AI summaries on social media are reshaping how users form opinions about political topics, yet their influence remains largely unexamined despite their widespread deployment. This paper investigates how two types of AI summaries affect user opinions and engagement: textual summaries of discussion narratives and percentage breakdowns of agreement/disagreement. Through a 144-participant experiment on simulated online discussion threads, we found that displaying commenter agreement percentages amplified social conformity towards the majority views beyond reading comments alone. Conversely, AI narrative summaries created misperceptions of balance in polarised threads, reducing opinion change. While these summaries did not influence participants’ willingness to engage, toxic discussions deterred participation even when participants held majority views. Based on our findings, we provide critical design interventions for industry and researchers to mitigate these tools’ polarising effects, paving the way for responsible AI deployment on social media platforms.

Short-form video platforms have become major channels for misinformation, with their rich multimodal features making false claims highly believable. HCI research shows that providing corrections in the same modality as the misinformation can be an effective solution. However, since corrections and misinformation convey contradicting information, the order in which one is exposed to them can impact what one believes. We conducted a between-subjects mixed-methods experiment where participants (N=120) rated the credibility of misinformation statements before and after viewing misinformation videos paired with correction videos. Corrections were shown either before, during, or after misinformation. Across all three timings, corrections reduced belief in misinformation, but post-exposure corrections proved most effective and mid-exposure corrections least effective. These findings suggest that correction mechanisms should appear after misinformation exposure, while avoiding mid-exposure interruptions that reduce impact. We outline design recommendations for integrating correction videos into short-form video platforms to improve resilience against misinformation.

As multi-agent Large Language Models (LLMs) gain traction, designers must consider how to surface their internal reasoning in ways that foster appropriate trust. We present a design-led, qualitative, comparative structured observation study, exploring how users interpret and evaluate transparency in multi-agent LLMs. Participants interacted with five interface variants, each instantiating different combinations of transparency-related design dimensions, across two task types: information-seeking and logical reasoning. We surface participants’ mental models, the cues they interpret as signals of transparency and trustworthiness, and how they weigh the costs and benefits of increasing process visibility. Transparency needs were dynamic and context-sensitive, with the ideal “Goldilocks” (i.e., “just right” transparency) level shaped jointly by task demands, interface affordances, and user characteristics such as task expertise and dispositional AI trust. We highlight tensions between process visibility, information sufficiency, and cognitive effort, and synthesise these insights into design considerations for aligning transparency with user needs in future multi-agent LLM interfaces.

Social media has blurred the line between professional journalism and personality-driven commentary, yet we know little about how users evaluate credibility and engage with news from influencers and legacy media when they appear in the same feed. This short paper investigates how political ideology and news source type shape perceived credibility and follow intentions on Instagram. We conducted a mixed-methods experiment where U.S.-based participants (N=120) viewed a set of real news posts and rated the credibility of four accounts (two legacy media–based, two influencer-based), balanced by ideology (two left-leaning, two right-leaning), and indicated whether they would follow each account. Our findings suggest that perceived credibility on Instagram is multi-dimensional, rooted in ideological alignment, yet moderated by institutional signals and perceived authenticity. These insights highlight how platform design and source dynamics can reinforce selective exposure, with implications for both mitigating polarisation and strengthening trust in online news ecosystems.

While short video platforms such as TikTok, YouTube Shorts, and Instagram Reels are frequently criticised for facilitating the spread of misinformation, they are also increasingly leveraged as tools for countering it through debunking content. Although video-based corrections have demonstrated effectiveness, their persuasive impact may depend on the richness of their audio-visual elements. This study examines the persuasive efficacy of three fundamental presentation styles commonly used in short-form video content: (1) videos featuring only captions, (2) captions accompanied by relevant images, and (3) captions presented alongside the creator’s visible face. Our results indicate that videos incorporating either relevant and engaging imagery or the creator’s facial presence are significantly more persuasive than those relying solely on captions. Based on these findings, we propose practical recommendations for improving the effectiveness of debunking videos, with the aim of promoting belief revision and mitigating misinformation on short video platforms.

Figure 1: Exemplified experimental setting for air pollution: participants were given a GUI to manipulate the different levels of particulate matter by moving a slider.The respective information was conveyed through the physical humanlike agent placed next to them.

Virtual AI-generated Influencers (VAIIs) are increasingly being used by corporations and public agencies, raising questions about how their visual design and communication strategies impact end-users’ propensity to accept the messages they deliver. We examined the impact of human-likeness (how closely a VAII resembles a human) and self-disclosure (whether the message contains personal information) on message acceptance, alongside dispositional factors like empathy and anthropomorphising tendencies. In a mixed-methods experiment, participants (N=120) watched short-form videos featuring VAIIs of varying human-likeness (High/Moderate-High/Moderate-Low/Low) and self-disclosure (present/absent). We observed the strongest message acceptance from the VAIIs with the lowest human-likeness, and message rejection for VAIIs with moderate to low human-likeness. Additionally, participants’ message acceptance was influenced by their empathy tendencies. Our qualitative analysis revealed further insights into participants’ perceptions of the human-likeness of VAIIs, their discomfort with self-disclosure, and their tendency to anthropomorphise VAIIs. These findings provide important implications for the design of VAIIs.

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.

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.

Social media has become a primary information source, with platforms evolving from text-based to multi-modal environments that include images and videos. While richer media modalities enhance user engagement, they also increase the spread and perceived credibility of misinformation. Most interventions to counter misinformation on social media are text-based, which may lack the persuasive power of richer modalities. This study explores whether the effectiveness of misinformation correction varies by modality, and if certain modalities of misinformation are better countered by a specific correction modality. We conducted a survey-based experiment where participants rated the credibility of misinformation tweets before and after exposure to corrections, across all combinations of text, images and video modalities. Our findings suggest that corrections are most effective when their modality richness matches that of the original misinformation. We discuss factors affecting the perceived credibility of corrections and offer strategies to optimise misinformation correction.

Location privacy leaks can lead to unauthorised tracking, identity theft, and targeted attacks, compromising personal security and privacy. This study explores LLM-powered location privacy leaks associated with photo sharing on social media, focusing on user awareness, attitudes, and opinions. We developed and introduced an LLM-powered location privacy intervention app to 19 participants, who used it over a two-week period. The app prompted users to reflect on potential privacy leaks that a widely available LLM could easily detect, such as visual landmarks & cues that could reveal their location, and provided ways to conceal this information. Through in-depth interviews, we found that our intervention effectively increased users' awareness of location privacy and the risks posed by LLMs. It also encouraged users to consider the importance of maintaining control over their privacy data and sparked discussions about the future of location privacy-preserving technologies. Based on these insights, we offer design implications to support the development of future user-centred, location privacy-preserving technologies for social media photos.

Humans naturally seek to identify causes behind outcomes through causal attribution, yet Human-AI research often overlooks how users perceive causality behind AI decisions. We examine how this perceived locus of causality—internal or external to the AI—influences trust, and how decision stakes and outcome favourability moderate this relationship. Participants (N=192) engaged with AI-based decision-making scenarios operationalising varying loci of causality, stakes, and favourability, evaluating their trust in each AI. We find that internal attributions foster lower trust as participants perceive the AI to have high autonomy and decision-making responsibility. Conversely, external attributions portray the AI as merely “a tool” processing data, reducing its perceived agency and distributing responsibility, thereby boosting trust. Moreover, stakes moderate this relationship—external attributions foster even more trust in lower-risk, low-stakes scenarios. Our findings establish causal attribution as a crucial yet underexplored determinant of trust in AI, highlighting the importance of accounting for it when researching trust dynamics.

Can you move it to

Recent advances in multi-agentic systems (e.g., AutoGen, OpenAI Agents) allow users to interact with a group of specialised AI agents rather than a single general-purpose agent. Despite the promise of this new paradigm, the HCI community has yet to fully examine the opportunities, risks, and user-centred challenges it introduces. We contribute to research on multi-agentic systems by exploring their architectures and key features through a human-centred lens. While literature and use cases are still emerging, we build on existing tools and frameworks available to developers to identify a set of overarching challenges, e.g., orchestration and conflict resolution, that can guide future research in HCI. We illustrate these challenges through examples, present potential design considerations, and provide research opportunities to spark interdisciplinary conversation. Our work lays the groundwork for future exploration and offers a research agenda focused on user-centred design in multi-agentic systems.

Social conformity occurs when individuals forego their personal judgements to agree with opposing judgements of a group majority. While conformity was initially observed and investigated in physical groups, recently there has been an increasing interest to understand dynamics of this phenomenon in online group settings. This survey summarises 36 years of social conformity literature (1988–2023) and reviews its occurrences, positive and negative outcomes, and contextual and personal determinants in online groups. By doing so, we identify gaps in the conformity literature that require further investigation, discuss common challenges and inform the design of future online conformity studies.

In recent years, the integration of location-based services into social media platforms has seen a significant surge, coinciding with the growing challenges posed by the proliferation of fake news online. However, the influence of location data on readers’ perceptions of online news credibility, particularly in relation to the reporters’ whereabouts, remains unclear. To investigate this relationship, we conducted a 3 (Topics: crime, science, health) × 2 (Location anchor: event-anchored or participant-anchored) × 4 (Proximity to location anchor - no, same, close-by or faraway location) mixed-method online study (N = 288) on Prolific. Our data collection involved presenting participants with news articles and assessing their credibility assessments and sharing intentions based on the proximity of those disseminating the news to both the subject matter of the news and the audience consuming it. Our findings reveal that the proximity of the reporter’s location to the readers’ location had a noticeable adverse impact on perceptions of news credibility and the likelihood of sharing it. Furthermore, we also identified a weak positive correlation between sharing intentions and trust in social media platforms. In addition, we observed that crime news were generally perceived as less credible compared to health and science news. Our research contributes significantly to a nuanced understanding of how location-based cues impact user behaviour when interacting with online news articles. Furthermore, it provides design insights for social media platforms aiming to enhance user trust and promote pro-social behaviours.

ChatGPT and other large language models (LLMs) have proven useful in crowdsourcing tasks, where they can effectively annotate machine learning training data. However, this means that they also have the potential for misuse, specifically to automatically answer surveys. LLMs can potentially circumvent quality assurance measures, thereby threatening the integrity of methodologies that rely on crowdsourcing surveys. In this paper, we propose a mechanism to detect LLM-generated responses to surveys. The mechanism uses ''prompt injection,'' such as directions that can mislead LLMs into giving predictable responses. We evaluate our technique against a range of question scenarios, types, and positions, and find that it can reliably detect LLM-generated responses with more than 98% effectiveness. We also provide an open-source software to help survey designers use our technique to detect LLM responses. Our work is a step in ensuring that survey methodologies remain rigorous vis-a-vis LLMs.

The start of the 2020s ushered in a new era of AI through the rise of Generative AI Large Language Models (LLMs) such as ChatGPT. These AI chatbots offer a form of interactive agency by enabling users to ask questions and query for more information. However, prior research only considers if LLMs have a political bias or agenda, and not how a biased LLM can impact a user’s opinion and trust. Our study bridges this gap by investigating a scenario where users read online news articles and then engage with an interactive AI chatbot, where both the news and the AI are biased to hold a particular stance on a news topic. Interestingly, participants were far more likely to adopt the narrative of a biased chatbot over news articles with an opposing stance. Participants were also substantially more inclined to adopt the chatbot’s narrative if its stance aligned with the news—all compared to a control news-article only group. Our findings suggest that the very interactive agency offered by an AI chatbot significantly enhances its perceived trust and persuasive ability compared to the ‘ static ’ articles from established news outlets, raising concerns about the potential for AI-driven indoctrination. We outline the reasons behind this phenomenon and conclude with the implications of biased LLMs for HCI research, as well as the risks of Generative AI undermining democratic integrity through AI-driven Information Warfare.

Online polarisation can tear the fabric of civility through reinforcing social media’s perceptions of division and discord. Social media platforms often rely on content-moderation to combat polarisation, contingent on the reactive removal or flagging of content. However, this approach often remains agnostic of the underlying debate’s ideas and stifles open discourse. In this study, we use prompt-tuned language models to mediate social media debates, applying the strategies of the Thomas-Kilmann Conflict Mode Instrument (TKI). We evaluate multiple mediation strategies in providing targeted responses to the debates, as shown to a debate audience. Our findings show that high-cooperativeness TKI strategies offered more persuasive arguments, while an accommodating argument strategy was the most successful at depolarising the audience’s opinion. Furthermore, high-cooperativeness strategies also increased the perception that the debaters will reach a consensus. Our work paves the way for scalable and personalised tools that mediate social media debates to encourage depolarisation.

Robots are embodied agents that act under several sources of uncertainty. When assisting humans in a collaborative task, robots need to communicate their uncertainty to help inform decisions. In this study, we examine the use of visualising a robot’s uncertainty in a high-stakes assisted decision-making task. In particular, we explore how different modalities of uncertainty visualisations (graphical display vs. the robot’s embodied behaviour) and confidence levels (low, high, 100%) conveyed by a robot affect the human decision-making and perception during a collaborative task. Our results show that these visualisations significantly impact how participants arrive to their decision as well as how they perceive the robot’s transparency across the different confidence levels. We highlight potential trade-offs and offer implications for robot-assisted decision-making. Our work contributes empirical insights on how humans make use of uncertainty visualisations conveyed by a robot in a critical robot-assisted decision-making scenario.

Augmented Reality (AR) offers unique opportunities for contributing to zoos’ objectives of public engagement and education about animal and conservation issues. However, the diversity of animal exhibits pose challenges in designing AR applications that are not encountered in more controlled environments, such as museums. To support the design of AR applications that meaningfully engage the public with zoo objectives, we first conducted two scoping reviews to interrogate previous work on AR and broader technology use at zoos. We then conducted a workshop with zoo representatives to understand the challenges and opportunities in using AR to achieve zoo objectives. Additionally, we conducted a field trip to a public zoo to identify exhibit characteristics that impacts AR application design. We synthesise the findings from these studies into a framework that enables the design of diverse AR experiences. We illustrate the utility of the framework by presenting two concepts for feasible AR applications.

Misinformation is rampant on social media, and existing platform-supplied interventions offer limited effectiveness. In this study, we examine the effectiveness of credibility labels that dispute the accuracy of information when they are supplied by one’s peers at different levels of relationship closeness and political agreement. We investigate four variants of these labels using a 2 (strong vs. weak tie strength) x 2 (high vs. low political agreement) between-subjects factorial design. We find that credibility disputes raised by one’s co-partisans (peers with similar political beliefs) significantly reduced belief in misinformation, irrespective of one’s relationship closeness with the peer. Our findings also reveal that in contrast to prior literature, a peer’s knowledgeability may be more potent than trustworthiness in causing belief change, and that trust can sometimes manifest even in the credibility judgement of distant peers, when perceived to have expertise or a fact-checking tendency. We further highlight the dual nature of these credibility labels, discussing scenarios in which disputes by hyper-partisan members of the opposite party can enforce belief in misinformation. We conclude by discussing how peer-supplied credibility disputes can benefit social media, especially echo chambers with high political homophily, where disputes by a co-partisan may be met with less resistance and persuade significantly reduced belief in fake news.

Recent work on augmented reality (AR) has explored the use of adaptive agents to overcome attentional issues that negatively impact task performance. However, despite positive technical evaluations, adaptive agents have shown no significant improvements to user task performance in AR. Furthermore, previous works have primarily evaluated such agents using abstract tasks. In this paper, we develop an agent that observes user behaviour and performs appropriate actions to mitigate attentional issues in a realistic sense-making task in AR. We employ mixed methods to evaluate our agent in a between-subject experiment (N=60) to understand the agent’s effect on user task performance and behaviour. While we find no significant improvements in task performance, our analysis revealed that users’ preferences and trust in the agent affected their receptiveness of the agent’s recommendations. We discuss the pitfalls of autonomous agents and highlight the need to shift from designing better Human–AI interactions to better Human–AI collaborations.

As the number of applications installed on smartphones continues to grow, the task of effectively managing location privacy has become increasingly complex. In this paper, we explore the factors that influence users' privacy-preserving intentions and contrast them with their actual behaviours. In addition, we compare location privacy concerns across different apps investigating the impact of app-specific features on the willingness to disclose location information. Our findings highlight significant challenges in privacy management due to privacy fatigue and perceived usability. Furthermore, participants raised the importance of more uniform standards regarding location privacy settings across various applications, calling for more detailed and interactive well-informed consent processes that highlight the risks instead of the benefits of disclosing location information. This research contributes important insights towards the development of more effective privacy settings that can foster increased user engagement in managing location privacy on smartphones.

As misinformation increasingly proliferates on social media platforms, it has become crucial to explore how to best convey automated news credibility assessments to end-users, and foster trust in fact-checking AIs. In this paper, we investigate how model-agnostic, natural language explanations influence trust and reliance on a fact-checking AI. We construct explanations from four Conceptualisation Validations (CVs) - namely consensual, expert, internal (logical), and empirical - which are foundational units of evidence that humans utilise to validate and accept new information. Our results show that providing explanations significantly enhances trust in AI, even in a fact-checking context where influencing pre-existing beliefs is often challenging, with different CVs causing varying degrees of reliance. We find consensual explanations to be the least influential, with expert, internal, and empirical explanations exerting twice as much influence. However, we also find that users could not discern whether the AI directed them towards the truth, highlighting the dual nature of explanations to both guide and potentially mislead. Further, we uncover the presence of automation bias and aversion during collaborative fact-checking, indicating how users' previously established trust in AI can moderate their reliance on AI judgements. We also observe the manifestation of a 'boomerang'/backfire effect often seen in traditional corrections to misinformation, with individuals who perceive AI as biased or untrustworthy doubling down and reinforcing their existing (in)correct beliefs when challenged by the AI. We conclude by presenting nuanced insights into the dynamics of user behaviour during AI-based fact-checking, offering important lessons for social media platforms.

While UbiComp research has steadily improved the performance of localisation systems, the analysis of such datasets remains largely unaddressed. In this paper, we present a tool to facilitate querying and analysis of localisation time-series with a focus on semantic localisation. Drawing on well-established models to represent movement and mobility, we first develop a query language for localisation datasets. We then develop a software library in R that implements this querying. We use case studies to demonstrate how our programming tool can be used to query localisation datasets. Our work addresses an important gap in localisation research, by providing a flexible tool that can model and analyse localisation data programmatically and in real time.

Temporal target selection requires users to wait and trigger the selection input within a bounded time window, with a selection cursor that is expected to be delayed. This task conceptualizes, for example, a variety of game scenarios such as determining the timing of shooting a projectile towards a moving object. In this work, we explore models that predict “when” users typically perform a selection (i.e., user selection distribution) and their selection error rates in such tasks. We hypothesize that users react to temporal factors including “distance”, “width”, and “delay” as how they treat the corresponding variables in spatial target selection. The derived models are evaluated in a controlled experiment and an MTurk-based online study. Our research contributes new knowledge on user behavior in temporal target selection tasks and its potential connection with its spatial correspondence. Our models and conclusions can benefit both users and designers of relevant interactive applications.

This paper explores using mixed reality (MR) mirrors for supporting improvisational dance making. Motivated by the prevalence of mirrors in dance studios and inspired by Forsythe’s Improvisation Technologies, we conducted workshops with 13 dancers and choreographers to inform the design of future MR visualisation and annotation tools for dance. The workshops involved using a prototype MR mirror as a technology probe that reveals the spatial and temporal relationships between the reflected dancing body and its surroundings during improvisation; speed dating group interviews around future design ideas; follow-up surveys and extended interviews with a digital media dance artist and a dance educator. Our findings highlight how the MR mirror enriches dancers’ temporal and spatial perception, creates multi-layered presence, and affords appropriation by dancers. We also discuss the unique place of MR mirrors in the theoretical context of dance and in the history of movement visualisation, and distil lessons for broader HCI research.

Optimising the assignment of tasks to workers is an effective approach to ensure high quality in crowdsourced data - particularly in heterogeneous micro tasks. However, previous attempts at heterogeneous micro task assignment based on worker characteristics are limited to using cognitive skills, despite literature emphasising that worker performance varies based on other parameters. This study is an initial step towards understanding whether and how multiple parameters such as cognitive skills, mood, personality, alertness, comprehension skill, and social and physical context of workers can be leveraged in tandem to improve worker performance estimations in heterogeneous micro tasks. Our predictive models indicate that these parameters have varying effects on worker performance in the five task types considered – sentiment analysis, classification, transcription, named entity recognition and bounding box. Moreover, we note 0.003 - 0.018 reduction in mean absolute error of predicted worker accuracy across all tasks, when task assignment is based on models that consider all parameters vs. models that only consider workers’ cognitive skills. Our findings pave the way for the use of holistic approaches in micro task assignment that effectively quantify worker context.

The topic of algorithmic fairness is of increasing importance to the Human–Computer Interaction research community following accumulating concerns regarding the use and deployment of Artificial Intelligence-based systems. How we conduct research on algorithmic fairness directly influences our inferences and conclusions regarding algorithmic fairness. To better understand the methodological decisions of studies focused on people’s perceptions of algorithmic fairness, we systematic analysed relevant papers from the CHI and FAccT conferences. We identified 200 relevant papers published between 1993 and 2022 and assessed their study design, participant sample, and geographical location of participants and authors. Our results highlight that studies are predominantly cross-sectional, cover a wide range of participant roles, and that both authors and participants are primarily from the United States. Based on these findings, we reflect on the potential pitfalls and shortcomings in how the community studies algorithmic fairness.

Living with Parkinson’s Disease introduces a range of significant challenges into one’s daily life. While medical interventions exist to overcome some of these challenges, patient self-care techniques often form an essential complement to the treatments recommended by medical doctors. Knowledge on these self-care techniques often originates from those living with Parkinson’s themselves or their close caregivers, as they have the knowledge and experience required to assess self-care techniques. This so-called ‘patient knowledge’ is usually exchanged in peer meetings or discussion forums. Although vital to the Parkinson’s Disease community, this information is often difficult to access due to its unstructured format and the difficulty of navigating through online forums. We present an online tool that allows for contributing, assessing, and finally discovering Parkinson’s Disease self-care techniques. The custom discovery tool was populated with self-care knowledge by over 300 people with Parkinson’s and dozens of their carers, spanning areas such as daily well-being and using assistive equipment. Then, we invited patients to explore the discover features in a smaller scale trial. While well-received, our deployment highlighted several challenges that we further discuss in this paper. Overall, our study contributes to crowdsourced digital health solutions and provides both design and research implications to this challenging domain with a vulnerable user group.

The rapid development and ubiquity of mobile and wearable devices promises to enable researchers to monitor users’ granular emotional data in a less intrusive manner. Researchers have used a wide variety of mobile and wearable devices for this purpose, and have proposed various approaches to sense users’ emotional states. In this survey, we utilise three established digital libraries (ACM Digital Library,IEEE Xplore Digital Library, andSpringer Nature). We analysed and critically assessed the different approaches used in the three stages (perception, learning, inference) of a typical mobile emotion sensing framework, following a structured paper selection process. The contribution of this survey is three-fold; first, we document all the latest relevant literature on mobile emotion sensing research; second, we describe how mobile and wearable devices use their sensing and computing capabilities to monitor human emotions; third, we discuss challenges and opportunities of mobile emotion sensing to demonstrate the potential of this thriving field of research.

With the rapid development of mobile and wearable devices, it is increasingly possible to access users’ affective data in a more unobtrusive manner. On this basis, researchers have proposed various systems to recognize user’s emotional states. However, most of these studies rely on traditional machine learning techniques and a limited number of signals, leading to systems that either do not generalize well or would frequently lack sufficient information for emotion detection in realistic scenarios. In this paper, we propose a novel attention-based LSTM system that uses a combination of sensors from a smartphone (front camera, microphone, touch panel) and a wristband (photoplethysmography, electrodermal activity, and infrared thermopile sensor) to accurately determine user’s emotional states. We evaluated the proposed system by conducting a user study with 45 participants. Using collected behavioral (facial expression, speech, keystroke) and physiological (blood volume, electrodermal activity, skin temperature) affective responses induced by visual stimuli, our system was able to achieve an average accuracy of 89.2 percent for binary positive and negative emotion classification under leave-one-participant-out cross-validation. Furthermore, we investigated the effectiveness of different combinations of data signals to cover different scenarios of signal availability.

Non-invasive inspection and imaging techniques are used to acquire non-visible information embedded in samples. Typical applications include medical imaging, defect evaluation, and electronics testing. However, existing methods have specific limitations, including safety risks (e.g., X-ray), equipment costs (e.g., optical tomography), personnel training (e.g., ultrasonography), and material constraints (e.g., terahertz spectroscopy). Such constraints make these approaches impractical for everyday scenarios. In this article, we present a method that is low-cost and practical for non-invasive inspection in everyday settings. Our prototype incorporates a miniaturized near-infrared spectroscopy scanner driven by a computer-controlled 2D-plotter. Our work presents a method to optimize content embedding, as well as a wavelength selection algorithm to extract content without human supervision. We show that our method can successfully extract occluded text through a paper stack of up to 16 pages. In addition, we present a deep-learning-based image enhancement model that can further improve the image quality and simultaneously decompose overlapping content. Finally, we demonstrate how our method can be generalized to different inks and other layered materials beyond paper. Our approach enables a wide range of content embedding applications, including chipless information embedding, physical secret sharing, 3D print evaluations, and steganography.

Mirror reflections offer an intuitive and realistic Mixed Reality (MR) experience comparable to other MR interfaces. Their high visual fidelity, and the sensorimotor contingency from the reflected moving body, make the mirror an ideal instrument for MR movement guidance. The translucent two-way mirror display enables users to follow a virtual humanoid instructor’s movement accurately by visually matching it with their reflections. In this work, we conduct the first formal evaluation of movement acquisition performance with simple motor tasks, using visual guidance from an MR mirror and a humanoid virtual instructor. Our results of performance and subjective ratings indicate that, comparing with simulated virtual mirror and with traditional screen-based movement guidance, the real MR mirror yields better acquisition performance and stronger sense of embodiment with the reflection, for upper-body movement. But the benefits diminish with larger-range head movements. We provide design guidelines for future mirror movement guidance interfaces and MR mirror experiences at large.

On-body interfaces, which leverage the human body’s surface as an input or output platform, can provide new opportunities for designing VR interaction. However, it remains unclear how on-body interfaces can best support current VR systems that mainly rely on mid-air interaction. We propose BodyOn, a collection of six design patterns that leverage combined on-body and mid-air interfaces to achieve more effective 3D interaction. Specifically, a user may use thumb-on-finger gestures, finger-on-arm gestures, or on-body displays with mid-air input, including hand movement and orientation, to complete an interaction task. To test our design concepts, we implemented example interaction techniques based on BodyOn that can assist users in various 3D interaction tasks. We further conducted an expert evaluation using the techniques as probes to elicit immediate design issues that emerge from the novel combination of on-body and midair interaction. We provide insights that can inspire and inform the design of future 3D user interfaces.

Recognising and monitoring emotional states play a crucial role in mental health and well-being management. Importantly, with the widespread adoption of smart mobile and wearable devices, it has become easier to collect long-term and granular emotion-related physiological data passively, continuously, and remotely. This creates new opportunities to help individuals manage their emotions and well-being in a less intrusive manner using off-the-shelf low-cost devices. Pervasive emotion recognition based on physiological signals is, however, still challenging due to the difficulty to efficiently extract high-order correlations between physiological signals and users' emotional states. In this paper, we propose a novel end-to-end emotion recognition system based on a convolution-augmented transformer architecture. Specifically, it can recognise users' emotions on the dimensions of arousal and valence by learning both the global and local fine-grained associations and dependencies within and across multimodal physiological data (including blood volume pulse, electrodermal activity, heart rate, and skin temperature). We extensively evaluated the performance of our model using the K-EmoCon dataset, which is acquired in naturalistic conversations using off-the-shelf devices and contains spontaneous emotion data. Our results demonstrate that our approach outperforms the baselines and achieves state-of-the-art or competitive performance. We also demonstrate the effectiveness and generalizability of our system on another affective dataset which used affect inducement and commercial physiological sensors.

The emergence of big data combined with the technical developments in Artificial Intelligence has enabled novel opportunities for autonomous and continuous decision support. While initial work has begun to explore how human morality can inform the decision making of future Artificial Intelligence applications, these approaches typically consider human morals as static and immutable. In this work, we present an initial exploration of the effect of context on human morality from a Utilitarian perspective. Through an online narrative transportation study, in which participants are primed with either a positive story, a negative story or a control condition (N = 82), we collect participants' perceptions on technology that has to deal with moral judgment in changing contexts. Based on an in-depth qualitative analysis of participant responses, we contrast participant perceptions to related work on Fairness, Accountability and Transparency. Our work highlights the importance of contextual morality for Artificial Intelligence and identifies opportunities for future work through a FACT-based (Fairness, Accountability, Context and Transparency) perspective.

Smart speakers have become exceedingly popular and entered many people's homes due to their ability to engage users with natural conversations. Researchers have also looked into using smart speakers as an interface to collect self-reported health data through conversations. Responding to surveys prompted by smart speakers requires users to listen to questions and answer in voice without any visual stimuli. Compared to traditional web-based surveys, where users can see questions and answers visually, voice surveys may be more cognitively challenging. Therefore, to collect reliable survey data, it is important to understand what types of questions are suitable to be administered by smart speakers. We selected five common survey questionnaires and deployed them as voice surveys and web surveys in a within-subject study. Our 24 participants answered questions using voice and web questionnaires in one session. They then repeated the same study session after 1 week to provide a "retest'' response. Our results suggest that voice surveys have comparable reliability to web surveys. We find that, when using 5-point or 7-point scales, voice surveys take about twice as long as web surveys. Based on objective measurements, such as response agreement and test-retest reliability, and subjective evaluations of user experience, we recommend that researchers consider adopting the binary scale and 5-point numerical scales for voice surveys on smart speakers.

Abstract Interactive public displays are versatile two-way interfaces between the digital world and passersby. They can convey information and harvest purposeful data from their users. Surprisingly little work has exploited public displays for collecting tagged data that might be useful beyond a single application. In this work, we set to fill this gap and present two studies: (1) a field study where we investigated collecting biometrically tagged video-selfies using public kiosk-sized screens, and (2) an online narrative transportation study that further elicited rich qualitative insights on key emerging aspects from the first study. In the first study, a 61-day deployment resulted in 199 video-selfies with consent to leverage the videos in any non-profit research. The field study indicates that people are willing to donate even highly sensitive data about themselves in public. The subsequent online narrative transportation study provides a deeper understanding of a variety of issues arising from the first study that can be leveraged in the future design of such systems. The two studies combined in this article pave the way forward towards a vision where volunteers can, should they so choose, ethically and serendipitously help unleash advances in data-driven areas such as computer vision and machine learning in health care.

We investigate the use of a miniaturized Near-Infrared Spectroscopy (NIRS) device in an assisted decision-making task. We consider the real-world scenario of determining whether food contains gluten, and we investigate how end-users interact with our NIRS detection device to ultimately make this judgment. In particular, we explore the effects of different nutrition labels and representations of confidence on participants’ perception and trust. Our results show that participants tend to be conservative in their judgment and are willing to trust the device in the absence of understandable label information. We further identify strategies to increase user trust in the system. Our work contributes to the growing body of knowledge on how NIRS can be mass-appropriated for everyday sensing tasks, and how to enhance the trustworthiness of assisted decision-making systems.

The uptake of artificial intelligence-based applications raises concerns about the fairness and transparency of AI behaviour. Consequently, the Computer Science community calls for the involvement of the general public in the design and evaluation of AI systems. Assessing the fairness of individual predictors is an essential step in the development of equitable algorithms. In this study, we evaluate the effect of two common visualisation techniques (text-based and scatterplot) and the display of the outcome information (i.e., ground-truth) on the perceived fairness of predictors. Our results from an online crowdsourcing study (N = 80) show that the chosen visualisation technique significantly alters people’s fairness perception and that the presented scenario, as well as the participant’s gender and past education, influence perceived fairness. Based on these results we draw recommendations for future work that seeks to involve non-experts in AI fairness evaluations.

Attentional tunneling describes a phenomenon in Augmented Reality (AR) where users excessively focus on virtual content while neglecting their physical surroundings. This leads to the concern that users could neglect hazardous situations when using AR applications. However, studies have often confounded the role of the virtual content with the role of the associated task in inducing attentional tunneling. In this paper, we disentangle the impact of the associated task and of the virtual content on the attentional tunneling effect by measuring reaction times to events in two user studies. We found that presenting virtual content did not significantly increase user reaction times to events, but adding a task to the content did. This work contributes towards our understanding of the attentional tunneling effect on handheld AR devices, and highlights the need to consider both task and context when evaluating AR application usage.

This paper investigates integration, coordination, and transition strategies of gaze and hand input for 3D object manipulation in VR. Specifically, this work aims to understand whether incorporating gaze input can benefit VR object manipulation tasks, and how it should be combined with hand input for improved usability and efficiency. We designed four gaze-supported techniques that leverage different combination strategies for object manipulation and evaluated them in two user studies. Overall, we show that gaze did not offer significant performance benefits for transforming objects in the primary working space, where all objects were located in front of the user and within the arm-reach distance, but can be useful for a larger environment with distant targets. We further offer insights regarding combination strategies of gaze and hand input, and derive implications that can help guide the design of future VR systems that incorporate gaze input for 3D object manipulation.

Designing computational support for dance is an emerging area of HCI research, incorporating the cultural, experiential, and embodied characteristics of the third-wave shift. The challenges of recognising the abstract qualities of body movement, and of mediating between the diverse parties involved in the idiosyncratic creative process, present important questions to HCI researchers: how can we effectively integrate computing with dance, to understand and cultivate the felt dimension of creativity, and to aid the dance-making process? In this work, we systematically review the past twenty years of dance literature in HCI. We discuss our findings, propose directions for future HCI works in dance, and distil lessons for related disciplines.

Text entry is an important and frequent task in interactive devices including augmented reality head-mounted displays (AR HMDs). In current AR HMDs, there are still two main open challenges to overcome for efficient and usable text entry: arm fatigue due to mid-air input and visual occlusion because of their small see-through displays. To address these challenges, we present iText, a technique for AR HMDs that is hands-free and is based on an imaginary (invisible) keyboard. We first show that it is feasible and practical to use an imaginary keyboard on AR HMDs. Then, we evaluated its performance and usability with three hands-free selection mechanisms: eye blinks (E-Type), dwell (D-Type), and swipe gestures (G-Type). Our results show that users could achieve an average text entry speed of 11.95, 9.03 and 9.84 words per minutes (WPM) with E-Type, D-Type, and G-Type, respectively. Given that iText with E-Type outperformed the other two selection mechanisms in text entry rate and subjective feedback, we ran a third, 5-day study. Our results show that iText with E-Type can achieve an average text entry rate of 13.76 WPM with a mean word error rate of 1.5%. In short, iText can enable efficient eyes-free text entry and can be useful for various application scenarios in AR HMDs.

Spatial experience, or how humans experience a given space, has been a pivotal topic especially in urban-scale environments. On the human scale, HCI researchers have mostly investigated personal meanings or aesthetic and embodied experiences. In this paper, we investigate the human scale as an ensemble of individual spatial features. Through large-scale online questionnaires we first collected a rich set of spatial features that people generally use to characterize their surroundings. Second, we conducted a set of field interviews to develop a more nuanced understanding of the feature identified as most important: perceived safety. Our combined quantitative and qualitative analysis contributes to spatial understanding as a form of context information and presents a timely investigation into the perceived safety of human scale spaces. By connecting our results to the broader scientific literature, we contribute to the field of HCI spatial understanding.

Abstract Despite large investments in smartwatch development, the market growth remains smaller than forecasted. The purpose of smartwatch use remains unclear, indicated by the lack of large-scale adoption. Thus, we aim to better understand the early adoption and everyday smartwatch use. We investigate a diverse usage data of smartwatches logged over a period of up to 14 months from 79 individuals between December 2015 and March 2017, one of the largest wearable datasets collected. First, we identify both explorative and accepted behaviours that users exhibit and further investigate how the individual usage traits and features differ between the two categories. Our analysis offers an insightful perspective on how smartwatch use evolves organically. Our results improve our shared understanding of smartwatch use and users adapting their use of smartwatch over time to match the capabilities of the technology by validating numerous findings from previous literature.

Inspired by the increasing prevalence of digital voice assistants, we demonstrate the feasibility of using voice interfaces to deploy and complete crowd tasks. We have developed Crowd Tasker, a novel system that delivers crowd tasks through a digital voice assistant. In a lab study, we validate our proof-of-concept and show that crowd task performance through a voice assistant is comparable to that of a web interface for voice-compatible and voice-based crowd tasks for native English speakers. We also report on a field study where participants used our system in their homes. We find that crowdsourcing through voice can provide greater flexibility to crowd workers by allowing them to work in brief sessions, enabling multi-tasking, and reducing the time and effort required to initiate tasks. We conclude by proposing a set of design guidelines for the creation of crowd tasks for voice and the development of future voice-based crowdsourcing systems.

In this paper, we demonstrate the existence of a bidirectional causal relationship between smartphone application use and user emotions. In a two-week long in-the-wild study with 30 participants we captured 502,851 instances of smartphone application use in tandem with corresponding emotional data from facial expressions. Our analysis shows that while in most cases application use drives user emotions, multiple application categories exist for which the causal effect is in the opposite direction. Our findings shed light on the relationship between smartphone use and emotional states. We furthermore discuss the opportunities for research and practice that arise from our findings and their potential to support emotional well-being.

Although crowd work is typically completed through desktop or laptop computers by workers at their home, literature has shown that crowdsourcing is feasible through a wide array of computing devices, including smartphones and digital voice assistants. An integrated crowdsourcing platform that operates across multiple devices could provide greater flexibility to workers, but there is little understanding of crowd workers’ perceptions on uptaking crowd tasks across multiple contexts through such devices. Using a crowdsourcing survey task, we investigate workers’ willingness to accept different types of crowd tasks presented on three device types in different scenarios of varying location, time and social context. Through analysis of over 25,000 responses received from 329 crowd workers on Amazon Mechanical Turk, we show that when tasks are presented on different devices, the task acceptance rate is 80.5% on personal computers, 77.3% on smartphones and 70.7% on digital voice assistants. Our results also show how different contextual factors such as location, social context and time influence workers decision to accept a task on a given device. Our findings provide important insights towards the development of effective task assignment mechanisms for cross-device crowd platforms.

Studies using augmented reality (AR) technology have suggested that users focus excessively on the virtual content in the AR environment at the expense of the physical world around them. This has implications related to the design of installations that aim to incorporate the user's physical environment as part of the AR experience. To better understand how user attention is managed in an AR environment, we present an observational study of Rewild Our Planet, a multi-modal installation that combined video, audio, a human docent and mobile AR to promote awareness about environmental issues. We found that, while AR was successful in engaging visitors, it drew attention away from other modalities within the installation. This impacts the work of the human docent and affects how visitors absorb information presented in the installation. Based on these observations, we present guidelines to inform the design of future AR-supported installations with the aim of minimizing or taking advantage of the observed attentional issues.

Selection studies are prevalent and indispensable for VR research. However, due to the tedious and repetitive nature of many such experiments, participants can become disengaged during the study, which is likely to impact the results and conclusions. In this work, we investigate participant disengagement in VR selection experiments and how this issue affects the outcomes. Moreover, we evaluate the usefulness of four engagement strategies to keep participants engaged during VR selection studies and investigate how they impact user performance when compared to a baseline condition with no engagement strategy. Based on our findings, we distill several design recommendations that can be useful for future VR selection studies or user tests in other domains that employ similar repetitive features.

Quantum computing (QC) is an intrinsically complex yet exciting discipline with increasing practical relevance. A deep understanding of QC requires the integration of knowledge across numerous technical fields, such as physics, computing and mathematics. This work aims to investigate how immersive Virtual Reality (VR) compares to a desktop environment (‘web-applet’) as an educational tool to help teach individuals QC fundamentals. We developed two interactive learning tutorials, one utilising the ‘Bloch sphere’ visualisation to represent a single-qubit system, and the other exploring multi-qubit systems through the lens of ‘quantum entanglement’. We evaluate the effectiveness of each medium to teach QC fundamentals in a user study with 24 participants. We find that the Bloch sphere visualisation was well-suited to VR over a desktop environment. Our results also indicate that mathematics literacy is an important factor in facilitating greater learning with this effect being notably more pronounced when using VR. However, VR did not significantly improve learning in a multi-qubit context. Our work provides valuable insights which contribute to the emerging field of Quantum HCI (QHCI) and VR for education.

While crowd workers typically complete a variety of tasks in crowdsourcing platforms, there is no widely accepted method to successfully match workers to different types of tasks. Researchers have considered using worker demographics, behavioural traces, and prior task completion records to optimise task assignment. However, optimum task assignment remains a challenging research problem due to limitations of proposed approaches, which in turn can have a significant impact on the future of crowdsourcing. We present 'CrowdCog', an online dynamic system that performs both task assignment and task recommendations, by relying on fast-paced online cognitive tests to estimate worker performance across a variety of tasks. Our work extends prior work that highlights the effect of workers' cognitive ability on crowdsourcing task performance. Our study, deployed on Amazon Mechanical Turk, involved 574 workers and 983 HITs that span across four typical crowd tasks (Classification, Counting, Transcription, and Sentiment Analysis). Our results show that both our assignment method and recommendation method result in a significant performance increase (5% to 20%) as compared to a generic or random task assignment. Our findings pave the way for the use of quick cognitive tests to provide robust recommendations and assignments to crowd workers.

Social conformity occurs when individuals in group settings change their personal opinion to be in agreement with the majority's position. While recent literature frequently reports on conformity in online group settings, the causes for online conformity are yet to be fully understood. This study aims to understand how social presencei.e., the sense of being connected to others via mediated communication, influences conformity among individuals placed in online groups while answering subjective and objective questions. Acknowledging its multifaceted nature, we investigate three aspects of online social presence: user representation (generic vs.user-specific avatars), interactivity (discussion vs.no discussion ), and response visibility (public vs.private ). Our results show an overall conformity rate of 30% and main effects from task objectivity, group size difference between the majority and the minority, and self-confidence on personal answer. Furthermore, we observe an interaction effect between interactivity and response visibility, such that conformity is highest in the presence of peer discussion and public responses, and lowest when these two elements are absent. We conclude with a discussion on the implications of our findings in designing online group settings, accounting for the effects of social presence on conformity.

The presence of fully-occluded targets is common within virtual environments, ranging from a virtual object located behind a wall to a datapoint of interest hidden in a complex visualization. However, efficient input techniques for locating and selecting these targets are mostly underexplored in virtual reality (VR) systems. In this paper, we developed an initial set of seven techniques techniques for fully-occluded target selection in VR. We then evaluated their performance in a user study and derived a set of design implications for simple and more complex tasks from our results. Based on these insights, we refined the most promising techniques and conducted a second, more comprehensive user study. Our results show how factors, such as occlusion layers, target depths, object densities, and the estimation of target locations, can affect technique performance. Our findings from both studies and distilled recommendations can inform the design of future VR systems that offer selections for fully-occluded targets.

Reaching towards out-of-sight objects during walking is a common task in daily life, however the same task can be challenging when wearing immersive Head-Mounted Displays (HMD). In this paper, we investigate the effects of spatial reference frame, walking path curvature, and target placement relative to the body on user performance of manually acquiring out-of-sight targets located around their bodies, as they walk in a spatial-mapping Mixed Reality (MR) environment wearing an immersive HMD. We found that walking and increased path curvature negatively affected the overall spatial accuracy of the performance, and that the performance benefited more from using the torso as the reference frame than the head. We also found that targets placed at maximum reaching distance yielded less error in angular rotation and depth of the reaching arm. We discuss our findings with regard to human walking kinesthetics and the sensory integration in the peripersonal space during locomotion in immersive MR. We provide design guidelines for future immersive MR experience featuring spatial mapping and full-body motion tracking to provide better embodied experience.

Crowdsourcing is a powerful approach for tapping into the collective insights of diverse crowds. Thus, crowdsourcing has potential to support designers in making sense of a design space. In this hands-on workshop, we will brainstorm and conceptualise new user interfaces and crowdsourcing systems for supporting designers in the design process. The workshop consists of developmental discussions of ideas contributed by the participants. In brainstorming and design sessions in groups, the participants will ideate new crowd-powered systems and user interfaces that support the designer's divergent and convergent thinking.

Mobile self-reports are a popular technique to collect participant labelled data in the wild. While literature has focused on increasing participant compliance to self-report questionnaires, relatively little work has assessed response accuracy. In this paper, we investigate how participant context can affect response accuracy and help identify strategies to improve the accuracy of mobile self-report data. In a 3-week study we collect over 2,500 questionnaires containing both verifiable and non-verifiable questions. We find that response accuracy is higher for questionnaires that arrive when the phone is not in ongoing or very recent use. Furthermore, our results show that long completion times are an indicator of a lower accuracy. Using contextual mechanisms readily available on smartphones, we are able to explain up to 13% of the variance in participant accuracy. We offer actionable recommendations to assist researchers in their future deployments of mobile self-report studies.

The increased reliance on algorithmic decision-making in socially impactful processes has intensified the calls for algorithms that are unbiased and procedurally fair. Identifying fair predictors is an essential step in the construction of equitable algorithms, but the lack of ground-truth in fair predictor selection makes this a challenging task. In our study, we recruit 90 crowdworkers to judge the inclusion of various predictors for recidivism. We divide participants across three conditions with varying group composition. Our results show that participants were able to make informed decisions on predictor selection. We find that agreement with the majority vote is higher when participants are part of a more diverse group. The presented workflow, which provides a scalable and practical approach to reach a diverse audience, allows researchers to capture participants' perceptions of fairness in private while simultaneously allowing for structured participant discussion.

Social conformity occurs when an individual changes their behaviour in line with the majority's expectations. Although social conformity has been investigated in small group settings, the effect of gender - of both the individual and the majority/minority - is not well understood in online settings. Here we systematically investigate the impact of groups' gender composition on social conformity in online settings. We use an online quiz in which participants submit their answers and confidence scores, both prior to and following the presentation of peer answers that are dynamically fabricated. Our results show an overall conformity rate of 39%, and a significant effect of gender that manifests in a number of ways: gender composition of the majority, the perceived nature of the question, participant gender, visual cues of the system, and final answer correctness. We conclude with a discussion on the implications of our findings in designing online group settings, accounting for the effects of gender on conformity.

Mobile crowd sensing (MCS) is a relatively new paradigm for collecting real-time and location-dependent urban sensing data. Given its applications, it is crucial to optimize the MCS process with the objective of maximizing the sensing quality and minimizing the sensing cost. While earlier studies mainly tackle this issue by designing different combinatorial optimization algorithms, there is a new trend to further optimize MCS by integrating learning techniques to extract knowledge, such as participants' behavioral patterns or sensing data correlation. In this paper, we perform an extensive literature review of learning-assisted optimization approaches in MCS. Specifically, from the perspective of the participant and the task, we organize the existing work into a conceptual framework, present different learning and optimization methods, and describe their evaluation. Furthermore, we discuss how different techniques can be combined to form a complete solution. In the end, we point out existing limitations, which can inform and guide future research directions.

Low back pain (LBP) is a globally common condition with no silver bullet solutions. Further, the lack of therapeutic consensus causes challenges in choosing suitable solutions to try. In this work, we crowdsourced knowledge bases on LBP treatments. The knowledge bases were used to rank and offer best-matching LBP treatments to end users. We collected two knowledge bases: one from clinical professionals and one from non-professionals. Our quantitative analysis revealed that non-professional end users perceived the best treatments by both groups as equally good. However, the worst treatments by non-professionals were clearly seen as inferior to the lowest ranking treatments by professionals. Certain treatments by professionals were also perceived significantly differently by non-professionals and professionals themselves. Professionals found our system handy for self-reflection and for educating new patients, while non-professionals appreciated the reliable decision support that also respected the non-professional opinion.

Online crowdsourcing enables the distribution of work to a global labor force as small and often repetitive tasks. Recently, situated crowdsourcing has emerged as a complementary enabler to elicit labor in specific locations and from specific crowds. Teamwork in online crowdsourcing has been recently shown to increase the quality of output, but teamwork in situated crowdsourcing remains unexplored. We set out to fill this gap. We present a generic crowdsourcing platform that supports situated teamwork and provide experiences from a laboratory study that focused on comparing traditional online crowdsourcing to situated team-based crowdsourcing. We built a crowdsourcing desk that hosts three networked terminal displays. The displays run our custom team-driven crowdsourcing platform that was used to investigate collocated crowdsourcing in small teams. In addition to analyzing quantitative data, we provide findings based on questionnaires, interviews, and observations. We highlight 1) emerging differences between traditional and collocated crowdsourcing, 2) the collaboration strategies that teams exhibited in collocated crowdsourcing, and 3) that a priori team familiarity does not significantly affect collocated interaction in crowdsourcing. The approach we introduce is a novel multi-display crowdsourcing setup that supports collocated labor teams and along with the reported study makes specific contributions to situated crowdsourcing research.

Near Infrared Spectroscopy (NIRS) is a sensing technique in which near infrared light is transmitted into a sample, followed by light absorbance measurements at various wavelengths. This technique enables the inference of the inner chemical composition of the scanned sample, and therefore can be used to identify or classify objects. In this paper, we describe how to facilitate the use of NIRS by non- expert users in everyday settings. Our work highlights the key challenges of placing NIRS devices in the hands of non-experts. We develop a system to mitigate these challenges, and evaluate it in a user study. We show how NIRS technology can be successfully utilised by untrained users in an unsupervised manner through a special enclosure and an accompanying smartphone app. Finally, we discuss potential future developments of commoditised NIRS.

We seek to quantify smartwatch use, and establish differences and similarities to smartphone use. Our analysis considers use traces from 307 users that include over 2.8 million notifications and 800,000 screen usage events, and we compare our findings to previous work that quantifies smartphone use. The results show that smartwatches are used more briefly and more frequently throughout the day, with half the sessions lasting less than 5 seconds. Interaction with notifications is similar across both types of devices, both in terms of response times and preferred application types. We also analyse the differences between our smartwatch dataset and a dataset aggregated from four previously conducted smartphone studies. The similarities and differences between smartwatch and smartphone use suggest effect on usage that go beyond differences in form factor.

Previous work suggests that Quantified-Self applications can retain long-term usage with motivational methods. These methods often require intermittent attention requests with manual data input. This may cause unnecessary burden to the user, leading to annoyance, frustration and possible application abandonment. We designed a novel method that uses on-screen alert dialogs to transform recurrent smartphone usage sessions into moments of data contributions and evaluate how accurately machine learning can reduce unintended interruptions. We collected sensor data from 48 participants during a 4-week long deployment and analysed how personal device usage can be considered in scheduling data inputs. We show that up to 81.7% of user interactions with the alert dialogs can be accurately predicted using user clusters, and up to 75.5% of unintended interruptions can be prevented and rescheduled. Our approach can be leveraged by applications that require self-reports on a frequent basis and may provide a better longitudinal QS experience.

We present a mixed-methods study aimed at assessing artists' experiences of a digital art exhibition service, called StreetGallery, on a network of interactive displays situated in public urban locations. We ground our analysis using survey responses and in-depth interviews of artists who have exhibited their art in StreetGallery over the years. Findings from these studies indicate that the artists highly value StreetGallery's open and egalitarian access to art, and its contribution towards fusing novel digital technologies and art in public urban spaces. We conclude that platforms such as StreetGallery have the potential to challenge traditional paradigms of art gallery practices and public urban spaces as a stage for consumption and commerce.

We present a study that quantifies the effect of cold temperature on smartphone input performance, particularly on tapping tasks. Our results show that smartphone input performance decreases when completing tapping tasks in cold temperatures. We show that colder temperature is associated with lower throughput and less accurate performance when using the phone in both one-handed and two-handed operations. We also demonstrate that colder temperature is related to higher error rate when using the phone in one-handed operation only, but not two-handed. Finally, we identify a number of design recommendations from the literature that can be considered as a countermeasure to poorer smartphone input performance in completing tapping tasks in cold temperature.

It is becoming increasingly easy for researchers to develop context-aware applications for smartphones. A perennial challenge, however, is to convince a large number of people to install them and donate contextual data for scientific purposes. Our empirical study seeks to address this challenge by investigating how people's perception and attitude affect their willingness to donate context data to researchers and quantifies the effects of social signals on donation action-taking. Our findings indicate that the perceived need for donation and perceived organization reputation are key determinants in deciding whether to donate: people with altruistic personality do not necessarily donate if they cannot see the need to take an action. Furthermore, we provide evidence that even if people indicate a willingness to donate, they are hesitant to take action towards donating data unless catalysts like social signals (hints about the actions of others) are present.

We present a crowdsourcing methodology to elicit highly structured knowledge for arbitrary questions. The method elicits potential answers (“options”), criteria against which those options should be evaluated, and a ranking of the top “options.” Our study shows that situated crowdsourcing markets can reliably elicit/moderate knowledge to generate a ranking of options based on different criteria that correlate with established online platforms. Our evaluation also shows that local crowds can generate knowledge that is missing from online platforms and on how a local crowd perceives a certain issue. Finally, we discuss the benefits and challenges of eliciting structured knowledge from local crowds.

We present a study on the use of audio-based cues to help overcome the well-known issue of display blindness, i.e. to help people become aware of situated interactive public displays. We used three different types of auditory cues based on suggestions from literature, namely spoken message, auditory icon, and random melody, and also included a no-audio condition as control. The study ran for 8 days on a university campus using an in-the-wild design, during which both qualitative and quantitative data were gathered. Results show that audio in general is good at attracting attention to the displays, and spoken message in particular also helps people understand that the display in question is interactive.

Researchers who analyse smartphone usage logs often make the assumption that users who lock and unlock their phone for brief periods of time (e.g., less than a minute) are continuing the same "session" of interaction. However, this assumption is not empirically validated, and in fact different studies apply different arbitrary thresholds in their analysis. To validate this assumption, we conducted a field study where we collected user-labelled activity data through ESM and sensor logging. Our results indicate that for the majority of instances where users return to their smartphone, i.e., unlock their device, they in fact begin a new session as opposed to continuing a previous one. Our findings suggest that the commonly used approach of ignoring brief standby periods is not reliable, but optimisation is possible. We therefore propose various metrics related to usage sessions and evaluate various machine learning approaches to classify gaps in usage.

Research claims that users value the battery life of their smartphones, but no study to date has attempted to quantify battery value and how this value changes according to users' current context and needs. Previous work has quantified the monetary value that smartphone users place on their data (e.g., location), but not on battery life. Here we present a field study and methodology for systematically measuring the monetary value of smartphone battery life, using a reverse second-price sealed-bid auction protocol. Our results show that the prices for the first and last 10% battery segments differ substantially. Our findings also quantify the tradeoffs that users consider in relation to battery, and provide a monetary model that can be used to measure the value of apps and enable fair ad-hoc sharing of smartphone resources.

We present the development and evaluation of a situated crowdsourcing mechanism that estimates queue length in real time. The system relies on public interactive kiosks to collect human estimations about their queue waiting time. The system has been designed as a standalone tool that can be retrospectively embedded in a variety of locations without interfacing with billing or customer systems. An initial study was conducted in order to determine whether people who just joined the queue would differ in their estimates from people who were at the front of the queue. We then present our system's evaluation in four different restaurants over 19 weekdays. Our analysis shows how our system is perceived by users, and we develop 2 ways to optimise the waiting time estimation: by correcting the estimations based on the position of the input mechanism, and by changing the sliding window considered inputs to provide better prediction. Our analysis shows that approximately 7% of restaurant customers provided estimations, but even so our system can provide predictions with up to 2 minute mean absolute error.

We develop a Markov state transition model of smartphone screen use. We collected use traces from real-world users during a 3-month naturalistic deployment via an app-store. These traces were used to develop an analytical model which can be used to probabilistically model or predict, at runtime, how a user interacts with their mobile phone, and for how long. Unlike classification-driven machine learning approaches, our analytical model can be interrogated under unlimited conditions, making it suitable for a wide range of applications including more realistic automated testing and improving operating system management of resources.

We develop and evaluate a data hiding method that enables smartphones to encrypt and embed sensitive information into carrier streams of sensor data. Our evaluation considers multiple handsets and a variety of data types, and we demonstrate that our method has a computational cost that allows real-time data hiding on smartphones with negligible distortion of the carrier stream. These characteristics make it suitable for smartphone applications involving privacy-sensitive data such as medical monitoring systems and digital forensics tools.

We evaluate the situational impairments caused by cold ambient temperature on fine-motor movement and vigilance during mobile interaction. For this purpose, we tested two mobile phone applications that measure fine motor skills and vigilance in controlled temperature settings. Our results show that cold adversely affected participants' fine-motor skills performance, but not vigilance. Based on our results we highlight the importance of correcting measurements when investigating performance of cognitive tasks to take into account the physical element of the tasks. Finally, we identify a number of design recommendations from literature that can mitigate the adverse effect of cold ambiance on interaction with mobile devices.

There is a limited amount of scavenging alternatives for smartphones. We assess the feasibility of using indoor light to extend smartphones' battery life. We build a prototype charger that demonstrates that indoor light scavenging is a practical method that can substantially extend battery life on smartphones. The results show that it is feasible and practical to extend battery life with this energy harvesting method. We finally discuss certain obstacles that need to be overcome, especially the redesign of operating systems to account for energy harvesting.

In information systems (IS) literature, there is ongoing debate as to whether the field has become fragmented and lost its identity in response to the rapid changes of the field. The paper contributes to this discussion by providing quantitative measurement of the fragmentation or cohesiveness level of the field. A co-word analysis approach aiding in visualization of the intellectual map of IS is applied through application of clustering analysis, network maps, strategic diagram techniques, and graph theory for a collection of 47,467 keywords from 9551 articles, published in 10 major IS journals and the proceedings of two leading IS conferences over a span of 20 years, 1993 through 2012. The study identified the popular, core, and bridging topics of IS research for the periods 1993–2002 and 2003–2012. Its results show that research topics and subfields underwent substantial change between those two periods and the field became more concrete and cohesive, increasing in density. Findings from this study suggest that the evolution of the research topics and themes in the IS field should be seen as part of the natural metabolism of the field, rather than a process of fragmentation or disintegration.

We present an empirical study that investigates crowdsourcing performance in a situated market. Unlike online markets, situated crowdsourcing markets consist of workers who become serendipitously available for work in a particular location and context. So far, the literature has lacked a systematic study of task performance and uptake in such markets under varying incentives. In a 3-week field study, we demonstrate that in a situated crowdsourcing market, task uptake and accuracy are generally comparable with online markets. We also show that increasing task rewards in situated crowdsourcing leads to increased task uptake but not accuracy, while decreasing task rewards leads to decreases in both task uptake and accuracy.

We present a revisitation analysis of smartphone use to investigate the question: do smartphones induce usage habits? We analysed three months of application launch logs from 165 users in naturalistic settings. Our analysis reveals distinct clusters of applications and users which share similar revisitation patterns. However, we show that much of smartphone usage on a macro-level is very similar to web browsing on desktops, and thus argue that smartphone usage is driven by innate service needs rather than technology characteristics. On the other hand, on a micro-level we identify unique characteristics in smartphone usage, and we present a rudimentary model that accounts for 92% in the variability of our smartphone use.

The interaction process with interactive public displays can be viewed as a set of interaction phases. In this paper we report a wizard-of-Oz study that explores the last two phases: 1) subtle interaction, where users can interact with the display through gestures or movement, and 2) direct interaction, when users interact with the display by directly manipulating it through e.g. a touch-screen interface. We investigate the effect of these two interaction phases on at-glance information acquisition, and demonstrate that the presentation of such at-glance information items can help shorten interaction times during the direct interaction phase.

In this paper we present Game of Words, a crowdsourcing game for public displays that allows the creation of a keyword dictionary to describe locations. It relies on crowdsourcing and gamification to identify, filter, and rank keywords based on their relevance to the location of the public display itself. We demonstrate that crowdsourcing on public displays can leverage users' knowledge of their environment, can work with a generic gaming task, and can be deployed on displays with multiple concurrent services. Our analysis shows that our approach has important benefits, such as the ability to identify undesired input, provide words of high semantic relevance, as well as a broader scope of keywords. Finally, our analysis also demonstrates that the chosen game design coped well with the challenges of this complex setting (i.e. public urban space) by disincentivising incorrect use of the system.

This study employs hierarchical cluster analysis, strategic diagrams and network analysis to map and visualize the intellectual landscape of the CHI conference on Human Computer Interaction through the use of co-word analysis. The study quantifies and describes the thematic evolution of the field based on a total of 3152 CHI articles and their associated 16035 keywords published between 1994 and 2013. The analysis is conducted for two time periods (1994-2003, 2004-2013) and a comparison between them highlights the underlying trends in our community. More significantly, this study identifies the evolution of major themes in the discipline, and highlights individual topics as popular, core, or backbone research topics within HCI.

Projective tests are personality tests that reveal individuals' emotions (e.g., Rorschach inkblot test). Unlike direct question-based tests, projective tests rely on ambiguous stimuli to evoke responses from individuals. In this paper we develop one such test, designed to be delivered automatically, anonymously and to a large community through public displays. Our work makes a number of contributions. First, we develop and validate in controlled conditions a quantitative projective test that can reveal emotions. Second, we demonstrate that this test can be deployed on a large scale longitudinally: we present a four-week deployment in our university's public spaces where 1431 tests were completed anonymously by passers-by. Third, our results reveal strong diurnal rhythms of emotion consistent with results we obtained independently using the Day Reconstruction Method (DRM), literature on affect, well-being, and our understanding of our university's daily routine.

We present Pulse, a wireless magnetic communication protocol for smartphones. Pulse is designed for off-the-shelf Android smartphones with magnetometers, and encodes data in magnetic fields. We present the design and evaluation of Pulse in various conditions (e.g., different voltages, number of transfer channels). The system provides security due to its short range (~1 cm), it can reach a speed of up to 44 bits per second, and it is possible to run it on most mobile phones with a magnetometer. We present our evaluation and discuss practical use cases where Pulse can be used today.

The rapid growth of the Ubicomp field has recently raised concerns regarding its identity. These concerns have been compounded by the fact that there exists a lack of empirical evidence on how the field has evolved until today. In this study we applied co-word analysis to examine the status of Ubicomp research. We constructed the intellectual map of the field as reflected by 6858 keywords extracted from 1636 papers published in the HUC, UbiComp and Pervasive conferences during 1999--2013. Based on the results of a correspondence analysis we identify two major periods in the whole corpus: 1999--2007 and 2008--2013. We then examine the evolution of the field by applying graph theory and social network analysis methods to each period. We found that Ubicomp is increasingly focusing on mobile devices, and has in fact become more cohesive in the past 15 years. Our findings refute the assertion that Ubicomp research is now suffering an identity crisis.

Research suggests smartphone users face 'application overload', but literature lacks an in-depth investigation of how users manage their time on smartphones. In a 3-week study we collected smartphone application usage patterns from 21 participants to study how they manage their time interacting with the device. We identified events we term application micro-usage: brief bursts of interaction with applications. While this practice has been reported before, it has not been investigated in terms of the context in which it occurs (e.g., location, time, trigger and social context). In a 2-week follow-up study with 15 participants, we captured participants? context while micro-using, with a mobile experience sampling method (ESM) and weekly interviews. Our results show that about approximately 40% of application launches last less than 15 seconds and happen most frequently when the user is at home and alone. We further discuss the context, taxonomy and implications of application micro-usage in our field. We conclude with a brief reflection on the relevance of short-term interaction observations for other domains beyond mobile phones.

In an increasingly connected world, users access personal or shared data, stored "in the cloud" (e.g., Dropbox, Skydrive, iCloud) with multiple devices. Despite the popularity of cloud storage services, little work has focused on investigating cloud storage users' Quality of Experience (QoE), in particular on mobile devices. Moreover, it is not clear how users' context might affect QoE. We conducted an online survey with 349 cloud service users to gain insight into their usage and affordances. In a 2-week follow-up study, we monitored mobile cloud service usage on tablets and smartphones, in real-time using a mobile-based Experience Sampling Method (ESM) questionnaire. We collected 156 responses on in-situ context of use for Dropbox on mobile devices. We provide insights for future QoE-aware cloud services by highlighting the most important mobile contextual factors (e.g., connectivity, location, social, device), and how they affect users' experiences while using such services on their mobile devices.

We report a pre-study and a three-week in-the-wild deployment of a non-moderated interactive public display prototype designed as a communication extension for an established community. A pre-study was conducted to map existing practices in order to ground the design. We explore the adoption process of the display prototype as well as rhythms of usage. We discuss findings related to extensions of presence within the community, the impact of the display on the community's activities, as well as aspects of appropriation and co-design. We illustrate how the display was used to extend one's presence within the community in addition to existing means of communication. This opens up new design possibilities when social dynamics are carefully negotiated.

Research is increasingly highlighting the potential for situated crowdsourcing to overcome some crucial limitations of online crowdsourcing. However, it remains unclear whether a situated crowdsourcing market can be sustained, and whether worker supply responds to price-setting in such a market. Our work is the first to systematically investigate workers' behaviour and response to economic incentives in a situated crowdsourcing market. We show that the market-based model is a sustainable approach to recruiting workers and obtaining situated crowdsourcing contributions. We also show that the price mechanism is a very effective tool for adjusting the supply of labour in a situated crowdsourcing market. Our work advances the body of work investigating situated crowdsourcing.

Transitioning from bespoke single-purpose displays to multipurpose public interactive displays entails a number of challenges. One challenge is the development of usable mechanisms that allow users to explore the functionality and services on such displays. This article presents a field trial that employs AutoCardSorter, a tool that uses semantic similarity and clustering algorithms, to automatically group the available applications of a public interactive display into categories based on the developer-provided descriptions of each application. The results demonstrate that the grouping generated by AutoCardSorter improved both performance and self-reported usability measures compared to practitioners' existing grouping. In addition, the study investigated the interplay between grouping and interaction modality (i.e., public display vs. desktop). Results tend to support that grouping affects more the user experience with a multipurpose interactive display, but findings were insignificant. This work provides a way for public displays to dynamically update their offered services without sacrificing usability.

Although advances in technology now enable people to communicate ‘anytime, anyplace’, it is not clear how citizens can be motivated to actually do so. This paper evaluates the impact of three principles of psychological empowerment, namely perceived self-efficacy, sense of community and causal importance, on public transport passengers’ motivation to report issues and complaints while on the move. A week-long study with 65 participants revealed that self-efficacy and causal importance increased participation in short bursts and increased perceptions of service quality over longer periods. Finally, we discuss the implications of these findings for citizen participation projects and reflect on design opportunities for mobile technologies that motivate citizen participation.

Safeguarding personally identifiable information (PII) is crucial because such information is increasingly used to engineer privacy attacks, identity thefts and security breaches. But is it likely that individuals may choose to just share this information with strangers? This study examines how reciprocation can lead to the disclosure of PII between strangers in online social networking. We demonstrate that the widespread use of public, one-to-many, communication channels such as ‘wall posts’ and profile pages in online social networks poses an exception to the assumption that reciprocation happens on one-to-one channels. We find that individuals not only reciprocate and share PII when the disclosure of such information is private and directed towards them by a stranger, but also when the stranger shares this information through a public channel that is not directed towards anyone in particular. Implications for privacy and design are discussed.

In this study, we demonstrate how analysis of users’ social network structure—a topic that has remained until recently inconspicuous within Human-Computer Interaction (HCI) research on social systems—can contribute to our understanding of Social Networking Services (SNS) effect on users. Despite a consensus that SNS enhance people's social capital, prior studies on SNS have provided inconsistent evidence on this process. In a multipronged study, we analyze personality, social capital, and Facebook data from a cohort of participants to model the extent to which one's SNS reflects aspects of his or personality and affects his bridging social capital. Our empirically validated model shows that empathy and conscientiousness influence the structural holes in one's social network, which in turn affects bridging social capital. These findings highlight the importance of network structure as an intermediary between one's personality and the social benefits one reaps from using SNS. Our work demonstrates how the implicit structural information embedded in users’ social networks can provide key insights into users’ personality and social capital.

Awareness campaigns aiming to highlight the accessibility challenges affecting people with disabilities face an important challenge. They often describe the environmental features that pose accessibility barriers out of context, and as a result public cannot relate to the problems at hand. In this paper we demonstrate that contextual cues can enhance people's perception and understanding of accessibility. We describe a two-week study where our participants submitted reports of inaccessible spots all over the city through a web application. Using a 2x2 factorial design we contrast the impact of two types of contextual cues, visual cues (i.e., displaying a picture of the inaccessible spot) and location cues (i.e., ability to zoom-in the exact location). We measure participants' perceptions of accessibility and how they are challenged to consider their own limitations and barriers that may also affect themselves in certain circumstances. Our results suggest that visual cues led to a bigger sense of urgency while also improving participants' attitude towards disability.

Narrowcasting refers to the targeted segmentation of media dissemination, and has been proposed as a counterpart to broadcasting. We present an explorative study that evaluates narrowcasting as an approach to sharing in online social media. We test a narrowcasting prototype for Facebook with 54 participants over a four-week period. We outline the various strategies that participants used to appropriate narrowcasting, and report on participants' use and perceptions. We also report on the effects of default sharing options and gender on sharing behavior. Our work provides implications for online sharing, suggesting that narrowcasting is an effective strategy for online social platforms.

In this paper we adopt a network science approach to investigate empathy and its implications for online social networks. We demonstrate that empathy is closely linked to social capital - the findings suggest that individuals higher on cognitive empathic skill are overall likely to report both higher bridging and higher bonding social capital. On the other hand, attributes of network structure around the individual, quantified through networks analysis metrics, were related to cognitive empathy. Further, an examination of the interplay between network structure, social capital and empathy suggests that empathy facilitates the relation between network structure and social capital previously reported in literature. We discuss the implications of our findings for the understanding of empathy in the context of online social networks and for the design of these systems.

Most studies take for granted the critical first steps that prelude interaction with a public display: awareness of the interactive affordances of the display, and enticement to interact. In this paper we investigate mechanisms for enticing interaction on public displays, and study the effectiveness of visual signals in overcoming the 'first click' problem. We combined 3 atomic visual elements (color/greyscale, animation/static, and icon/text) to form 8 visual signals that were deployed on 8 interactive public displays on a university campus for 8 days. Our findings show that text is more effective in enticing interaction than icons, color more than greyscale, and static signals are more effective than animated. Further, we identify gender differences in the effectiveness of these signals. Finally, we identify a behavior termed "display avoidance" that people exhibit with interactive public displays.

Mobile phone user interfaces typically show an icon to indicate remaining battery, but not the amount of time the device can be used for, often forcing users to make faulty estimates and predictions about battery life. Here we report on two studies that capture users' experiences with a user-centered battery interface design. In Study 1, we analyze 12 participants' use of mobile phones, demonstrating that mobile phone users do not know how or what to do to extend their mobile's battery life. We further identify the information they rely on to assess battery life. In Study 2, we use this information to design, prototype and evaluate an interactive battery interface (IBI) with another 22 participants. Our findings describe how users perceive battery life and how we used their mental models of mobile phone batteries to create IBI. Lastly, we report on the users' experiences and IBI's effect on battery lifetime, showing gains of approximately 27% over the course of a day.

This study is the first attempt to investigate altruistic use of interactive public displays in natural usage settings as a crowdsourcing mechanism. We test a non-paid crowdsourcing service on public displays with eight different motivation settings and analyse users' behavioural patterns and crowdsourcing performance (e.g., accuracy, time spent, tasks completed). The results show that altruistic use, such as for crowdsourcing, is feasible on public displays, and through the controlled use of motivational design and validation check mechanisms, performance can be improved. The results shed insights on three research challenges in the field: i) how does crowdsourcing performance on public displays compare to that of online crowdsourcing, ii) how to improve the quality of feedback collected from public displays which tends to be noisy, and iii) identify users' behavioural patterns towards crowdsourcing on public displays in natural usage settings.