Objection Overruled! Lay People can Distinguish Large Language Models from Lawyers, but still Favour Advice from an LLMLarge Language Models (LLMs) are seemingly infiltrating every domain, and the legal context is no exception. In this paper, we present the results of three experiments (total N = 288) that investigated lay people's willingness to act upon, and their ability to discriminate between, LLM- and lawyer-generated legal advice. In Experiment 1, participants judged their willingness to act on legal advice when the source of the advice was either known or unknown. When the advice source was unknown, participants indicated that they were significantly more willing to act on the LLM-generated advice. The result of the source unknown condition was replicated in Experiment 2. Intriguingly, despite participants indicating higher willingness to act on LLM-generated advice in Experiments 1 and 2, participants discriminated between the LLM- and lawyer-generated texts significantly above chance-level in Experiment 3. Lastly, we discuss potential explanations and risks of our findings, limitations and future work.2025ESEike Schneiders et al.University of Nottingham, School of Computer Science; University of Southampton, School of Electronics and Computer ScienceHuman-LLM CollaborationAI-Assisted Decision-Making & AutomationAI Ethics, Fairness & AccountabilityCHI
How Humans Communicate Programming Tasks in Natural Language and Implications For End-User Programming with LLMsLarge language models (LLMs) like GPT-4 can convert natural-language descriptions of a task into computer code, making them a promising interface for end-user programming. We undertake a systematic analysis of how people with and without programming experience describe information-processing tasks (IPTs) in natural language, focusing on the characteristics of successful communication. Across two online between-subjects studies, we paired crowdworkers either with one another or with an LLM, asking senders (always humans) to communicate IPTs in natural language to their receiver (either a human or LLM). Both senders and receivers tried to answer test cases, the latter based on their sender's description. While participants with programming experience tended to communicate IPTs more successfully than non-programmers, this advantage was not overwhelming. Furthermore, a user interface that solicited example test cases from senders often, but not always, improved IPT communication. Allowing receivers to request clarification, though, was less successful at improving communication.2025MPMadison Pickering et al.University of ChicagoHuman-LLM CollaborationProgramming Education & Computational ThinkingCHI
Supportive Fintech for Individuals with Bipolar Disorder: Financial Data Sharing Preferences for Longitudinal Care ManagementFinancial stability is a key challenge for individuals living with bipolar disorder (BD). Symptomatic periods in BD are associated with poor financial decision-making, contributing to a negative cycle of worsening symptoms and an increased risk of bankruptcy. There has been an increased focus on designing supportive financial technologies (fintech) to address varying and intermittent needs across different stages of BD. However, little is known about this population’s expectations and privacy preferences related to financial data sharing for longitudinal care management. To address this knowledge gap, we have deployed a factorial vignette survey using the Contextual Integrity framework. Our data from individuals with BD (N=480) shows that they are open to sharing financial data for long term care management. We have also identified significant differences in sharing preferences across age, gender, and diagnostic subtype. We discuss the implications of these findings in designing equitable fintech to support this marginalized community.2024JBJeff Brozena et al.Pennsylvania State UniversityCognitive Impairment & Neurodiversity (Autism, ADHD, Dyslexia)Universal & Inclusive DesignCHI
Computational Notebooks as Co-Design Tools: Engaging Young Adults Living with Diabetes, Family Carers, and Clinicians with Machine Learning ModelsEngaging end user groups with machine learning (ML) models can help align the design of predictive systems with people’s needs and expectations. We present a co-design study investigating the benefits and challenges of using computational notebooks to inform ML models with end user groups. We used a computational notebook to engage young adults, carers, and clinicians with an example ML model that predicted health risk in diabetes care. Through co-design workshops and retrospective interviews, we found that participants particularly valued using the interactive data visualisations of the computational notebook to scaffold multidisciplinary learning, anticipate benefits and harms of the example ML model, and create fictional feature importance plots to highlight care needs. Participants also reported challenges, from running code cells to managing information asymmetries and power imbalances. We discuss the potential of leveraging computational notebooks as interactive co-design tools to meet end user needs early in ML model lifecycles.2023AAAmid Ayobi et al.University College LondonAI-Assisted Decision-Making & AutomationInteractive Data VisualizationIntelligent Tutoring Systems & Learning AnalyticsCHI
Hummer: Text Entry by Gaze and HumText entry by gaze is a useful means of hands-free interaction that is applicable in settings where dictation suffers from poor voice recognition or where spoken words and sentences jeopardize privacy or confidentiality. However, text entry by gaze still shows inferior performance and it quickly exhausts its users. We introduce text entry by gaze and hum as a novel hands-free text entry. We review related literature to converge to word-level text entry by analysis of gaze paths that are temporally constrained by humming. We develop and evaluate two design choices: “HumHum” and “Hummer.” The first method requires short hums to indicate the start and end of a word. The second method interprets one continuous humming as an indication of the start and end of a word. In an experiment with 12 participants, Hummer achieved a commendable text entry rate of 20.8 words per minute and outperformed HumHum and the gaze-only method EyeSwipe in both quantitative and qualitative measures.2021RHRamin Hedeshy et al.University of StuttgartEye Tracking & Gaze InteractionVoice User Interface (VUI) DesignCHI
Understanding the Design and Effectiveness of Peripheral Breathing Guide Use During Information WorkPeripheral breathing guides – tools designed to influence breathing while completing another primary task – have been proposed to provide physiological benefits during information work. While research has shown that guides can influence breathing rates under ideal conditions, there is little evidence that they can lead to underlying markers of physiological benefit under interrupted work conditions. Further, even if guides are effective during work tasks, it is unclear how personal and workplace factors affect peoples' willingness to adopt them for everyday use. In this paper, we present the results of a comparative, mixed-methods study of five different peripheral breathing guides. Our findings show that peripheral breathing guides are viable and can provide physiological markers of benefit during interrupted work. Further, we show that guides are effective – even when use is intermittent due to workplace distractions. Finally, we contribute guidelines to support the design of breathing guides for everyday information work.2021ATAaron Tabor et al.University of New BrunswickNotification & Interruption ManagementWorkplace Wellbeing & Work StressCHI
TAGSwipe: Touch Assisted Gaze Swipe for Text EntryThe conventional dwell-based methods for text entry by gaze are typically slow and uncomfortable. A swipe-based method that maps gaze path into words offers an alternative. However, it requires the user to explicitly indicate the beginning and ending of a word, which is typically achieved by tedious gaze-only selection. This paper introduces TAGSwipe, a bi-modal method that combines the simplicity of touch with the speed of gaze for swiping through a word. The result is an efficient and comfortable dwell-free text entry method. In the lab study TAGSwipe achieved an average text entry rate of 15.46 wpm and significantly outperformed conventional swipe-based and dwell-based methods in efficacy and user satisfaction.2020CKChandan Kumar et al.University of Koblenz–LandauEye Tracking & Gaze InteractionVoice User Interface (VUI) DesignCHI
Introducing Peripheral Awareness as a Neurological State for Human-computer IntegrationIn this work we introduce peripheral awareness as a neurological state for real-time human-computer integration, where the human is assisted by a computer to interact with the world. Changes to the field of view in peripheral awareness have been linked with quality of human performance. This instinctive narrowing of vision that occurs as a threat is perceived has implications in activities that benefit from the user having a wide field of view, such as cycling to navigate the environment. We present "Ena", a novel EEG-eBike system that draws from the user's neural activity to determine when the user is in a state of peripheral awareness to regulate engine support. A study with 20 participants revealed various themes and tactics suggesting that peripheral awareness as a neurological state is viable to align human-machine integration with internal bodily processes. Ena suggests that our work facilitates a safe and enjoyable human-computer integration experience.2020JAJosh Andres et al.Monash University & IBM ResearchMicromobility (E-bike, E-scooter) InteractionHand Gesture RecognitionBrain-Computer Interface (BCI) & NeurofeedbackCHI
Leveraging Error Correction in Voice-based Text Entry by Talk-and-GazeWe present the design and evaluation of Talk-and-Gaze (TaG), a method for selecting and correcting errors with voice and gaze. TaG uses eye gaze to overcome the inability of voice-only systems to provide spatial information. The user's point of gaze is used to select an erroneous word either by dwelling on the word for 800 ms (D-TaG) or by uttering a "select" voice command (V-TaG). A user study with 12 participants compared D-TaG, V-TaG, and a voice-only method for selecting and correcting words. Corrections were performed more than 20% faster with D-TaG compared to the V-TaG or voice-only methods. As well, D-TaG was observed to require 24% less selection effort than V-TaG and 11% less selection effort than voice-only error correction. D-TaG was well received in a subjective assessment with 66% of users choosing it as their preferred choice for error correction in voice-based text entry.2020KSKorok Sengupta et al.University of Koblenz-LandauHand Gesture RecognitionEye Tracking & Gaze InteractionCHI
Evaluating the Effect of Feedback from Different Computer Vision Processing Stages: A Comparative Lab StudyComputer vision and pattern recognition are increasingly being employed by smartphone and tablet applications targeted at lay-users. An open design challenge is to make such systems intelligible without requiring users to become technical experts. This paper reports a lab study examining the role of visual feedback. Our findings indicate that the stage of processing from which feedback is derived plays an important role in users' ability to develop coherent and correct understandings of a system's operation. Participants in our study showed a tendency to misunderstand the meaning being conveyed by the feedback, relating it to processing outcomes and higher level concepts, when in reality the feedback represented low level features. Drawing on the experimental results and the qualitative data collected, we discuss the challenges of designing interactions around pattern matching algorithms.2019JKJacob Kittley-Davies et al.University of SouthamptonExplainable AI (XAI)Algorithmic Transparency & AuditabilityPrivacy by Design & User ControlCHI
Efficient, but Effective? Volunteer Engagement in Short-term Virtual Citizen Science ProjectsVirtual citizen science (VCS) projects have proven to be a highly effective method to analyse large quantities of data for scientific research purposes. Yet if these projects are to achieve their goals, they must attract and maintain the interest of sufficient numbers of active, dedicated volunteers. Although CSCW and HCI research has typically focussed on designing platforms to support long-term engagement, in recent years a new project format has been trialled -- using short-term crowdsourcing activities lasting as little as 48 hours. In this paper, we explore two short-term projects to understand how they influence participant engagement in the task and discussion elements of VCS. We calculate descriptive statistics to characterise project participants. Additionally, using calculation of correlation coefficients and hypothesis testing, we identify factors influencing volunteer task engagement and the effect this has on project outcomes. Our findings contribute to the understanding of volunteer engagement in VCS.2019NRNeal T Reeves et al.Open CollaborationsCSCW
A Review & Analysis of Mindfulness Research in HCI: Framing Current Lines of Research and Future OpportunitiesMindfulness is a term seen with increasing frequency in HCI literature, and yet the term itself is used almost as variously as the number of papers in which it appears. This diversity makes comparing or evaluating HCI approaches around mindfulness or understanding the design space itself a challenging task. We conducted a structured ACM literature search based on the term mindfulness. Our selection process yielded 38 relevant papers, which we analyzed for their definition, motivation, practice, evaluation and technology use around mindfulness. We identify similarities, divergences and areas of interest for each aspect, resulting in a framework composed of four perspectives and seven lines of research. We highlight challenges and opportunities for future HCI research and design.2019NTNađa Terzimehić et al.Ludwig Maximilian University of MunichVisualization Perception & CognitionMental Health Apps & Online Support CommunitiesCHI
Beyond monetary incentives: experiments in paid microtask contestsIn this paper, we aim to gain a better understanding into how paid microtask crowdsourcing could leverage its appeal and scaling power by using contests to boost crowd performance and engagement. We introduce our microtask-based annotation platform Wordsmith, which features incentives such as points, leaderboards and badges on top of financial remuneration. Our analysis focuses on a particular type of incentive, contests, as a means to apply crowdsourcing in near-real-time scenarios, in which requesters need labels quickly. We model crowdsourcing contests as a continuous-time Markov chain with the objective to maximise the output of the crowd workers, while varying a parameter which determines whether a worker is eligible for a reward based on their present rank on the leaderboard. We conduct empirical experiments in which crowd workers recruited from CrowdFlower carry out annotation microtasks on Wordsmith - in our case, to identify named entities in a stream of Twitter posts. In the experimental conditions, we test different reward spreads and record the total number of annotations received. We compare the results against a control condition in which the same annotation task was completed on CrowdFlower without a time or contest constraint. The experiments show that rewarding only the best contributors in a live contest could be a viable model to deliver results faster, though quality might suffer for particular types of annotation tasks. Increasing the reward spread leads to more work being completed, especially by the top contestants. Overall, the experiments shed light on possible design improvements of paid microtasks platforms to boost task performance and speed, and make the overall experience more fair and interesting for crowd workers.2019OFOluwaseyi Feyisetan et al.MoneyCSCW
Tracking the Consumption of Home EssentialsPredictions of people's behaviour increasingly drive interactions with a new generation of IoT services designed to support everyday life in the home, from shopping to heating. Based on the premise that such automation is difficult due to the contingent nature of people's practices, in this work we explore the nature of these contingencies in depth. We have designed and conducted a technology probe that made use of simple linear predictions as a provocation, and invited people to track the life of their household essentials over a two-month period. Through a mixed-method approach we demonstrate the challenges of simple predictions, and in turn identify eight categories of contingencies that influenced prediction accuracy. We discuss strategies for how designers of future predictive IoT systems may take the contingencies into account by removing, hiding, revealing, managing, or exploiting the system uncertainty at the core of the issue.2019CFCarolina Fuentes et al.University of NottinghamContext-Aware ComputingHome Energy ManagementCHI
Collaborative Practices with Structured Data: Do Tools Support What Users Need?Collaborative work with data is increasingly common and spans a broad range of activities - from creating or analysing data in a team, to sharing it with others, to reusing someone else's data in a new context. In this paper, we explore collaboration practices around structured data and how they are supported by current technology. We present the results of an interview study with twenty data practitioners, from which we derive four high-level user needs for tool support. We compare them against the capabilities of twenty systems that are commonly associated with data activities, including data publishing software, wikis, web-based collaboration tools, and online community platforms. Our findings suggest that data-centric collaborative work would benefit from: structured documentation of data and its lifecycle; advanced affordances for conversations among collaborators; better change control; and custom data access. The findings help us formalise practices around data teamwork, and build a better understanding how people's motivations and barriers when working with structured data.2019LKLaura Koesten et al.University of SouthamptonCrowdsourcing Task Design & Quality ControlKnowledge Management & Team AwarenessCHI
The Body as Starting Point: Exploring Inside and Around Body Boundaries for Body-Centric Computing DesignMore HCI designs and devices are embracing what is being dubbed “body centric computing,” where designs both deliberately engage the body as the locus of interest, whether to move the body into play or relaxation, or to track and monitor its performance, or to use it as a surface for interaction. Most HCI researchers are engaging in these designs, however, with little direct knowledge of how the body itself works either as a set of complex internal systems or as sets of internal and external systems that interact dynamically. The science of how our body interacts with the microbiome around us also increasingly demonstrates that our presumed boundaries between what is inside and outside us may be misleading if not considered harmful. Developing both (1) introductory knowledge and (2) design practice of how these in-bodied and circum-bodied systems work with our understanding of the em-bodied self, and how this gnosis/praxis may lead to innovative new body-centric computing designs is the topic of this workshop.2018MSm.c. schraefel et al.University of SouthamptonFull-Body Interaction & Embodied InputHuman-Nature Relationships (More-than-Human Design)CHI
Is Virtual Citizen Science A Game?The use of game elements within virtual citizen science is increasingly common, promising to bring increased user activity, motivation, and engagement to large-scale scientific projects. However, there is an ongoing debate about whether or not gamifying systems such as these is actually an effective means by which to increase motivation and engagement in the long term. While gamification itself is receiving a large amount of attention, there has been little beyond individual studies to assess its suitability or success for citizen science; similarly, while frameworks exist for assessing citizen science performance, they tend to lack any appreciation of the effects that game elements might have had. We therefore review the literature to determine what the trends are regarding the performance of particular game elements or characteristics in citizen science, and survey existing projects to assess how popular different game features are. Investigating this phenomenon further, we then present the results of a series of interviews carried out with the EyeWire citizen science project team to understand more about how gamification elements are introduced, monitored, and assessed in a live project. Our findings suggest that projects use a range of game elements with points and leaderboards the most popular, particularly in projects that describe themselves as “games.” Currently, gamification appears to be effective in citizen science for maintaining engagement with existing communities, but shows limited impact for attracting new players.2018ESElena Simper et al.Citizen Science and Scientific ResearchCSCW
Who models the world? Collaborative Ontology Creation and User Roles in WikidataWikidata is a collaborative knowledge graph which is central to many academic and industry IT projects. Its users are responsible for maintaining the schema that organises this knowledge into classes, properties, and attributes, which together form the Wikidata 'ontology'. In this paper, we study the relationship between different Wikidata user roles and the quality of the Wikidata ontology. To do so we first propose a framework to evaluate the ontology as it evolves. We then cluster editing activities to identify user roles in monthly time frames. Finally, we explore how each role impacts the ontology. Our analysis shows that the Wikidata ontology has uneven breadth and depth. We identified two user roles: contributors and leaders. The second category is positively associated to ontology depth, with no significant effect on other features. Further work should investigate other dimensions to define user profiles and their influence on the knowledge graph.2018APAlessandro Piscopo et al.Collaboration in Online CommunitiesCSCW
The Body as Starting Point: Exploring Inside and Around Body Boundaries for Body-Centric Computing DesignMore HCI designs and devices are embracing what is being dubbed “body centric computing,” where designs both deliberately engage the body as the locus of interest, whether to move the body into play or relaxation, or to track and monitor its performance, or to use it as a surface for interaction. Most HCI researchers are engaging in these designs, however, with little direct knowledge of how the body itself works either as a set of complex internal systems or as sets of internal and external systems that interact dynamically. The science of how our body interacts with the microbiome around us also increasingly demonstrates that our presumed boundaries between what is inside and outside us may be misleading if not considered harmful. Developing both (1) introductory knowledge and (2) design practice of how these in-bodied and circum-bodied systems work with our understanding of the em-bodied self, and how this gnosis/praxis may lead to innovative new body-centric computing designs is the topic of this workshop.2018MSm.c. schraefel et al.University of SouthamptonFull-Body Interaction & Embodied InputEarly Childhood Education TechnologyHuman-Nature Relationships (More-than-Human Design)CHI
Common Barriers to the Use of Patient-Generated Data Across Clinical SettingsPatient-generated data, such as data from wearable fitness trackers and smartphone apps, are viewed as a valuable information source towards personalised healthcare. However, studies in specific clinical settings have revealed diverse barriers to their effective use. In this paper, we address the following question: are there barriers prevalent across distinct workflows in clinical settings to using patient-generated data? We conducted a two-part investigation: a literature review of studies identifying such barriers; and interviews with clinical specialists across multiple roles, including emergency care, cardiology, mental health, and general practice. We identify common barriers in a six-stage workflow model of aligning patient and clinician objectives, judging data quality, evaluating data utility, rearranging data into a clinical format, interpreting data, and deciding on a plan or action. This workflow establishes common ground for HCI practitioners and researchers to explore solutions to improving the use of patient-generated data in clinical practices.2018PWPeter West et al.University of SouthamptonMental Health Apps & Online Support CommunitiesTelemedicine & Remote Patient MonitoringCHI