We evaluated the viability of using Large Language Models (LLMs) to trigger and personalize content in Just-in-Time Adaptive Interventions (JITAIs) in digital health. As an interaction pattern representative of context-aware computing, JITAIs are being explored for their potential to support sustainable behavior change, adapting interventions to an individual's current context and needs. Challenging traditional JITAI implementation models, which face severe scalability and flexibility limitations, we tested GPT-4 for suggesting JITAIs in the use case of heart-healthy activity in cardiac rehabilitation. Using three personas representing patients affected by CVD with varying severeness and five context sets per persona, we generated 450 JITAI decisions and messages. These were systematically evaluated against those created by 10 laypersons (LayPs) and 10 healthcare professionals (HCPs). GPT-4-generated JITAIs surpassed human-generated intervention suggestions, outperforming both LayPs and HCPs across all metrics (i.e., appropriateness, engagement, effectiveness, and professionalism). These results highlight the potential of LLMs to enhance JITAI implementations in personalized health interventions, demonstrating how generative AI could revolutionize context-aware computing.

Despite the growing popularity and applications of AI-driven chatbots across various domains, their potential to support people with dementia remains relatively under-explored in human–computer interaction (HCI). To address this gap, we conducted four workshops with 17 participants (5 with mild dementia and 12 informal caregivers) to explore the needs, concerns, and preferences of people with dementia regarding chatbot use. This study focuses specifically on the insights gained from participants with dementia. Reflexive thematic analysis revealed that participants expressed interest in AI chatbots for casual interactions, particularly to alleviate social isolation and maintain language skills. They also saw the chatbots’ potential to act as ‘digital diaries’ for memory triggers and to offer memory training exercises to help with challenges in recalling names and words. However, concerns were raised about over-reliance, awareness of conversing with a chatbot, and trust.

Many people from diverse backgrounds are interested in learning programming independently and often share their experiences on social media platforms. Despite this, little is known about their specific self-regulation needs, which are crucial for achieving their learning goals. This study employed qualitative analyses to examine the textual narratives of individuals engaged in informal programming learning across four social media platforms: Twitter (X), YouTube comments, Reddit, and the DEV Community. The aim was to understand the self-regulation challenges they face in their informal, interestbased learning as articulated through their textual narratives and to pinpoint potential support mechanisms for self-regulation in this context. We extracted a purposive sample from these platforms, targeting posts rich in meaning or indicative of self-regulation, and then conducted a thematic analysis. Our findings indicate that learners require personal learning spaces that support three critical self-regulation aids: refining learning goals, resuming learning during spaced sessions, and supporting reflective practice throughout the learning process. Furthermore, we observed that informal learners generally lack specifically designed tools to support their self-regulation, relying instead on their varying skills and utilising diverse, non-specific tools. This gap unveils significant design opportunities to address the specific self-regulation needs of this underserved population of learners on social media. We discuss the potential of a storytelling-centric approach in design to address the identified issues for this population, which not only aids in sense-making from their learning experiences through reflection but also promotes social learning and engagement.

Previous research on programming education has mainly focused on formal learning environments and the programming communities of children and professionals, leaving the daily practices and learning needs of adult novices who learn to program informally online under-explored. This study aims to address this gap by providing a background that contextualizes the challenges and practices of adult novices learning informally online, with a particular emphasis on self-regulation. A scoping review methodology was employed to explore existing literature on this topic. Of 510 papers returned in the search, only 8 papers met the inclusion criteria for the review. The findings indicate a lack of research on how to support self-regulation in online informal programming learning settings. Additionally, it revealed that adult novices may face challenges such as loneliness, frustration, and uncertainty when learning informally online. It also highlights the potential importance of peer support, metacognitive skills, and self-efficacy in fostering learning in this population. The study presents a novel focus on the role of self-regulation in the context of informal online programming learning and differs from previous literature that largely focuses on formal learning settings. The findings have the potential to inform the field of computing education by shedding light on this population and their specific learning needs.

Digital health technology utilizing wearables, IoT and mobile devices has been successfully applied in the monitoring of numerous diseases and conditions. However, intervention, in response to monitored data, is yet to benefit from technological support and continues to follow a traditional point-of-care delivery model by providers and health professionals. Mental health is an example of a critical health area in dire need for technology solutions to enable timely, effective and scalable interventions. This is especially the case with an increasing prevalence of mental health conditions and a declining capacity of the healthcare professional workforce. Numerous studies reveal the potential for peer support groups as an effective, scalable, cost-effective, first-line of response in mental health interventions. Peer support helps participants, at low and moderate risk, better understand their diseases or conditions and empowers them to take control of their own health. Peer support interactions also seems to inform health professionals with insights and intricate knowledge, making it effectively a learning health system. This paper proposes a software architecture to better enable "peer-sourcing". We present related work and show how the proposed architecture might draw similarity to and differences from crowd-sourcing architectures. We also present a study in which we interacted with service users (mental health patients) and mental healthcare professionals to better understand and elicit the key requirements for the software architecture.

We report experience in requirements elicitation of domain knowledge from experts in clinical and cognitive neurosciences. The elicitation target was a causal model for early signs of dementia indicated by changes in user behaviour and errors apparent in logs of computer activity. A Delphi-style process consisting of workshops with experts followed by a questionnaire was adopted. The paper describes how the elicitation process had to be adapted to deal with problems encountered in terminology and limited consensus among the experts. In spite of the difficulties encountered, a partial causal model of user behavioural pathologies and errors was elicited. This informed requirements for configuring data- and text-mining tools to search for the specific data patterns. Lessons learned for elicitation from experts are presented, and the implications for requirements are discussed as “unknown unknowns”, as well as configuration requirements for directing data-/text-mining tools towards refining awareness requirements in healthcare applications.

This paper reports on the first year of a three-year-long co-creation project with older adults. We focus our analysis on one particular workshop in which participants stopped designing and began to think about promoting the app we were co-creating. The workshop proved uniquely important for examining assumptions we had made about how and why the co-creation process would be successful. This paper concedes flaws in these assumptions and in the execution of the methodology as a way of illuminating dynamics that act on research projects in ways that are antithetical to effective co-creation. Reporting on the unexpected results of our participant engagements, we reveal new insights into the challenges in executing co-creation methodology.

Stammering is a speech disorder affecting approximately 1% of the worldwide population. It can have associated impacts on daily life, such as loss of confidence in social situations and increased anxiety levels (particularly when speaking to strangers). Work exploring the development of digital tools to support people who stammer (PwS) is emerging. However, there is a paucity of research engaging PwS in the design process, with participation being facilitated mainly in testing phases. In this paper, we describe the user-centered design, development and evaluation of StammerApp, a mobile application to support PwS. We contribute insights into the challenges and barriers that PwS experience day-to-day and reflect on the complexities of designing with this diverse group. Finally, we present a set of design recommendations for the development of tools to support PwS in their everyday interactions, and provide an example of how these might be envisioned through the StammerApp prototype.

We have been generally successful for transferring software engineering knowledge to industry through various forms of education. However, many challenges in software engineering training remain. A key amongst these is how best to energise software engineering education with real-world software engineering practices. This paper describes our experience of delivering a radically different approach based on the notion of a Software Design Studio. The Software Design Studio is both a lab for students engaged in conceiving, designing and developing software products as well as an approach for teaching software engineering in the lab which emphasizes practical hands-on work and experimentation. The feedback on the Software Design Studio -- from both staff and students -- has been outstanding. Although the programme is designed as a small, elite programme there is interest to see if the teaching methods can be transferred across to the much larger undergraduate programme in Computer Science. In this paper, we provide a detailed description of how our studio works in practice so that others, thinking of tak-ing a studio or studio-inspired approach, can use in designing their own courses.

Dementia is a serious threat to social sustainability. As life expectancy increases, more people are developing dementia. At the same time, demographic change is reducing the economically active part of the population. Care of people with dementia imposes great emotional and financial strain on sufferers, their families and society at large. In response, significant research resources are being focused on dementia. One research thread is focused on using computer technology to monitor people in at-risk groups to improve rates of early diagnosis. In this paper we provide an overview of dementia monitoring research and identify a set of scientific challenges for the engineering of dementia-monitoring software, with implications for other mental health self-management systems.

Studio-based learning for software engineering is a well-received concept, despite its apparent lack of uptake across institutions worldwide. Studio education affords a variety of highly desirable benefits, and is also popular amongst its students. This paper presents Lancaster University's software engineering studio, details of its implementation, observations made throughout its first year, evidence of its successes, and reflections against the recently defined studio framework. This paper aims to provide useful information for anyone that is considering utilizing a studio-based approach.

Novel envisioned systems face the risk of rejection by their target user community and the requirements engineer must be sensitive to the factors that will determine acceptance or rejection. Conventionally, technology acceptance is determined by perceived usefulness and ease-of-use, but in some domains other factors play an important role. In healthcare systems, particularly, ethical and emotional factors can be crucial. In this paper we describe an approach to requirements discovery that we developed for such systems. We describe how we have applied our approach to a novel system to passively monitor users for signs of cognitive decline consistent with the onset of dementia. A key challenge was eliciting users' reactions to emotionally charged events never before experienced by them at first hand. Our goal was to understand the range of users' emotional responses and their values and motivations, and from these formulate requirements that would maximise the likelihood of acceptance of the system. The problem was heightened by the fact that the key stakeholders were elderly people who represent a poorly studied user constituency. We discuss the elicitation and analysis methodologies used, and our experience with tool support. We conclude by reflecting on the affect issues for RE and for technology acceptance.

Studio-based teaching is a method commonly used in arts and design that emphasizes a physical "home" for students, problem-based and peer-based learning, and mentoring by academic staff rather than formal lectures. There have been some attempts to transfer studio-based teaching to software engineering education. In many ways, this is natural as software engineering has significant practical elements. However, attempts at software studios have usually ignored experiences and theory from arts and design studio teaching. There is therefore a lack of understanding of what "studio" really means, how well the concepts transfer to software engineering, and how effective studios are in practice. Without a clear definition of "studio", software studios cannot be properly evaluated for their impact on student learning nor can best and worst practices be shared between those who run studios. In this paper, we address this problem head-on by conducting a qualitative analysis of what "studio" really means in both arts and design. We carried out 15 interviews with a range of people with studio experiences and present an analysis and model for evaluation here. Our results suggest that there are many intertwined aspects that define studio education, but it is primarily the people and the culture that make a studio. Digital technology on the other hand can have an adverse effect on studios, unless properly recognised.

Software engineering is a team activity yet the programmer's key tool, the IDE, is still largely that of a soloist. This paper describes the vision, implementation and initial evaluation of CoffeeTable - a fully featured research prototype resulting from our reflections on the software design process. CoffeeTable exchanges the traditional IDE for one built around a shared interactive desk. The proposed solution encourages smooth transitions between agile and traditional modes of working whilst helping to create a shared vision and common reference frame - key to sustaining a good design. This paper also presents early results from the evaluation of CoffeeTable and offers some insights from the lessons learned. In particular, it highlights the role of developer tools and the software constructions that are shaped by them.