalability.
To address this, we developed a LangChain-based pipeline that automatically transforms a story into a graphic novel. Through a user study with 76 participants, we investigated (1) how this adaptation influences ESL readers' comprehension and narrative engagement, and (2) readers' perception of AI's role in the creative process. Results showed no significant differences in comprehension or engagement between the AI-generated graphic novel and traditional text. Although 70\% of participants recognized AI involvement, attitudes toward its role as illustrator were generally positive, despite a few cross-domain concerns. This work contributes to the understanding of AI-powered storytelling from a human-centered perspective, identifying key insights for effectively supporting readers through AI-generated visual narratives.

Active and healthy ageing depends on maintaining physical and cognitive activity, but it is still challenging to motivate older adults to participate in regular training. This paper describes the iterative design and evaluation of a digital platform for increasing older adults' motivation to perform physical and cognitive exercises. The digital solution was designed and evaluated in four iterations with a total of 13 older adults. The first stage focused on identifying effective communication methods, including different formats of instructional delivery and feedback, as well as tone. The second stage explored the combination of physical activity with cognitively stimulating activities, such as brain games, sport, and hobbies, to find the most motivating combinations. The final stage developed the prototype further by integrating motivational elements into one coherent design, emphasizing clarity, guidance, and user agency. The final evaluation reviewed the overall design, including the importance of adaptive systems that dynamically adjust the difficulty level to align with users' physical and cognitive abilities to increase motivation. This study contributes to the growing field of participatory design within digital health interventions, aligning with best practices that emphasize the need for dynamic user involvement in all stages of development.

In the recent years, studies on health consequences of smartphone usage time have increased, yet findings on the effectiveness of usage interventions remain unclear. This preregistered study investigates the effectiveness of a planning intervention to reduce total smartphone usage time. Additionally, it examines the interventions’ underlying mechanisms of self-efficacy, intention, action, and coping planning. A primary analysis of a randomized controlled trial, with data collected at three measurement points was conducted. Three cohorts of university students were recruited during the period prior to the end-of-term exams. A total of N = 787 participants were allocated to either an intervention condition (n = 389) or a control condition (n = 398). At baseline measurement (T1) the intervention condition formed up to three actions and three coping plans. Self-reported self-efficacy, intention, action, and coping planning as well as objectively measured smartphone usage were assessed up to a three-weeks follow-up. The effectiveness of the intervention and the mediating mechanisms were evaluated using linear mixed models. The analysis revealed no significant effect on total smartphone usage time. With respect to the interventions underlying mechanisms, results show a significant indirect effect of self-efficacy at T2, on a reduction in total smartphone usage time at T3 but no evidence for intention, action, or coping planning.

In today’s workplaces, digital distractions and frequent task-switching significantly impact productivity and well-being. With the integration of AI tools and Large Language Models (LLMs), we face new challenges in managing attention between human cognition and AI-assisted workflows. This workshop will explore the dual nature of AI tools, examining their potential to enhance productivity and mitigate the risks of "metacognitive laziness" and digital distraction. Through collaborative design activities and discussions, participants will examine strategies for developing attention management that support mindful task-switching between human and AI-driven work. We will focus on designing workplaces that leverage AI capabilities while preserving deep work and cognitive well-being. The workshop aims to generate practical insights for creating sustainable work practices in an AI-augmented workplace.

Designers can immerse themselves into the world of users by using 360° video leading to richer insights and better solutions. However, 360° video is challenging to share and incompatible with existing tools, preventing designers from effectively integrating it into their iterative and collaborative workflows. To address these challenges, we developed D360, a tool that enables designers to view, annotate, and collaboratively analyze 360° video. D360 features a web-based 360° video viewing and annotation tool, a database, and Miro integration to analyze 360° video using a familiar collaborative process. We evaluated D360 using walk-throughs with six professional designers that verified its utility and identified improvements to creating and presenting annotations. By providing both design directions for future 360° video tools for designers and our open source tool, we enable practitioners and researchers to leverage the rich interaction and visual context of 360° video for more impactful insights.

Novel consumer neurotechnologies allow users to track their cognitive states and processes, such as attention and mental workload (MWL). However, data on these inherently complex, abstract, and invisible cognitive processes can be challenging to interpret, and little is known about how users make sense of their data. In this work, we explore how people understand and reflect on MWL through six semi-structured interviews and a follow-up experience sampling study. We examine how people conceptualize MWL, distinguish it from related concepts such as stress, what they consider high and low workload in their daily lives, and how they connect workload to emotional states. We discuss these user perceptions and identify barriers to MWL self-tracking, such as lack of trust in the data and ambiguity of the MWL concept, and propose five design guidelines to make cognitive tracking tools more intelligible and meaningful for users.

Background:
Heart failure (HF) is a chronic, progressive condition where the heart cannot pump enough blood to meet the body’s needs. In addition to the daily challenges that HF poses, acute exacerbations can lead to costly hospitalizations and increased mortality. High health care costs and the burden of HF have led to the emerging application of new technologies to support people living with HF to stay well while living in the community. However, many digital solutions have not involved consumers and health care professionals in their design, leading to poor adoption. The SmartHeart project aimed to codevelop a smart health ecosystem to support the early detection of HF deterioration and encourage self-care, potentially preventing hospitalizations.

Objective:
This study aims to provide an overview of the SmartHeart project by describing our approach to designing the SmartHeart system, outlining its features, and describing the planned pilot study to determine the feasibility of the system.

Methods:
We used the Integrate, Design, Assess, and Share (IDEAS) framework to guide the development of the SmartHeart system, involving users (people with HF and their caregivers) and stakeholders (health care providers involved in the management of HF) in its design. SmartHeart is a complete remote heart health monitoring and automated feedback delivery system. It includes 2 user interfaces for patients: an Amazon Alexa conversational agent and a smartphone app. The system collects physiological, symptom, and behavioral data through wireless sensors and self-reports from users. These data are processed and analyzed to provide personalized health insights, self-care support, and alerts in case of health deterioration. The system also includes a web-based user interface for health care professionals, allowing them to access data, send messages to users, and receive notifications about potential health deterioration. A single-arm, multicenter pilot trial (N=20) is planned to determine the feasibility and acceptability of SmartHeart before evaluation through a randomized controlled trial. The primary outcome will be a description of the study's feasibility (recruitment, attrition, engagement, and changes in self-care).

Results:
The SmartHeart study started in January 2021 on procurement of funding. Recruitment for the pilot trial started in August 2024 and will be completed by March 2025. We have currently enrolled 12 participants. Follow-up of all participants will be completed by the end of May 2025.

Conclusions:
We have co-designed and developed a complete remote heart health monitoring and automated feedback delivery system for the early detection of HF deterioration and prevention of HF-related hospitalizations. The next step is a pilot study, which will provide valuable information on feasibility and preliminary effects to inform a larger evaluation trial. SmartHeart has the potential to augment existing health services and help people with HF stay well while living in the community.

Rapid Serial Visual Presentation (RSVP) has been both praised and scorned as an effective reading technique. While it enables text rendering on small screens, its dynamic text display tends to demand heightened attention from users. Previous work has implemented RSVP reading interfaces on a range of devices, such as watches, glasses, and phones, but none have gained widespread adoption. Considering its ability to display information on devices with small screen real-estate effectively, we explored its applicability to reading smartphone notifications. Following a user-centered design process, we first elicited the main design parameters through a focus group. Informed by these insights, we implemented our RSVP notification prototype, SpeedNotification. We tested SpeedNotification's specific design parameters in a lab study and evaluated RSVP notifications more broadly in a field study. We present the essential functionalities and challenges for creating a feasible RSVP user experience for mobile notifications and discuss associated issues and opportunities.