With the rapid expansion of smart technologies, people using devices such as smartphones and tablets are becoming ubiquitous. This growth has also expanded communication methods, such as social media platforms, that can be utilized for seeking support. However, parents of autistic children often receive inadequate support from existing technology. In this paper, we report findings from a qualitative study with 14 parents of autistic children from five different countries. We analyzed these parents’ experience with receiving and utilizing social support, whether they consider received support adequate, challenges faced, and the types of support these parents seek. Our findings reveal a complex dynamic in the way parents of autistic children seek social support. For instance, the higher stress levels experienced by these parents motivate them to seek support from their families first. However, a lack of autism awareness among family members often leads them to seek support elsewhere, especially from online autism groups. Interestingly, unhealthy competition among parents of autistic children in many online autism groups led some parents to discontinue interacting in these groups and work independently. We provide guidelines for designing new interventions and offer guidelines to redesign existing technologies that may better align with the needs of parents of autistic children.

This research examines the use of advanced Natural Language Processing (NLP) technology, specifically GPT-3.5/4 models, to enhance text-based learning for neurodiverse students within higher education. Recognizing that conventional pedagogical resources often fail to meet the distinct needs of learners with neurodevelopmental differences, our research explores the hypothesis that NLP-enabled personalization of academic content can facilitate effective, equitable, and engaging educational experiences for college students. We present AImpathizer, a tool designed to adapt academic content into representations more congruent with the cognitive styles of neurodiverse students. AImpathizer is designed using a user-centric approach, highlighting the challenges and needs of neurodiverse college students. AImpathizer aims to incorporate more accessibility modifications in comparison to traditional academic content. The outcomes of this research aim to provide empirical evidence supporting the integration of NLP tools into educational environments in synergy with Universal Design for Learning (UDL) framework, paving the way for more inclusive educational practices.

Recent advances in large language models (LLMs) have brought AI-based systems into higher education, raising critical concerns regarding ethical and pedagogical implications. As AI-based systems become widely integrated in educational settings, questions about these concerns become increasingly important, yet current literature on best practices is still developing. Our work aims to guide researchers and educators by exploring the challenges and opportunities of AI-based systems in computer science higher education. We conducted a PRISMA-inspired literature review (N = 45), applying qualitative thematic analysis. Our findings suggest AI can foster a judgment-free space for underrepresented groups, especially considering the often competitive and defensive climate of computer science education, yet AI overuse may simultaneously undermine students’ belief in their competencies. We examine student and educator perspectives, linguistic nuances, policy considerations, and next steps. We finalize the discussion through classroom recommendations that reject the perceived dichotomy between academic AI policy and teaching AI literacy.

With the rapid growth and availability of smart technology, the idea of households using smart technology (smartphones, computers) is becoming popular. This availability of smart technology inside homes can potentially help with critical problems such as understanding the emotions of people with autism. Despite their high potential in solving complex problems, such technological solutions are largely unexplored. We take a first step in this direction by conducting a systematic literature review using three major databases: Google Scholar, ACM Digital Library, and IEEE Xplore. Our findings indicate that most of the existing approaches only utilize facial expressions for model development and collect data from children with autism, indicating a need for utilizing technology to capture more than facial expressions, capture contextual and other available information, and collect data from a diverse group of individuals with autism as it will improve the effectiveness of the machine learning models.

In recent years an increasing number of autistic students enrolled in college, many choosing computer science (CS) and related majors. However, the retention and graduation rates for autistic students are considerably lower than for neurotypical students. Research investigating factors that influence autistic students' success in college is sparse. More importantly, there is a scarcity of research that examines assessment techniques used and factors that influence the success of autistic CS students. Here, we report findings based on a systematic literature review$(\mathrm{N}=44)$that aims to examine the experience of autistic CS students to identify factors that influence their success given current assessment techniques. Our findings indicate that traditional accommodations, which in general are difficult to access and lack personalization, fall short of addressing many of the challenges autistic students face. The absence of a common vocabulary and communication style (e.g., verbal, instructions, teaching materials) and differences in learning style and cognitive processing between autistic students and neurotypical instructors hinders autistic students' academic success. Executive functioning (EF) challenges impact both academic and social aspects. However, accommodations targeted at addressing challenges related to EF are overlooked. Finally, despite many autistic students' affinity for technological interventions, their incorporation in accommodation is almost non-existent. We propose a set of actionable guidelines that could aid many of the challenges autistic students experience in CS programs, including designing a curriculum inspired by universal design, designing personalized accommodations, and incorporating technological interventions as part of the accommodation process.

Music Therapy (MT) has been well established as beneficial for people with Autism Spectrum Condition (ASC), commonly known as Autism Spectrum Disorder (ASD), by improving attention, motor synchrony, and prosocial behaviors. It has been shown to be increasing in popularity as an intervention choice for ASC due to its relative accessibility. Recent research has also shown promising results with integrating technology into traditional music therapy interventions to broaden and improve therapeutic modalities. In this paper, we conduct a systematic literature review on the relationship between MT and ASC, as well as current MT practices and use this to inform a set of design recommendations on technology designed for autistic people. This design framework will guide developers and therapists working to incorporate technology into a therapeutic setting with a strength-based (SB) approach that centers participants as stakeholders. Our review indicates that much of the current research focuses on the medical treatment of autistic behaviors. The research generally supports the efficacy of MT for autism and identifies it as a particular strength of the autistic community. However, it also highlights concerns regarding common outcome measures and standardization. We then put forth three design implications, supported by various findings, to further promote accessible MT for autistic people across the spectrum.

For many students, attending college is a dramatic but necessary change. To gain a better understanding of experiences that are unique to autistic college students, we conducted a mixed-method study with 20 students (10 autistic and 10 neurotypical). We collected physiological, contextual, experience, and environmental data from their natural environment using Fitbit and smartphones. We found that stress patterns, emotional states, and physical states are similar for both groups. Our autistic participants prioritized academic success over everything else, often intentionally confining their movements among academic, resident, and work locations to engage themselves with academic work as much as possible. They had a small number of friends, always preferring quality over quantity and sometimes regarding friends as close as family members. To maintain a better social life, they extensively used social media. They slept more than neurotypical participants per day; however, they experienced lower sleep quality.

As one of the most important and stressful steps towards gaining employment, job interviews are a social barrier that poses a unique challenge for individuals with Autism Spectrum Disorder (ASD). There are existing tools and in-person training available that show promise. However, these solutions are limited in the degree to which they can collect performance data, provide specific feedback, offer options to intelligently customize training, and simulate a variety of interviews in a realistic manner. InterViewR is a mixed reality job interview training simulator that addresses these shortcomings to reduce barriers in entering the workforce. The integration of Virtual Reality and wearable smart technology enables users to practice customizable interview simulations and receive both real-time and retrospective feedback. Findings are reported from a cognitive walkthrough (N=33) that highlights areas that need to be considered when designing such interview training platforms for individuals with autism.

Autistic college students face significant challenges in college settings and have a higher dropout rate than neurotypical college students. High physiological distress, depression, and anxiety are identified as critical challenges that contribute to this less than optimal college experience. In this paper, we leverage affordable mobile and wearable devices to collect large amounts of physiological and contextual data (biomarkers) and leverage a data-driven analysis approach for building stress prediction models. Such models can be used to provide real-time intervention for better stress management. We conducted a mixed-method study where we collected physiological and contextual data from 20 college students (10 neurotypical and 10 autistic). Our proposed data-driven analysis pipeline leverages an unsupervised representation learning technique with a semi-supervised label approximation method to predict the onset of stress based on biomarkers for autistic students, neurotypical students, and both populations with accuracies 69%, 72%, and 70%, respectively.

College students with autism have an estimated retention rate of 38.8% in the USA, which is attributed to challenges resulting from mental and behavioral problems, lack of adequate support from colleges, and a lack of awareness of their unique needs. We propose visualizations guided by the self-determination theory to help understand the experiences and needs of college students with autism and to empower them to manage their behavior. The visualizations are created based on physiological and contextual data collected from 20 college students (10 with autism, 10 neurotypical) using Fitbit and smartphone, and are incorporated in a dashboard to offer seamless access. We evaluated the effectiveness of these visualizations by conducting a lab study (N=18) and learned that such visualizations can highlight experiences of students with autism, can enhance awareness about individuals’ behaviors and habits, and can help to identify areas that can potentially improve the college experience.

Job interviews are uniquely challenging for individuals with Autism Spectrum Disorder. While digital interview training tools have shown promising results for improving vocational outcomes for individuals with Autism, existing solutions are largely limited in the degree to which they can simulate a realistic interview environment, collect performance data from the user, and provide actionable feedback for continued improvement. To address these shortcomings and understand how to create an effective training tool, we designed InterViewR, a simulation-based interview training system that combines virtual reality and wearable smart technology in an integrated platform. Our design emphasizes an immersive user experience for training effectiveness and utilizes physiological sensing to provide intelligent affective biofeedback. We report findings from a usability study (N=11) where participants evaluated InterViewR on feasibility, usability, and perceived usefulness.

Tobacco addiction is one of the most challenging behavioral health problems with successful cessation rates remaining in single digits. With increased availability of commercially available mobile and wearable devices we have the opportunity to infer lapse vulnerability and intervene in near real-time. In this work we present findings from a mixed-method study where we collected around 398.2 hours of physiological data from regular smokers in the field (N is 5) using commercial wearable devices along with contextual data using EMAs. We applied quantitative and qualitative analysis techniques to identify key factors contributing to smoking lapse and developed a statistical model capable of inferring lapse vulnerability from physiological and contextual signals collected from the natural environment. Our methodology and findings show promise in designing practical, near real-time smoking intervention systems that can be used by regular smokers in their everyday living situation.

Many neurodiverse individuals experience isolation in adult life with a significant percentage reporting never engaging in long term romantic relationships. Such social isolation can result in critical behavioral and mental health problems such as anxiety, addiction, and depression. Online socialization platforms can aid in relationship building as they are void of many complexities present in face-to-face communication. We conducted an exploratory study (N=10) with neurodiverse and neurotypical young adults to investigate how to design online socialization platforms that enables creating and maintaining lasting relationships, especially for neurodiverse individuals. We focused on understanding their experiences surrounding online socialization platforms, challenges in using them, and unmet needs. To elicit requirements, we designed a prototype socialization platform, Connection. Our findings reveal that there is indeed a great desire for such platforms among neurotypical and neurodiverse individuals.

Individuals with Autism Spectrum Disorder (ASD) experience difficulty in social interaction resulting in a lack of close social connections. Online social networking sites show promise in enabling individuals with ASD to make connections as the media of communication is often void of complexities present in real-life situations (e.g., real-time interpretation of complex social and emotional cues). We conducted a systematic content analysis of the Autism and Asperger's subreddits to understand user types, interaction patterns and trends, and user's experience, expectations, and unmet needs pertaining to these online communities. Our analysis revealed several interesting findings including the presence of 15 different topic areas, the tone and preference of the community, the difference in communication style between individuals with ASD and other related members (e.g., parents, therapists). Our analysis signified a need for an accessible and inclusive social networking site design that will cater to the specific needs of this community.

Smart technologies (wearable and mobile devices) show tremendous potential in the detection, diagnosis, and management of Autism Spectrum Disorder (ASD) by enabling continuous real-time data collection, identifying effective treatment strategies, and supporting intervention design and delivery. Though promising, effective utilization of smart technology in aiding ASD is still limited. We propose a set of implications to guide the design of ASD-support technology by analyzing 149 peer-reviewed articles focused on children with autism from ACM Digital Library, IEEE Xplore, and PubMed. Our analysis reveals that technology should facilitate real-time detection and identification of points-of-interest, adapt its behavior driven by the real-time affective state of the user, utilize familiar and unfamiliar features depending on user-context, and aid in revealing even minuscule progress made by children with autism. Our findings indicate that such technology should strive to blend-in with everyday objects. Moreover, gradual exposure and desensitization may facilitate successful adaptation of novel technology.

We present the design, implementation, and evaluation of Bangladesh Emergency Services (BES), a mobile application that provides 911-like services in Bangladesh. The involvement of A2I, a government agency, helped in the nationwide deployment of BES. To date, 120,382 unique users downloaded BES and 27,117 users are actively using it. Analysis of user ratings (N=3,788) and reviews (N=1,231) from Google Play Store, an online survey (N=137), and face-to-face interviews (N=10) revealed that users considered BES effective. Currently, BES is used to report aggressive behavior, street fight, harassment, fire-related incidents, and for locating nearby hospitals for medical emergencies. Our findings highlight the importance of trustworthiness, scalability of service, and iterative and participatory design for successful deployment of such services. We believe our research will inspire the design of mobile-based emergency applications in other developing countries that lack 911-like services.

The advent of mobile health (mHealth) technologies challenges the capabilities of current visualizations, interactive tools, and algorithms. We present Chronodes, an interactive system that unifies data mining and human-centric visualization techniques to support explorative analysis of longitudinal mHealth data. Chronodes extracts and visualizes frequent event sequences that reveal chronological patterns across multiple participant timelines of mHealth data. It then combines novel interaction and visualization techniques to enable multifocus event sequence analysis, which allows health researchers to interactively define, explore, and compare groups of participant behaviors using event sequence combinations. Through summarizing insights gained from a pilot study with 20 behavioral and biomedical health experts, we discuss Chronodes's efficacy and potential impact in the mHealth domain. Ultimately, we outline important open challenges in mHealth, and offer recommendations and design guidelines for future research.

In this paper, we present the design and implementation of a location-aware mobile-based emergency service for Bangladesh, a developing country, which lacks any central 911-like emergency service. Our goal was to investigate the feasibility and acceptability of such services that do not require building any new infrastructure or changing the existing infrastructure available in Bangladesh. To achieve our goal, we iteratively designed and deployed two location-aware mobile-based emergency services in Dhaka, the capital city of Bangladesh. These deployments provided a deep insight about user experience, acceptance, and sustainability issues. We learned that users considered these services effective, felt comfortable using these during emergencies, and expressed a need for integration of additional services. Our findings indicate that it is feasible to design a location-aware mobile-based emergency service in Bangladesh. We believe that our proposed design will help to provide a low-cost alternative to the central 911-services, especially for the developing countries.

Presents the introductory welcome message from the conference proceedings. May include the conference officers' congratulations to all involved with the conference event and publication of the proceedings record.

Smoking is one of the most challenging behavioral health problems. In the past, failed quit attempts have been attributed to factors including stress, presence of smoking cues, and negative affect – most of which were self-reported and prone to recall-bias. The first step in designing effective smoking cessation systems is to objectively identify factors that contribute to lapse. In our research, we collected physiological data utilizing wearable sensors from a four day pre-quit, post-quit study (N=55). We also collected self-report measures (n=3120), which offer rich contextual information about users' social, emotional, geographical, and physiological conditions. Analysis of collected data informed the design of MyQuitPal, a participant-centric cessation support system, which aims to assist individuals to better understand their smoking behavior. The design of MyQuitPal is also grounded on theories of long term health-behavior change. We believe our research advances understanding of complexities and opportunities surrounding the design of smoking cessation systems.

Management of daily stress can be greatly improved by delivering sensor-triggered just-in-time interventions (JITIs) on mobile devices. The success of such JITIs critically depends on being able to mine the time series of noisy sensor data to find the most opportune moments. In this paper, we propose a time series pattern mining method to detect significant stress episodes in a time series of discontinuous and rapidly varying stress data. We apply our model to 4 weeks of physiological, GPS, and activity data collected from 38 users in their natural environment to discover patterns of stress in real-life. We find that the duration of a prior stress episode predicts the duration of the next stress episode and stress in mornings and evenings is lower than during the day. We then analyze the relationship between stress and objectively rated disorder in the surrounding neighborhood and develop a model to predict stressful episodes.

Presents the introductory welcome message from the conference proceedings. May include the conference officers' congratulations to all involved with the conference event and publication of the proceedings record.

We investigate needs, challenges, and opportunities in visualizing time-series sensor data on stress to inform the design of just-in-time adaptive interventions (JITAIs). We identify seven key challenges: massive volume and variety of data, complexity in identifying stressors, scalability of space, multifaceted relationship between stress and time, a need for representation at multiple granularities, interperson variability, and limited understanding of JITAI design requirements due to its novelty. We propose four new visualizations based on one million minutes of sensor data (n=70). We evaluate our visualizations with stress researchers (n=6) to gain first insights into its usability and usefulness in JITAI design. Our results indicate that spatio-temporal visualizations help identify and explain between- and within-person variability in stress patterns and contextual visualizations enable decisions regarding the timing, content, and modality of intervention. Interestingly, a granular representation is considered informative but noise-prone; an abstract representation is the preferred starting point for designing JITAIs.

Mobile sensor data-to-knowledge (MD2K) was chosen as one of 11 Big Data Centers of Excellence by the National Institutes of Health, as part of its Big Data-to-Knowledge initiative. MD2K is developing innovative tools to streamline the collection, integration, management, visualization, analysis, and interpretation of health data generated by mobile and wearable sensors. The goal of the big data solutions being developed by MD2K is to reliably quantify physical, biological, behavioral, social, and environmental factors that contribute to health and disease risk. The research conducted by MD2K is targeted at improving health through early detection of adverse health events and by facilitating prevention. MD2K will make its tools, software, and training materials widely available and will also organize workshops and seminars to encourage their use by researchers and clinicians.

Driving is known to be a daily stressor. Measurement of driver's stress in real-time can enable better stress management by increasing self-awareness. Recent advances in sensing technology has made it feasible to continuously assess driver's stress in real-time, but it requires equipping the driver with these sensors and/or instrumenting the car. In this paper, we present "GStress", a model to estimate driver's stress using only smartphone GPS traces. The GStress model is developed and evaluated from data collected in a mobile health user study where 10 participants wore physiological sensors for 7 days ( for an average of 10.45 hours/day) in their natural environment. Each participant engaged in 10 or more driving episodes, resulting in a total of 37 hours of driving data. We find that major driving events such as stops, turns, and braking increase stress of the driver. We quantify their impact on stress and thus construct our GStress model by training a Generalized Linear Mixed Model (GLMM) on our data. We evaluate the applicability of GStress in predicting stress from GPS traces, and obtain a correlation of 0.72. By obviating any burden on the driver or the car, we believe, GStress can make driver's stress assessment ubiquitous.

Wearable wireless sensors for health monitoring are enabling the design and delivery of just-in-time interventions (JITI). Critical to the success of JITI is to time its delivery so that the user is available to be engaged. We take a first step in modeling users' availability by analyzing 2,064 hours of physiological sensor data and 2,717 self-reports collected from 30 participants in a week-long field study. We use delay in responding to a prompt to objectively measure availability. We compute 99 features and identify 30 as most discriminating to train a machine learning model for predicting availability. We find that location, affect, activity type, stress, time, and day of the week, play significant roles in predicting availability. We find that users are least available at work and during driving, and most available when walking outside. Our model finally achieves an accuracy of 74.7% in 10-fold cross-validation and 77.9% with leave-one-subject-out.

Reflection-in- and -on-action are key elements of creative design. While reflection-in-action has been often studied, reflection-on-action has received less attention. In this paper, we fill this gap by first reporting results from in-depth interviews (N=12) with practicing designers aimed at understanding activities related to reflection-on-action. We found that activities related to reflection-on-action are intentional, repetitive and frequent, design materials are the primary aids for reflective activities, and there is a strong need for better reflection support tools. To address this need, we designed, implemented and evaluated a novel interactive visualization tool -- ReflectionSpace. The tool uses file meta-data and naming conventions to map design materials to the appropriate design phase and context of use and places corresponding representations in a time- and activity-centric visualization that can be navigated at different levels of detail. User feedback from a comparative study reveals that such a representation of design process is more beneficial and preferred relative to the traditional file-centric approach for fostering reflection-on-action.

Reuse of existing presentation materials is prevalent among knowledge workers. However, finding the most appropriate material for reuse is challenging. Existing information management and search tools provide inadequate support for reuse due to their dependence on users' ability to effectively categorize, recall, and recognize existing materials. Based on our findings from an online survey and contextual interviews, we designed and implemented a slide-based contextual recommender, ConReP, for supporting reuse of presentation materials. ConReP utilizes a user-selected slide as a search-key, recommends materials based on similarity to the selected slide, and provides a local-context-based visual representation of the recommendations. Users input provides new insight into presentation reuse and reveals that slide-based search is more effective than keyword-based search, local-context-based visual representation helps in better recall and recognition, and shows the promise of this general approach of exploiting individual slides and local-context for better presentation reuse.

Reflection is an integral part of the creative design process. However, reflection is often considered as an enigmatic, un-orderly, and irregular process, leading to a lack of support from existing design and creativity support tools. Through a contextual inquiry (N=12) of designers' reflection practice, we found that reflection is predominantly a systematic, intentional, and repeatedly practiced activity. We also found that artifacts play an imperative role in supporting reflective process. We offer guidelines for the design of better reflection support tools.

Prior knowledge is a critical resource for design, especially when designers are striving to generate new ideas for complex problems. Systems that improve access to relevant prior knowledge and promote reuse can improve design efficiency and outcomes. Unfortunately, such systems have not been widely adopted indicating that user needs in this area have not been adequately understood. In this paper, we report the results of a contextual inquiry into the practices of and attitudes toward knowledge management and reuse during early design. The study consisted of interviews and surveys with professional designers in the creative domains. A novel aspect of our work is the focus on early design, which differs from but complements prior works' focus on knowledge reuse during later design and implementation phases. Our study yielded new findings and implications that, if applied, will help bring the benefits of knowledge management systems and reuse into early design activity.

To address the challenges posed by device capacity and capability, and also the nature of ad-hoc network of pervasive computing, a resource discovery model is needed that can resolve security and privacy issues with simple solutions. The use of complex algorithms and powerful fixed infrastructure is infeasible due to the volatile nature of pervasive environment and tiny pervasive devices. In this paper, we present a risk- aware trust based secure resource discovery model, RTSRD (risk-aware trust based secure resource discovery) for a truly pervasive environment. Our model is an adaptive hybrid one that allows both secure and non-secure discovery of services on adaptive trust. RTSRD also incorporates a risk model for sharing resources with unknown devices. Hence, the two contributions of this paper are: adaptive trust, and risk model for resource discovery in pervasive computing environments.

Requests for communication via mobile devices can be disruptive to the current task or social situation. To reduce the frequency of disruptive requests, one promising approach is to provide callers with cues of a receiver's context through an awareness display, allowing informed decisions of when to call. Existing displays typically provide cues based on what can be readily sensed, which may not match what is needed during the call decision process. In this paper, we report results of a four week diary study of mobile phone usage, where users recorded what context information they considered when making a call, and what information they wished others had considered when receiving a call. Our results were distilled into lessons that can be used to improve the design of awareness displays for mobile devices, e.g., show frequency of a receiver's recent communication and distance from a receiver to her phone. We discuss technologies that can enable cues indicated in these lessons to be realized within awareness displays, as well as discuss limitations of such displays and issues of privacy.

SSRD is a secure resource discovery model for devices running in a pervasive computing environment. SSRD is based on a lightweight trust model. SSRD+ is an extension of the existing SSRD model. In SSRD+, we enhance the trust model by adding dynamic trust relationship and also specifying behavioral characteristics that determine the level of trust among devices. We also add a risk model to address challenges posed by the pervasive and ad hoc nature of the network. These models work together to make the entire discovery process lightweight and secure. In this paper we present details of the trust and risk models. We illustrate the design and implementation of SSRD+ as a whole that optimally explores resources without degrading the performance of the devices while ensuring user security and privacy

A secure resource discovery model for devices running in pervasive computing environments has become significant. Due to the constrained nature of these devices, the need for a model that allows discovery and sharing of resources without putting much overhead on the operation of devices is needed. The dependency on fixed powerful machines for ensuring security is not desired and also not feasible. In this paper, we present a resource discovery model that provides security whenever needed without degrading the performance of the device. We also propose a trust based service-oriented adaptive security mechanism named SSRD (simple and secure resource discovery)

In this paper, primarily, we delineate the numerous challenges that can arise in mobile-business (m-business) to provide the highest degree of proliferation of pervasive advertisement. Later, we demonstrate the feasibility of PerAd-Service, an integral part of, MARKS (middleware adaptability for knowledge usability, resource discovery, and self-healing) to address those challenges.

Usability of pervasive computing features along with the detection of a meeting is very effective for the users. In this paper, we present the design of a smart meeting room (SMR) that not only detect the starting and ending of a meeting but also provide the support of participant's context information, knowledge usability and ephemeral group communication. Our approach will play a vital role for not only for meeting detection but other closely related applications of pervasive computing.