Transparency in AI is crucial for fostering user trust and acceptance, yet achieving it through explanations presents significant design challenges, particularly regarding how much detail to provide. For example, in-depth explanations can convey accurate and comprehensive information, but they also risk overwhelming users. This paper considers this important design tradeoff in the context of training dataset explanations, which describe the data used to train AI systems and differ from most model-centric explanations in terms of what and how much information they communicate. Specifically, we investigate how information depth in training dataset explanations and the use of Progressive Disclosure impact users’ understanding of an AI system (assessed via their critiques of the system), their system assessments, and their cognitive load. Findings from a study with 32 participants show advantages to providing users with comprehensive information on training datasets. Detailed explanations not only enhanced perceived trust, fairness, and understanding, but were also preferred by participants despite the increased cognitive load. While Progressive Disclosure did not effectively mitigate cognitive load, it improved users’ perception of learning. These findings suggest that effective transparency does not come from minimizing detail, but from embracing it, as participants consistently valued clarity and completeness over brevity, even at the cost of higher cognitive load.

The widespread use of digital devices among children and teenagers has raised concerns about overuse, particularly for early adolescents, who have unique developmental needs and engage with technology more frequently than other age groups. A challenge for designers and researchers interested in contributing solutions is a lack of synthesized design guidelines and characterization of the current state-of-the-art. In this paper, we present a systematic scoping review of academic literature and an analysis of 47 apps, providing a comprehensive characterization of existing tech-mediated solutions for early adolescents. Our review covers literature from two major databases (ACM DL and IEEE Xplore) spanning the past 10 years (2014-May 2024), following the scope of prior similar reviews. The app analysis includes Google Play and Apple App Store apps with features targeting tech overuse, excluding general-purpose apps (e.g., social media, games) and apps without a free trial version. Our findings highlight researchers' design recommendations for promoting tech disengagement in this demographic (e.g., supporting collaborative rule-setting and self-monitoring, maintaining privacy, addressing diverse user needs), while revealing that existing apps tend to prioritize restrictive measures, overlooking self-regulation and active parental engagement. Our findings also identify areas of agreement and potential misalignments between current digital interventions and prior research on target users' preferences.

Live streaming platforms have become established communication channels where streamers and viewers can communicate through the chatbox. While there are numerous benefits to streamer-viewer interactions, managing messages can become challenging for streamers, especially during high-activity periods. In this paper, we investigate a strategy to supporting streamers that involves providing them with real-time summaries of chat messages. We investigate the feasibility of this approach through a prototype called the Stream Assistant, which provides automated poll summaries, a Word Cloud depiction of chat messages, and an overview of popular Emotes in the chat. We explore the potential utility of this approach in a multi-session study with 10 streamers, where we interviewed participants on their current chat management approaches and perceptions of the Stream Assistant after using it during one of their live streams. Our findings highlight the role of the chat in boosting engagement and audience growth and illustrate how streamers from different domains balance the chat with their activity. Our findings also indicate that many participants were enthusiastic about the Stream Assistant’s lightweight polling, whereas the utility of the other features might depend on the pace of the chat and the intensity of the streamer’s activities.

Training data has a profound impact on the performance of Machine Learning (ML) systems. To help interested parties develop informed trust and appropriate reliance, recent work has proposed providing users with training dataset explanations. While this prior work showed promising subjective impacts of increasing system transparency in this manner, little is known as to how users might use this information to critique a system. In this work, we investigate how two presentation styles for such explanations (a narrative-driven Data Story and a Q&A format) support users’ critique of an automated hiring system. Findings from a between-subjects study with 39 participants provide insights into the aspects of training dataset explanations that participants leveraged in their critiques. Our findings also show that presentation style can impact critique emphasis, critique accuracy, and subjective impressions of explanation utility.

Recruiting participants is a pivotal yet challenging aspect of Human-Computer Interaction (HCI) research. In this position paper, we reflect on our experiences recruiting participants for multiple user studies within a human-centered Explainable AI project. Emphasizing the importance of participant diversity in terms of their AI knowledge and experiences, we discuss challenges, trade-offs, and strategies associated with recruiting for both in-person and remote participation.

Collaborative learning has been suggested as a promising approach to help older adults learn new technology, however, its effectiveness has been understudied in the context of feature-rich applications. We conducted an observational study with the aim of identifying aspects of collaborative learning, including characteristics of collaborative partners, that impact older adults' exploratory learning behaviour in feature-rich software. We recruited 22 participants (6 younger adults and 16 older adults) who formed 5 same-age and 6 mixed-age dyads. These dyads worked together remotely to explore a feature-rich application, which was new to them. We classified dyadic interactions into four different collaboration dynamics characterized by distinct attributes. We discovered that effective communication and the ability to navigate the software independently enabled a successful collaboration dynamic that empowered learners. We showed that trust between partners enabled effective communication and we observed that the existing relationship between partners strongly impacted their communication patterns. The more complicated study tasks required participants to validate the correctness of their work and this validation was particularly difficult for some novice older adults who did not benefit from transfer learning and struggled with navigation issues.

Children's excessive use of technology is a growing concern, and despite taking various measures, parents often find it difficult to limit their children's device use. Limiting tech usage can be especially challenging with early adolescents as they start to develop a sense of autonomy. While numerous tech-based mediation solutions exist, in this paper, we aim to learn from early adolescents directly by having them contribute to co-design activities. Through a multi-session, group-based, online co-design study with 21 early adolescents (ages 11-14), we explore their perceptions towards tech overuse and what types of solutions they propose to help with disengagement. Findings from these co-design sessions contribute insights into how the participants conceptualized the problem of tech overuse, how they envisioned appropriate mediation strategies, and important design considerations. We also reflect on our study methods, which encouraged active participation from our participants and facilitated valuable contributions during the online co-design sessions.

Online Question and Answer communities (Q&As) are popular spaces for learning and sharing knowledge. However, prior research suggests that Q&As may not be appealing to and inclusive of men and women, with absent social considerations listed as a potential contributing factor. We investigate how additional community presence information can affect users' perceptions of and engagement with a Q&A for graphic design software. Through a 10-day task-based field study with 30 participants (14 women, 14 men, 2 non-binary), we uncover how community presence information can humanize the Q&A and play a role in promoting an inclusive environment. On the other hand, some participants question if community presence information belongs in a Q&A and describe some privacy implications. The women in our sample also talked about the importance of diverse community demographics, while we did not observe this sentiment expressed by the men. Our findings contribute an understanding of how users perceive the role of community presence information within a Q&A. We also discuss how this information might impact women's future participation and engagement.

Training datasets fundamentally impact the performance of machine learning (ML) systems. Any biases introduced during training (implicit or explicit) are often reflected in the system's behaviors leading to questions about fairness and loss of trust in the system. Yet, information on training data is rarely communicated to stakeholders. In this work, we explore the concept of data-centric explanations for ML systems that describe the training data to end-users. Through a formative study, we investigate the potential utility of such an approach, including the information about training data that participants find most compelling. In a second study, we investigate reactions to our explanations across four different system scenarios. Our results suggest that data-centric explanations have the potential to impact how users judge the trustworthiness of a system and to assist users in assessing fairness. We discuss the implications of our findings for designing explanations to support users’ perceptions of ML systems.

Citizen science game developers often choose to transform or simplify the scientific problem of interest to make the game accessible to a wider audience, which is critical for collecting enough data. A potential drawback of this design decision is the disconnect that it creates between the players and the science. We are interested in studying ways to re-establish that connection and give back to the players in the form of educational content embedded in those games. Here, we briefly describe a preliminary study focused on analyzing different formats (text, cartoon or video), then we mainly focus on studying ways to present educational material (full versus gradual access) and evaluating their effects on players in a citizen science game. The results indicate that although there is no clear consensus on which format is more appropriate, providing gradual access to the content as a reward for completing levels can increase the engagement, the retention and the performance of players.

People increasingly access cross-device applications from their smartphones while on the go. Yet, they do not fully use the mobile versions for complex tasks, preferring the desktop version of the same application. We conducted a survey (N=77) to identify challenges when switching back and forth between devices. We discovered significant cross-device learnability issues, including that users often find exploring the mobile version frustrating, which leads to prematurely giving up on using the mobile version. Based on the findings, we created four design concepts as video prototypes to explore how to support cross-device learnability. The concepts vary in four key dimensions: the device involved, automation, temporality, and learning approach. Interviews (N=20) probing the design concepts identified individual differences affecting cross-device learning preferences, and that users are more motivated to use cross-device applications when offered the right cross-device learnability support. We conclude with future design directions for supporting seamless cross-device learnability.

Augmented Reality (AR) and Virtual Reality (VR) devices are becoming easier to access and use, but the barrier to entry for creating AR/VR applications remains high. Although the recent spike in HCI research on novel AR/VR tools is promising, we lack insights into how AR/VR creators use today's state-of-the-art authoring tools as well as the types of challenges that they face. We interviewed 21 AR/VR creators, which we grouped into hobbyists, domain experts, and professional designers. Despite having a variety of motivations and skillsets, they described similar challenges in designing and building AR/VR applications. We synthesize 8 key barriers that AR/VR creators face nowadays, starting from prototyping the initial experiences to dealing with "the many unknowns" during implementation, to facing difficulties in testing applications. Based on our analysis, we discuss the importance of considering end-user developers as a growing population of AR/VR creators, how we can build learning opportunities into AR/VR tools, and the need for building AR/VR toolchains that integrate debugging and testing.

Feature-rich applications such as word processors and spreadsheets are not only being used by adults but increasingly by children and older adults as well. Learning these applications is challenging as they offer hundreds of commands throughout the interface. We investigate how newcomers from different age groups explore the user interface of a feature-rich application to determine, locate, and use relevant features. We conducted an in-lab observational study with 10 children (10-12), 10 adults (20-35) and 10 older adults (60-75) who were first-time users of Microsoft OneNote. Our results illustrate key exploration differences across age groups, including that children were careful and performed as efficiently as the adults, whereas older adults spent a longer time and repeated sequences of failed selections. Further, their exploration style was negatively influenced by their past knowledge of similar applications. We discuss design interventions to accommodate these exploration differences and to improve software onboarding for newcomers.

Design professionals increasingly need to learn and use multiple feature-rich software to design complex artifacts in a variety of domains. In this work, we aimed to understand how design professionals learn new software features and seek help in today's modern workplace of interconnected spaces and increased access to shared online learning resources. We conducted 20 interviews with design professionals working in different disciplines such as architecture and planning, construction, structural design, and broadcast media, who used complex feature-rich applications on a daily basis. We found that despite the wide availability of learning resources on the web and decades of research dedicated to workplace knowledge management tools, design professionals still rely heavily on interpersonal help-seeking. Furthermore, while companies have devised a variety of technological channels to answer employees' help requests, such intermediate tools were often abandoned, and help-seeking conversations shifted to face-to-face learning episodes. Our findings point to design opportunities and shortcomings in the design of remote communication tools when seeking help for feature-rich software applications.

Online step-by-step text and video tutorials play an integral role in learning feature-rich software applications. However, when searching, users can find it difficult to assess whether a tutorial is designed for their level of software expertise. Novice users can struggle when a tutorial is out of their reach, whereas more advanced users can end up wasting time with overly simple, first-principles instruction. To assist users in selecting tutorials, we investigate the feasibility of using machine-learning techniques to automatically assess a tutorial's difficulty. Using Photoshop as our primary testbed, we develop a set of distinguishable tutorial features, and use these features to train a classifier that can label a tutorial as either Beginner or Advanced with 85% accuracy. To illustrate a potential application, we developed a tutorial browsing interface called TutVis. Our initial user evaluation provides insight into TutVis's ability to support users in a range of tutorial selection scenarios.

Question and answer (Q&A) sites can capture a range of user perspectives on using complex, feature-rich software. Little is known, however, on who is contributing to the sites. We look at contribution diversity from the perspective of gender in a domain with near gender parity: graphic design. Through content analysis of 330 answers from two popular Q&A sites and semi-structured interviews with 24 graphic designers, we examine who is contributing, what content, how the community shows appreciation towards their answers, and perceived motivations and barriers to participation. We find that despite gender balance in the field, women contribute far less frequently than men. We also see gender differences in contribution styles and user appreciation. Our interviews shed further light on how Q&A community cultures might be impacting men and women differently and how design choices made by the sites? developers might be exacerbating these differences. We suggest implications for design for improving gender inclusivity.

For most modern feature-rich software, considerable external help and learning resources are available on the web (e.g., documentation, tutorials, videos, Q&A forums). But, how do users new to an application discover and make use of such resources? We conducted in-lab and diary studies with 26 software newcomers from a variety of different backgrounds who were all using Fusion 360, a 3D modeling application, for the first time. Our results illustrate newcomers' diverse needs, perceptions, and help-seeking behaviors. We found a number of distinctions in how technical and non-technical users approached help-seeking, including: when and how they initiated the help-seeking process, their struggles in recognizing relevant help, the degree to which they made coordinated use of the application and different resources, and in how they perceived the utility of different help formats. We discuss implications for moving beyond "one-size-fits-all" help resources towards more structured, personalized, and curated help and learning materials.

Graphic designers often use the Web to collect images to use as inspiration and references for their work. Their resulting collections of images, however, typically do not retain important aspects of their visual research, such as their thought process when searching and all explored design avenues. Guided by an exploratory study with 14 expert graphic designers, we developed Prism - a system that supports a graphic designer's visual research on the Internet by automatically capturing all inspected images and annotating them with the designer's search trails. We evaluated Prism through a two-week field study with 11 expert designers. Our findings suggest that Prism's capture and display capabilities helped the designers to reify their design thinking, to better reflect on and compare alternative design ideas, and to collaborate with their colleagues and clients.

The use of crowdsourcing platforms for data collection in HCI research is attractive in their ability to provide rapid access to large and diverse participant samples. As a result, several researchers have conducted studies investigating the similarities and differences between data collected through crowdsourcing and more traditional, laboratory-style data collection. We add to this body of research by examining the feasibility of conducting social acceptability studies via crowdsourcing. Social acceptability can be a key determinant for the early adoption of emerging technologies, and as such, we focus our investigation on social acceptability for Head-Worn Display (HWD) input modalities. Our results indicate that data collected via a crowdsourced experiment and a laboratory-style setting did not differ at a statistically significant level. These results provide initial support for crowdsourcing platforms as viable options for conducting social acceptability research.

We investigated interactive agents using subliminal priming - the act of exposing a person to stimuli that they may not consciously notice, but are still processed subliminally in their mind - in an attempt to shape a person's mood and behavior. We present an overview of the psychology of subliminal priming from the perspective of how it applies to human-agent interaction, including a discussion of the potential ethical and practical implications. We further present the results from two exploratory studies (one in-lab, one crowdsourced) that present potential subliminal-priming interfaces. Our results suggest that subliminal priming may impact how participants perceive an agent and how much they enjoy a task, but we failed to find any effect of priming on participant mood or agent persuasiveness. This work aims to raise awareness of the dangers of subliminal methods of priming and contributes to the discussion on the ethics of social agents.

Instructors of 3D design workshops for children face many challenges, including maintaining awareness of students' progress, helping students who need additional attention, and creating a fun experience while still achieving learning goals. To help address these challenges, we developed Maestro, a workshop orchestration system that visualizes students' progress, automatically detects and draws attention to common challenges faced by students, and provides mechanisms to address common student challenges as they occur. We present the design of Maestro, and the results of a case-study evaluation with an experienced facilitator and 13 children. The facilitator appreciated Maestro's real-time indications of which students were successfully following her tutorial demonstration, and recognized the system's potential to "extend her reach" while helping struggling students. Participant interaction data from the study provided support for our follow-along detection algorithm, and the capability to remind students to use 3D navigation.

The popularity of head-worn displays (HWD) technologies such as Virtual Reality (VR) and Augmented Reality (AR) headsets is growing rapidly. To predict their commercial success, it is essential to understand the acceptability of these new technologies, along with new methods to interact with them. In this vein, the evaluation of social acceptability of interactions with these technologies has received significant attention, particularly from the performer's (i.e., user's) viewpoint. However, little work has considered social acceptability concerns from observers' (i.e., spectators') perspective. Although HWDs are designed to be personal devices, interacting with their interfaces are often quite noticeable, making them an ideal platform to contrast performer and observer perspectives on social acceptability. Through two studies, this paper contrasts performers' and observers' perspectives of social acceptability interactions with HWDs under different social contexts. Results indicate similarities as well as differences, in acceptability, and advocate for the importance of including both perspectives when exploring social acceptability of emerging technologies. We provide guidelines for understanding social acceptability specifically from the observers' perspective, thus complementing our current practices used for understanding the acceptability of interacting with these devices.

The rise of open platforms and public APIs has enabled more users of customizable software to customize by installing plugins or add-ons that are created and shared by others. Despite the prevalence of online customization sharing, we know little about how and why online customizations are shared. Through interviews with 20 users of four diverse systems who have extensive experience with sharing their customizations, we reveal the concept of customization sharing ecosystems. These ecosystems include multiple components for hosting customizations, discussing, and managing them; the ecosystems are sustained through users acting in a diverse set of roles (e.g., sharers, re-users, reviewers, problem reporters, requesters, helpers, publicizers, and packers). Our interviews also reveal motivations for creating and sharing customizations online which overlap considerably with those in open source software. Our findings highlight tradeoffs and design considerations in these ecosystems.

There is always something new to learn about feature-rich software, even for experienced users. This paper focuses on a specific type of learning activity that we refer to as ad libitum exploration. Based on an interview study with 11 experienced software users, we define ad libitum exploration as the process of routinely seeking new software knowledge, without necessarily having a specific problem to solve. To support this activity, we designed Switter, an alternative Twitter client embedded in a replica of Photoshop's user interface. Given a tweet referencing a tutorial, Switter highlights the interface elements mentioned in the tutorial in the interface replica. Switter also allows users to filter tweets by clicking tools in the interface replica. Through a weeklong field study with nine Photoshop experts, we found that Switter supports a range of software learning objectives, from focused exploration targeting known weaknesses, to the discovery of novel command combinations.

Feature-rich software applications offer users hundreds of commands, yet most people use only a very small fraction of the available command set. Command recommenders aim to increase awareness of an application's capabilities by generating personalized recommendations for new commands. A primary distinguishing characteristic of these recommenders concerns the manner in which they determine command relevance. Social approaches do so by analyzing community usage logs, whereas, task-based approaches mine web documentation for logical command clusters. Through a qualitative study with sixteen participants, in this work we explored user attitudes towards these different approaches and the supplemental information they enable.

Family communication and technology designed to support it is a widely studied topic. However, most research that focuses on family communication in North America tends to assume high degrees of connectivity and Internet access. We present a study of family communication practices in rural and northern areas of Manitoba, Canada where Internet connectivity is intermittent or severely limited in some communities. Our results show the ways in which individuals stay connected with outside relatives can be hampered by communication infrastructure challenges. In particular, these challenges can dictate how, where and how often conversations with loved ones take place. Our results also indicate that these experiences, many of which are negative, can create lasting impressions that may be difficult to alter as infrastructure improves. This suggests opportunities for designing family communication technologies for outdoor locations with better connectivity, scheduling communication during times with better connectivity, and combating social isolation.

This paper presents an original scenario design specifically created for exploring gender-related issues surrounding collaborative human-robot teams on assembly lines. Our methodology is grounded squarely in the need for increased gender work in human-robot interaction. As with most research in social human-robot interaction, investigating and exploring gender issues relies heavily on an evaluation methodology and scenario that aims to maximize ecological validity, so that the lab results can generalize to a real-world social scenario. In this paper, we present our discussion on study elements required for ecological validity in our context, present an original study design that meets these criteria, and present initial pilot results that reflect on our approach and study design.

Prior work on command recommendations for feature-rich software has relied on data supplied by a large community of users to generate personalized recommendations. In this work, we explore the feasibility of using an alternative data source: web documentation. Specifically, our approach uses QF-Graphs, a previously proposed technique that maps higher-level tasks (i.e., search queries) to commands referenced in online documentation. Our approach uses these command-to-task mappings as an automatically generated plan library, enabling our prototype system to make personalized recommendations for task-relevant commands. Through both offline and online evaluations, we explore potential benefits and drawbacks of this approach.

Identifying a target command can be difficult and time-consuming when the user is unfamiliar with a software system. One technique for assisting command identification is to provide a subset interface that contains only a limited set of the system's capabilities. We examine the design of subset interfaces, showing that subsets can be presented separately to the full user interface (UI) (e.g. in a palette) or in place, with in-place methods using either static or dynamic methods to identify the subset. We introduce the StencilMap and EphemeralMap as in-place subset UIs that, respectively, use static and dynamic highlighting. Both StencilMaps and EphemeralMaps make all of an application's commands concurrently available for selection within a grid. To highlight subset items StencilMaps use a static dark semi-transparent ‘stencil’ overlay to de-emphasise all but the subset items; EphemeralMaps, in contrast, use a short delay, with subset items shown immediately, and other items gradually faded in. A first experiment compares user performance with the in-place presentation of StencilMaps against that of the separate presentation of a subset palette. Results confirm the predicted spatial memory benefits for StencilMaps. A second experiment analyses the performance impact of three approaches to highlighting: none, static highlighting in StencilMaps, and dynamic highlighting in EphemeralMaps. Results show an interesting trade-off – while highlighting can offer benefits in assisting rapid target identification (particularly when the user is unfamiliar with the interface layout), there can also be longer-term performance benefits when highlighting is absent because the increased difficulty of visual search promotes the use and formation of spatial memory.

User comments posted to popular online tutorials constitute a rich additional source of information for readers, yet current designs for displaying user comments on tutorial webpages do little to support their use. Instead, comments are separated from the tutorial content they reference and tend to be ordered according to post date. We propose and evaluate the TaggedComments system, a new approach to displaying comments that users post to online tutorials. Using tags supplied by commenters, TaggedComments seeks to enhance the role of user comments by 1) improving their visibility, 2) allowing users to personalize their use of the comments according to their particular information needs, and 3) providing direct access to potentially helpful comments from the tutorial content. A laboratory evaluation with 16 participants shows that, in comparison to the standard comment layout, TaggedComments significantly improves users' subjective impressions of comment utility when interacting with Photoshop tutorials.

We present an approach to content control where parents and children collaboratively configure restrictions and filters, an approach that focuses on education rather than simple rule setting. We conducted an initial exploratory qualitative study with results highlighting the importance that parents place on avoiding inappropriate content. Building on these findings, we designed an initial prototype which allows parents and children to work together to select appropriate applications, providing an opportunity for parents to educate their children on what is appropriate. A second qualitative study with parents and children in the six to eight year-old age group revealed a favorable response to this approach. Our results suggest that parents felt that this approach helped facilitate discussions with their children and made the education more enjoyable and approachable, and that children may have also learned from the interaction. In addition, the approach provided some parents with insights into their children's interests and understanding of their notions of appropriate and inappropriate content.

CommandMaps are a promising interface technique that flattens command hierarchies and exploits human spatial memory to provide rapid access to commands. CommandMaps have performed favorably in constrained cued-selection studies, but have not yet been tested in the context of real tasks. In this paper we present two real-world implementations of CommandMaps: one for Microsoft Word and one for an image editing program called Pinta. We use these as our experimental platforms in two experiments. In the first, we show that CommandMaps demonstrate performance and subjective advantages in a realistic task. In the second, we observe naturalistic use of CommandMaps over the course of a week, and gather qualitative data from interviews, questionnaires, and conversations. Our results provide substantial insight into users' reactions to CommandMaps, showing that they are positively received by users and allowing us to provide concrete recommendations to designers regarding when and how they should be implemented in real applications.

Remote control robots are being found in an increasing number of application domains, including search and rescue, exploration, and reconnaissance. There is a large body of HRI research that investigates interface design for remote navigation, control, and sensor monitoring, while aiming for interface enhancements that benefit the remote operator such as improving ease of use, reducing operator mental load, and maximizing awareness of a robot's state and remote environment. Even though many remote control robots have multi-degree-of-freedom robotic manipulator arms for interacting with the environment, there is only limited research into easy-to-use remote control interfaces for such manipulators, and many commercial robotic products are still using simplistic interface technologies such as keypads or gamepads with arbitrary mappings to arm morphology. In this paper, we present an original interface for the remote control of a multi-degree of freedom robotic arm. We conducted a controlled experiment to compare our interface to an existing commercial keypad interface and detail our results that indicate our interface was easier to use, required less cognitive task load, and enabled people to complete tasks more quickly. In this paper, we present an original interface for the remote control of a multi-degree of freedom robotic arm. We conducted a controlled experiment to compare our interface to an existing commercial keypad interface and detail our results that indicate our interface was easier to use, required less cognitive task load, and enabled people to complete tasks more quickly.

Computer Algebra Systems and matrix-based mathematics packages provide sophisticated functionality to assist with mathematical problem solving. However, despite their widespread adoption, little work in the human–computer interaction community has examined the extent to which these computational tools support expert problem solving. In this article, we report findings from a qualitative study comparing and contrasting the work practices and software use of practicing researchers in mathematics and engineering who share the goal of developing and defending new mathematical formulations. Our findings indicate that although computational tools are used by both groups to support their work, current mathematics software plays a relatively minor, somewhat untrusted role in the process. Our data suggest that five primary factors limit the applicability of current mathematics software to expert work practices: (a) a lack of transparency in how current software derives its computed results; (b) the lack of clearly defined operational boundaries indicating whether the system can meaningfully operate on the user's input (whether expressions or data); (c) the need for free-form two-dimensional input to support annotations, diagrams, and in-place manipulation of objects of interest; (d) the potential for transcription problems when switching between physical and computational media; and (e) the need for collaboration, particularly in early stages of problem solving. Each of these issues suggests a concrete direction for future improvement of mathematics software for experts. These findings also have more general implications for the design of computational systems intended to support complex problem solving.

While workers in an urban environment typically enjoy full speed, always available, broadband access, those in rural and remote environments do not necessarily have access to the same level of service. In this paper we describe insights from a qualitative study examining the benefits and continued challenges of using networked technologies for work purposes in rural and remote communities. Our findings indicate that work in these areas increasingly depends on networked technology to support in-situ and geographically distributed work practices, and to ameliorate health and safety issues, but that participants experience significant challenges in obtaining signal access and stability. We discuss implications for design and future research that arise from our findings.

The human hand can naturally coordinate multiple finger joints, and simultaneously tilt, press and roll a pen to write or draw. For this reason, digital pens are now embedded with auxiliary input sensors to capture these actions. Prior research on auxiliary input channels has mainly investigated them in isolation of one another. In this work, we explore the coordinated use of two auxiliary channels, a class of interaction techniques we refer to as a-coord input. Through two separate experiments, we explore the design space of a-coord input. In the first study we identify if users can successfully coordinate two auxiliary channels. We found a strong degree of coordination between channels. In a second experiment, we evaluate the effectiveness of a-coord input in a task with multiple steps, such as multi-parameter selection and manipulation. We find that a-coord input facilitates coordination even with a complex, aforethought sequential task. Overall our results indicate that users can control at least two auxiliary input channels in conjunction which can facilitate a number of common tasks can on the pen.

Designers of GUI applications typically arrange commands in hierarchical structures, such as menus, due to screen space limitations. However, hierarchical organisations are known to slow down expert users. This paper proposes the use of spatial memory in combination with hierarchy flattening as a means of improving GUI performance. We demonstrate these concepts through the design of a command selection interface, called CommandMaps, and analyse its theoretical performance characteristics. We then describe two studies evaluating CommandMaps against menus and Microsoft's Ribbon interface for both novice and experienced users. Results show that for novice users, there is no significant performance difference between CommandMaps and traditional interfaces -- but for experienced users, CommandMaps are significantly faster than both menus and the Ribbon.

Intelligent interactive systems (IIS) have great potential to improve users' experience with technology by tailoring their behaviour and appearance to users' individual needs; however, these systems, with their complex algorithms and dynamic behaviour, can also suffer from a lack of comprehensibility and transparency. We present the results of two studies examining the comprehensibility of, and desire for explanations with deployed, low-cost IIS. The first study, a set of interviews with 21 participants, reveals that i) comprehensibility is not always dependent on explanations, and ii) the perceived cost of viewing explanations tends to outweigh the anticipated benefits. Our second study, a two-week diary study with 14 participants, confirms these findings in the context of daily use, with participants indicating a desire for an explanation in only 7% of diary entries. We discuss the implications of our findings for the design of explanation facilities.

Computer Algebra Systems (CAS) provide sophisticated functionality to assist with mathematical problem solving. Despite their widespread adoption, however, little work in the HCI community has examined the extent to which these computational tools support domain experts. In this paper, we report findings from a qualitative study investigating the work practices and tools of nine mathematicians in a research setting. Counter to our expectations, our data suggests that computational tools play only a minor role in their workflow, with the limited use of CAS owing primarily to four factors: (1) the need for transparency in CAS's reasoning to explain computed results; (2) the problem of rigidity and formality in CAS's input/output style dialogue; (3) the need for 2D input to support a wide range of annotations, diagrams, and in-place manipulation of objects of interest; and (4) the need for collaboration, particularly in early stages of problem solving. While grounded in the study of mathematicians, these findings (particularly the first) have implications for the design of computational systems intended to support complex problem solving.

This research examines a design guideline that aims to increase the positive perception of interruptions. The guideline advocates matching the amount of attention attracted by an interruption's notification method (attentional draw) to the utility of the interruption content. Our first experiment examined a set of 10 visual notification signals in terms of their detection times and established a set of three significantly different signals along the spectrum of attentional draw. Our second experiment investigated matching these different signals to interruption content with different levels of utility. Results indicate that the matching strategy decreases annoyance and increases perception of benefit compared to a strategy that uses the same signal regardless of interruption utility, with no significant impact on workload or performance. Design implications arising from the second experiment as well as recommendations for future work are discussed.

We describe a mixed-initiative framework designed to sup-port the customization of complex graphical user interfaces. The framework uses an innovative form of online GOMS analysis to provide the user with tailored customization sug-gestions aimed at maximizing the user’s performance with the interface. The suggestions are presented non-intrusively, minimizing disruption and allowing the user to maintain full control. The framework has been applied to a general user-productivity application. A formal user evaluation of the system provides encouraging evidence that this mixed-initiative approach is preferred to a purely adaptable alterna-tive and that the system’s suggestions help improve task per-formance.

As computer applications become larger with every new version, there is a growing need to provide some way for users to manage the interface complexity. There are three different potential solutions to this problem: 1) an adaptable interface that allows users to customize the application to suit their needs; 2) an adaptive interface that performs the adaptation for the users; or 3) a combination of the adaptive and adaptable solutions, an approach that would be suitable in situations where users are not customizing effectively on their own. In this paper we examine what it means for users to engage in effective customization of a menu-based graphical user interface. We examine one aspect of effective customization, which is how characteristics of the users' tasks and customization behaviour affect their performance on those tasks. We do so by using a process model simulation based on cognitive modelling that generates quantitative predictions of user performance. Our results show that users can engage in customization behaviours that vary in efficiency. We use these results to suggest how adaptive support could be added to an adaptable interface to improve the effectiveness of the users' customization.