Social accounts of technological change make the flexibility and openness of interpretations the starting point of an argument against technological determinism. They suggest that technological change unfolds in the semantic domain, but they focus on the social processes around the interpretations of new technologies, and do not address the conceptual processes of change in interpretations. The dissertation presents an empirically grounded case study of the design process of an open-source online software platform based on the framework of distributed cognition to argue that the cognitive perspective is needed for understanding innovation in software, because it allows us to describe the reflexive and expansive contribution of conceptual processes to new software and the significance of professional epistemic practices in framing the direction of innovation. The framework of distributed cognition brings the social and cognitive perspectives together on account of its understanding of conceptual processes as distributed over time, among people, and between humans and artifacts. The dissertation argues that an evolving open-source software landscape became translated into the open-ended local design space of a new software project in a process of infrastructural implosion, and the design space prompted participants to outline and pursue epistemic strategies of sense-making and learning about the contexts of use. The result was a process of conceptual modeling, which resulted in a conceptually novel user interface. Prototyping professional practices of user-centered design lent directionality to this conceptual process in terms of a focus on individual activities with the user interface. Social approaches to software design under the broad umbrella of human-centered computing have been seeking to inform the design on the basis of empirical contributions about a social context. The analysis has shown that empirical engagement with the contexts of use followed from conceptual modeling, and concern about real world contexts was aligned with the user-centered direction that design was taking. I also point out a social-technical gap in the design process in connection with the repeated performance challenges that the platform was facing, and describe the possibility of a social-technical imagination.

Experiences of interpersonal racism persist as a prevalent reality for BIPOC (Black, Indigenous, People of Color) in the United States. One form of racism that often goes unnoticed is racial microaggressions. These are subtle acts of racism that leave victims questioning the intent of the aggressor. The line of offense is often unclear, as these acts are disguised through humor or seemingly harmless intentions. In this study, we analyze the language used in online racial microaggressions ("Acts") and compare it to personal narratives recounting experiences of such aggressions ("Recalls") by Black social media users. We curated a corpus of acts and recalls from social media discussions on platforms like Reddit and Tumblr. Additionally, we collaborated with Black participants in a workshop to hand-annotate and verify the corpus. Using natural language processing techniques and qualitative analysis, we examine the language underlying acts and recalls of racial microaggressions. Our goal is to understand the lexical patterns that differentiate the two in the context of racism in the U.S. Our findings indicate that neural language models can accurately classify acts and recalls, revealing contextual words that associate Blacks with objects that perpetuate negative stereotypes. We also observe overlapping linguistic signatures between acts and recalls, serving different purposes, which have implications for current challenges in social media content moderation systems. / Master of Science / Racial Microaggressions are expressions of human biases that are subtly disguised. The differences in language and themes used in instances of Racial Microaggressions ("Acts") and the discussions addressing them ("Recalls") on online communities have made it difficult for researchers to automatically quantify and extract these differences. In this study, we introduce a tool that can effectively distinguish acts and recalls of microaggressions. We utilize Natural Language Processing techniques to classify and identify key distinctions in language usage and themes. Additionally, we employ qualitative methods and engage in workshop discussions with Black participants to interpret the classification results. Our findings reveal common linguistic patterns between acts and recalls that serve opposing purposes. Acts tend to stereotype and degrade Black people, while recalls seek to portray their discomfort and seek validation for their experiences. These findings highlight why recalls are often considered toxic in online communities. This also represents an initial step towards creating a socio-technical system that safeguards the experiences of racial minority groups.

Computer Science is a complex field, and even experts do not always agree how the field should be defined. Though a moderate amount is known about how precollege students think about the field of CS, less is known about how CS majors' conceptions of the field develop during the undergraduate curriculum. Given the difficulty of understanding CS, how do students make educational decisions like what electives or specializations to pursue? This work presents a theory of student conceptions of CS, based on 37 interviews with students and student advisers and analyzed with a grounded theory approach. Students tend to have one of three main views about CS: CS as an academic discipline focused on the mathematical study of algorithms, CS as mostly about programming but also incorporating supporting subfields, and CS as a broad discipline with many different (programming and non-programming) subfields. I have also developed and piloted a survey instrument to determine how prevalent each kind of conception is in the undergraduate population. I also present a theory of student educational decisions in CS. Students do not usually have specific educational goals in CS and instead take an exploratory approach to their classes. Particularly enjoyable or unenjoyable classes cause them to narrow their educational focus. As a result, students do not reason very deeply about the CS content of their classes when they make educational decisions. This work makes three main contributions: the theory of student conceptions, the theory of student educational decisions, and the preliminary survey instrument for evaluating student conceptions. This work has applications in CS curriculum design as well as for future research in the CS education community.

The practice of health self-management offers behavioral and problem-solving strategies that can effectively promote responsibility for one's own wellbeing, improve one's health outcomes, and decrease the cost of health services. Personal informatics applications support health self-management by allowing their users to easily track personal health information, and to review the changes and patterns in this information. Over the course of the past several years, I have pursued a research agenda centered on understanding how personal health informatics applications can further support the strategies of health self-management--specifically those relating to goal-management and behavior change. I began by developing a flexible personal informatics tool, called Salud!, that I could use to observe real-world goal management and behavior change strategies, as well as use to evaluate new interfaces designed to assist in goal management. Unlike existing personal informatics tools, Salud! allows users to self-define the information that they will track, which allows tracking of highly personal and meaningful data that may not be possible to track given other tools. It also enables users to share their account data with facilitators (e.g. fitness grainers, nutritionists, etc.) who can provide input and feedback. Salud! was built on top of an infrastructure consisting of a stack of modular services that make it easier for others to develop and/or evaluate a variety of personal informatics applications. Several research teams used this infrastructure to develop and deploy a variety of custom projects. Informal analysis of their efforts showed an unmet need for data storage and visualization services for home- and health-based sensor data. In order to design a goal management support tool for Salud!, I first, I conducted a meta-analysis of relevant research literature to cull a set of proven goal management strategies. The key outcome of this work was an operationalization of Action Plans--goal management strategies that are effective at supporting behavior change. I then deployed Salud! in two fitness-related contexts to observe and understand the breadth of health-related behavior change and goal management practices. Findings from these deployments showed that personal informatics tools are most helpful to individuals who are able to articulate short-term, actionable goals, and who are able to integrate self-tracking into their daily activities. The literature meta-analysis and the two Salud! deployments provided formative requirements for a goal management interaction that would both incorporate effective goal management strategies and support the breadth of real-world goals. I developed a model of the goal management process as the framework for such an interaction. This model enables goals to be represented, evaluated, and visualized, based on a wide range of user objectives and data collection strategies. Using this model, I was able to develop a set of interactions that allow users of Salud! to manage their personal goals within the application. The generalized goal management model shows the inherent difficulty in supporting open-ended, highly personalized goal management. To function generically, Salud! requires facilitator input to correctly process goals and meaningfully classify their attributes. However, for specific goals represented by specific data collection strategies, it is possible to fully- or semi-automate the goal management process. I ran a large-scale evaluation of Salud! with the goal management interaction to evaluate the effectiveness of a fully-automated goal management interaction. The evaluation consisted of a common health self-management intervention: a simple fitness program to increase participants' daily step count. The results of this evaluation suggest that the goal management interaction may improve the rate of goal realization among users who are initially less active and less confident in their ability to succeed. Additionally, this evaluation showed that, while it can significantly increase participants' step count, a fully automated fitness program is not as effective as traditional, instructor-led fitness programs. However, it is much easier to administer and much less resource intensive, showing that it can be utilized to rapidly evaluate concrete goal management strategies.

In order to support people's everyday health and wellness goals, health practitioners and organizations are embracing a more holistic approach to medicine---supporting patients both as individuals and members of their families and communities, and meeting people where they are: at home, work, and school. This 'everyday' approach to health has been enabled by new technologies, both dedicated-devices and services designed specifically for health sensing and feedback -- and multipurpose --such as smartphones and broadband-connected computers. Our physical relationship with computing has also become more intimate, and personal health devices can now track and report an unprecedented amount of information about our bodies, following their users around to an extent no doctor, coach or dietitian ever could. But we still have much to learn about how pervasive health devices can actually help promote the adoption of new health practices in daily life. Once they're `in the wild,' such devices interact with their users, but also the physical, social and political worlds in which those users live. These external factors---such as the walkablity of a person's neighborhood or the social acceptability of exercise and fitness activities---play a significant role in people's ability to change their health behaviors and sustain that change. Specifically, social theories of behavior change suggest that peer support may be critical in changing health attitudes and behaviors. These theories---Social Support Theory, Social Cognitive Theory and Social Comparison Theory among them---offer both larger frameworks for understanding the social influences of health behavior change and specific mechanisms by which that behavior change could be supported through interpersonal interaction. However, we are only beginning to understand the role that pervasive health technologies can play in supporting and mediating social interaction to motivate people's exploration and adoption of healthy behaviors. In this dissertation I seek to better understand how social computing technologies can help people help each other live healthier lives. I ground my research in a participant-led investigation of a specific population and condition: adolescents and obesity prevention. I want to understand how social behavior change theories from psychology and sociology apply to pervasive social health technology. Which mechanisms work and why? How does introducing a pervasive social health system into a community affect individuals' behaviors and attitudes towards their health? Finally, I want to contribute back to those theories, testing their effectiveness in novel technologically mediated situations. Adolescent obesity is a particularly salient domain in which to study these issues. In the last 30 years, adolescent obesity rates in the US alone have tripled, and although they have leveled off in recent years they remain elevated compared to historical norms. Habits formed during adolescence can have lifelong effects, and health promotion research shows that even the simple act of walking more each day has lasting benefits. Everyday health and fitness research in HCI has generally focused on social comparison and "gamified" competition. This is especially true in studies focused on adolescents and teens. However, both theory from social psychology and evidence from the health promotion community suggest that these direct egocentric models of behavior change may be limited in scope: they may only work for certain kinds of people, and their effects may be short-lived once the competitive framework is removed. I see an opportunity for a different approach: social tools for everyday adolescent health. These systems, embedded in existing school and community practices, can leverage scalable, non-competitive social interaction to catalyze positive perceptions of physical activity and social support for fitness, while remaining grounded in the local environment. Over the last several years I have completed a series of field engagements with middle school students in the Atlanta area. I have focused on students in a majority-minority low-income community in the Atlanta metropolitan area facing above-average adult obesity levels, and I have involved the students as informants throughout the design process. In this dissertation, I report findings based on a series of participatory design-based formative explorations; the iterative design of a pedometer-based pervasive health system to test these theories in practice; and the deployment of this system---StepStream---in three configurations: a prototype deployment, a `self-tracking' deployment, and a `social' deployment. In this dissertation, I test the following thesis: A school-based social fitness approach to everyday adolescent health can positively influence offline health behaviors in real-world settings. Furthermore, a noncompetitive social fitness system can perform comparably in attitude and behavior change to more competitive or direct-comparison systems, especially for those most in need of behavior change}. I make the following contributions: (1) The identification of tensions and priorities for the design of everyday health systems for adolescents; (2) A design overview of StepStream, a social tool for everyday adolescent health; (3) A description of StepStream's deployment from a socio-technical perspective, describing the intervention as a school-based pervasive computing system; (4) An empirical study of a noncompetitive awareness system for physical activity; (5) A comparison of this system in two configurations in two different middle schools; (6) An analysis of observational learning and collective efficacy in a pervasive health system.

Studies assessing novice programming proficiency have often found that many students coming out of introductory-level programming courses still struggle with programming. To address this, some researchers have attempted to find and develop ways to better help students succeed in learning to program. This dissertation research contributes to this area by studying the programming processes of students trained through a specific program design curriculum, How to Design Programs (HTDP). HTDP is an introductory-level curriculum for teaching program design that teaches a unique systematic process called the design recipe that leverages the structure of input data to design programs. The design recipe explicitly scaffolds learners through the program design process by asking students to produce intermediate artifacts that represent a given problem in different ways up to a program solution to the problem. Although HTDP is used in several higher-education institutions and some K-12 programs, how HTDP-trained students design programs towards problems, particularly ones with multiple task-components, has not been thoroughly studied. The overarching goal of this dissertation is to gain an understanding and insight into how students use the techniques put forth by the design recipe towards designing solutions for programming problems. I conducted a series of exploratory user studies with HTDP-trained student cohorts from HTDP course instances across two different universities to collect and analyze students’ programming process data in situ. I synthesized findings from each study towards an overall conceptual framework, which serves as a data-grounded theory that captures several facets of HTDP-trained students’ program design process. The main contribution of this work is this theory, which describes: (1) the program design-related skills that students used and the levels of complexity at which they applied these skills, (2) how students’ use of design skills evolve during a course, (3) the interactions between program design skills and course contexts that influenced how students applied their skills, and (4) the programming process patterns by which students approached the programming problems we gave and how these approaches relate towards students’ success with the problems. Using insights from the theory, I describe recommendations toward pedagogical practices for teaching HTDP-based courses, as well as broader reflections towards teaching introductory CS.

No description available.

The opportunity for online engagement increases possible exposure to potentially risky behaviors for teens, which may have significant negative consequences (Hair et al., 2009). Effective family communication about online safety can help reduce the risky adolescent behavior and limit the consequences after it occurs. This paper contributes a theory of communication factors that positively influence teen and parent perception of communication about online safety and provides design implications based on those findings. Previous work identified gaps in family communication, however, this study seeks to empirically identify factors that would close the communication gap from the perspective of both teens and parents. I analyzed data from a survey of 215 teen-parent pairs with a cross-sectional design and examined the factors that contribute to increased family communication about online safety. For parents, active mediation, technical monitoring of their teens' devices, and a perceived positive affect schedule of the teen were associated with higher levels of family communication. Our results were similar for teens, except that parental monitoring and the teen's online safety concern were also positively associated with increased family communication, while restrictive mediation was associated with lower levels of family communication. A key implication of these findings is that teens do not want to be left alone, but desire active mediation and monitoring. Teens do not want technological based restriction. As the first study to explore specific mechanisms which may improve family communication between parents and teens regarding online safety, I am able to recommend design solutions that allow teens an active role in their own online safety and facilitate effective family communication from the perspectives of both parties by assisting parents to adopt active mediation techniques rather than developing technologies that encourage restrictive parenting. Many designs for parents and teen monitoring historically support a restrictive approach (P. Wisniewski et al., 2017). Rather than focus on parental control applications, I advance both analytical support for a more nuanced theoretical and practical applications.

Evaluating User Experiences early in a development process can save both time and money by pro-actively mapping out user needs and behavior patterns. However, since most well-known UX-evaluation methods applies during or after user interaction, due to the “second-order” design problem of experiences being a byproduct of interaction, there is a desire within HCI for more early-stage UX-evaluation methods that could be applied to concept stages as well. This paper investigates the experiential evaluation of a storyboard and narrative through the Anticipated eXperience Method (AxE) and discusses how it compares to a re-iterated high-fidelity prototype created in Figma. The process of this study is described out of the context it has been executed in, which is together with the company Scania AB at their cabin production facility in Oskarshamn, Sweden. The study explores what insights can be found from evaluating anticipated user experiences in early concept development and how these insights can apply towards further development of a touchpad user Interface. The underlying foundation of this study has followed the approach of a design-inclusive UX-research project, which heavily incorporates design activities into the process of conducting research. Thus, the storyboard, narrative and interactive Figma prototype have been created along the process and takes center stage in the investigation of experiential evaluation at early stages of interactive product development.

As robots have become more pervasive in our everyday life, social aspects of robots have attracted researchers' attention. Because emotions play a key role in social interactions, research has been conducted on conveying emotions via speech, whereas little research has focused on the effects of non-speech sounds on users' robot perception. We conducted a within-subjects exploratory study with 40 young adults to investigate the effects of non-speech sounds (regular voice, characterized voice, musical sound, and no sound) and basic emotions (anger, fear, happiness, sadness, and surprise) on user perception. While listening to the fairytale with the participant, a humanoid robot (Pepper) responded to the story with a recorded emotional sound with a gesture. Participants showed significantly higher emotion recognition accuracy from the regular voice than from other sounds. The confusion matrix showed that happiness and sadness had the highest emotion recognition accuracy, which aligns with the previous research. Regular voice also induced higher trust, naturalness, and preference compared to other sounds. Interestingly, musical sound mostly showed lower perceptions than no sound. A further exploratory study was conducted with an additional 49 young people to investigate the effect of regular non-verbal voices (female voices and male voices) and basic emotions (happiness, sadness, anger, and relief) on user perception. We also further explored the impact of participants' gender on emotion and social perception toward robot Pepper. While listening to a fairy tale with the participants, a humanoid robot (Pepper) responded to the story with gestures and emotional voices. Participants showed significantly higher emotion recognition accuracy and social perception from the voice + Gesture condition than Gesture only conditions. The confusion matrix showed that happiness and sadness had the highest emotion recognition accuracy, which aligns with the previous research. Interestingly, participants felt more discomfort and anthropomorphism in male voices compared to female voices. Male participants were more likely to feel uncomfortable when interacting with Pepper. In contrast, female participants were more likely to feel warm. However, the gender of the robot voice or the gender of the participant did not affect the accuracy of emotion recognition. Results are discussed with social robot design guidelines for emotional cues and future research directions. / Master of Science / As robots increasingly appear in people's lives as functional assistants or for entertainment, there are more and more scenarios in which people interact with robots. More research on human-robot interaction is being proposed to help develop more natural ways of interaction. Our study focuses on the effects of emotions conveyed by a humanoid robot's non-speech sounds on people's perception about the robot and its emotions. The results of our experiments show that the accuracy of emotion recognition of regular voices is significantly higher than that of music and robot-like voices and elicits higher trust, naturalness, and preference. The gender of the robot's voice or the gender of the participant did not affect the accuracy of emotion recognition. People are now not inclined to traditional stereotypes of robotic voices (e.g., like old movies), and expressing emotions with music and gestures mostly shows a lower perception. Happiness and sadness were identified with the highest accuracy among the emotions we studied. Participants felt more discomfort and human-likeness in the male voices than in female voices. Male participants were more likely to feel uncomfortable when interacting with the humanoid robot, while female participants were more likely to feel warm. Our study discusses design guidelines and future research directions for emotional cues in social robots.