How do user-defined gestures for mixed reality change when users’ hands are engaged in tasks? To address this question, we conducted a gesture elicitation study to understand user preferences and the characteristics of gestures conceptualized in three scenarios with varying levels of hand constraints, namely: "both hands free", "one hand fixed", and "both hands busy". We analyzed these gestures across multiple dimensions and compared our findings with those from prior research. Our results indicate that when both hands are occupied, users tend to favor head gestures over those involving other body parts, such as the eyes or legs. Additionally, we found that most of the proposed gestures were metaphorical, with many influenced by legacy bias. These insights enhance our understanding of how hand constraints influence gesture choices in mixed reality scenarios.

Fine-scale pixel-level annotation of minirhizotron root images is a less common and challenging task. We present an interactive segmentation framework to accelerate root annotation. We leverage the concept of few-shot segmentation so that the pretrained model can be effectively fine-tuned and transferred to an unseen category. To provide immediate feedback for real-time interaction, we adapted a UNet architecture by attaching lightweight embedding layers which leveraged a prototype learning (PL) approach to efficiently learn the data metric in the embedding space. The prototypes optimized by the prototype loss preserve the within-class data variation, enabling effective fine-tuning. Furthermore, we designed a system with our interactive annotation framework and experimented with real users to validate the approach.

To design intuitive and effective context-aware task guidance systems, we must understand users’ thought processes and the obstacles they experience when they perform tasks. Though task guidance systems have proven beneficial in many domains for improving task performance and reducing user frustration, there is a lack of general guidelines and design principles for their development. Prior work has shown that recipe-based cooking is a strong medium for studying task planning and execution. In response, we conducted a contextual inquiry study in home kitchens, observing eight different participants’ cooking sessions. We used affinity diagramming of our notes and transcripts to identify common obstacles faced by participants and establish user needs in the areas of object interaction, safety, knowledge base, and task coordination. We discuss how these findings can inform the design of technology-driven solutions for task guidance systems beyond cooking.

Mobile health applications for weight maintenance offer self-monitoring as a tool to empower users to achieve health goals (e.g., losing weight); yet maintaining consistent self-monitoring over time proves challenging for users. These apps use push notifications to help increase users’ app engagement and reduce long-term attrition, but they are often ignored by users due to appearing at inopportune moments. Therefore, we analyzed whether delivering push notifications based on time alone or also considering user context (e.g., current activity) affected users’ engagement in a weight maintenance app, in a 4-week in-the-wild study with 30 participants. We found no difference in participants’ overall (across the day) self-monitoring frequency between the two conditions, but in the context-based condition, participants responded faster and more frequently to notifications, and logged their data more timely (as eating/exercising occurs). Our work informs the design of notifications in weight maintenance apps to improve their efficacy in promoting self-monitoring.

Artificial Intelligence (AI) has been enhancing data analysis efficiency and accuracy during plant phenotyping, which is vital for tackling global agricultural and environmental challenges. Designing a reliable AI system to assist precise plant phenotyping begins with high-quality phenotypic feature annotation, which usually involves collaboration between plant scientists and AI specialists. However, due to the high level of diversity in these researchers’ backgrounds, it is likely that they have differing user needs from a fine-grained plant feature annotation system. We conducted semi-structured interviews with eight experienced annotators from diverse backgrounds, and observed how they interact with their preferred annotation system, to elucidate the challenges faced when annotating plant features and identify user needs. We collected qualitative responses to the interview questions, and conducted a quantitative evaluation of the agreement of their annotations on the given images. By analyzing the participants’ behaviors and the collected data, we identified common user needs and derived implications for the design of an AI-assisted annotation system, including providing a range of annotation options, the flexibility to adapt annotations, and functions to help addressing uncertainty. Our research contributes to the design of systems that make annotations efficient and reliable, not only benefiting plant phenotyping, but also other interdisciplinary fields that rely on user-driven annotations.

Physical multi-touch spherical displays can provide a direct, hands-on, embodied interaction experience with global visualization data like ocean temperatures and currents. However, current commercially available displays may be cost-prohibitive for educational institutions and/or non-profits to acquire. Virtual globe-based visualizations like Google Earth are a potential alternative, but it is not clear how well the interactive affordances of physical spheres may transfer to the virtual. We conducted a within-subjects comparative study with 21 participants who completed similar tasks on a physical and a virtual spherical interface platform, which were designed to be as similar as possible, in order to allow us to compare the interaction experiences. Our results overall showed no significant difference be-tween usability or task time on the two platforms. In their qualitative feedback, participants noticed the differences between the physical sphere and virtual sphere in terms of effort and motor demand. Our research implies that, in resource-constrained environments, a virtual globe can be a sufficient substitute for a physical sphere from a usability perspective.

Children are being presented with augmented reality (AR) in different contexts, such as education and gaming. However, little is known about how children conceptualize AR, especially AR headsets. Prior work has shown that children's interaction behaviors and expectations of technological devices can be quite different from adults’. It is important to understand children's mental models of AR headsets to design more effective experiences for them. To elicit children's perceptions, we conducted four participatory design sessions with ten children on designing content for imaginary AR headsets. We found that children expect AR systems to be highly intelligent and to recognize and virtually transform surroundings to create immersive environments. Also, children are in favor of using these devices for difficult tasks but prefer to work on their own for easy tasks. Our work contributes new understanding on how children comprehend AR headsets and provides recommendations for designing future headsets for children.

The goal of this paper is to understand differences between children's and adults’ motions in order to improve future motion recognition algorithms for children. Motion-based applications are becoming increasingly popular among children (e.g., games). These applications often rely on accurate recognition of users’ motions to create meaningful interactive experiences. Motion recognition systems are usually trained on adults’ motions. However, prior work has shown that children move differently from adults. Therefore, these systems will likely perform poorly on children's motions, negatively impacting their interactive experiences. Although prior work has established that there are perceivable differences between child and adult motion, these differences are yet to be quantified. If we can quantify these differences, then we can gain new insights about how children perform motions (i.e., their motion qualities). We present 24 articulation features (11 of which we newly developed) that describe motions quantitatively; we then evaluate them on a subset of child and adult motions from the publicly available Kinder-Gator dataset to reveal differences; motions in this dataset are represented as postures, each of which is defined by 3D positions of 20 joints tracked by a Kinect at a specific time instance. Our results showed that children perform motions that are quantifiably faster, more intense, less smooth, and less coordinated as compared to adults. Based on our results, we propose guidelines for improving motion recognition algorithms and designing motion applications for children.

Dual-modality blocks-text programming environments have shown promise in helping students learn programming and computational thinking. These environments link blocks-based visualizations to text-based representations, which are more typical of production languages. Since prior work shows that some students who learn in dual-modality environments outperform those who learn in text on assessments, we sought to understand specifically how students use dual-modality environments and what support these environments provide to the learning process. We analyzed survey responses and tool logs collected during a study at a large public university in a CS1 course (N=425). We found that students from all prior programming experience backgrounds made use of the ability to visualize code structures by using blocks. Students with prior experience in blocks or no prior experience said they felt the dual-modality instruction helped them understand code structure and meaning. As students progressed through the class, we found that they made more use of the blocks mode's reference palettes than to its drag-and-drop facilities or mode-switching features. By identifying how students interact with dual-modality tools and how they impact student understanding, this work provides guidance for classroom instructors.

Abstract Gesture recognition algorithms help designers create intelligent user interfaces for a number of application areas. However, these recognition algorithms are usually designed to recognize the gestures of adults, not children, and as such they generally do not perform as well for children as adults. Recognition of younger children’s gestures is particularly poor when compared to recognition of older children’s and adults’ gestures. Researchers have begun to examine the aspects of children’s gesture articulation patterns that make recognition difficult. This paper extends the initial work examining child-specific recognition approaches by considering general-purpose approaches and how they might apply to the problem of recognizing children’s touchscreen gestures. This paper presents a survey of existing recognition and analysis techniques for gestures of both adults and children from a human-centered perspective, highlighting ways in which improved recognition can lead to a better experience for children using touchscreen gestures in a variety of contexts.

Multi-touch spherical displays that enable groups of people to collaboratively interact are increasingly being used in informal learning settings such as museums. Prior research on large flatscreen displays has examined group collaboration patterns in museum settings to inform the design of group learning experiences around these displays. However, previous research has shown differences in how users conceptualize interacting with spherical and flatscreen displays, thereby making it important to separately investigate how groups naturally collaborate around spherical displays in a museum setting. The spherical form factor of the display affords new forms of collaboration: unlike flatscreen displays, spherical displays do not have a definite front or center, thus intrinsically creating both shared and private touch interaction areas on the display based on users' viewing angles or physical arrangements. We conducted a 5-day long field study at a local science museum during which 571 visitors (370 adults and 201 children) in 211 groups interacted with a walk-up-and-use collaborative learning application showing global science data visualizations, on a multi-touch spherical display. We qualitatively analyzed groups' natural collaboration patterns including their physical arrangements (F-formations), their collaboration profiles (e.g., turn-taker or independent), and the nature of group discussion around the display. Our results show that groups often engaged in both independent as well as closely collaborative group explorations when interacting around the sphere: physical spacing between group members around the sphere was strongly linked to the way groups collaborated. It was less common for group members to make and accept suggestions or coordinate touch interactions when they did not share the same fields-of-view or touch interaction space with each other around the sphere. We discuss implications for supporting group collaboration in this context which will inform the design of future walk-up-and-use multi-touch spherical display applications for use in public settings.

Fitts' law has accurately modeled both children's and adults' pointing movements, but it is not as precise for modeling movement to small targets. To address this issue, prior work presented FFitts' law, which is more exact than Fitts' law for modeling adults' finger input on touchscreens. Since children's touch interactions are more variable than adults, it is unclear if FFitts' law should be applied to children. We conducted a 2D target acquisition task with 54 children (ages 5-10) to examine if FFitts' law can accurately model children's touchscreen movement time. We found that Fitts' law using nominal target widths is more accurate, with a R2 value of 0.93, than FFitts' law for modeling children's finger input on touchscreens. Our work contributes new understanding of how to accurately predict children's finger touch performance on touchscreens.

Interactive spherical displays offer numerous opportunities for engagement and education in public settings. Prior work established that users' touch-gesture patterns on spherical displays differ from those on flatscreen tabletops, and speculated that these differences stem from dissimilarity in how users conceptualize interactions with these two form factors. We analyzed think-aloud data collected during a gesture elicitation study to understand adults' and children's (ages 7 to 11) conceptual models of interaction with spherical displays and compared them to conceptual models of interaction with tabletop displays from prior work. Our findings confirm that the form factor strongly influenced users' mental models of interaction with the sphere. For example, participants conceptualized that the spherical display would respond to gestures in a similar way as real-world spherical objects like physical globes. Our work contributes new understanding of how users draw upon the perceived affordances of the sphere as well as prior touchscreen experience during their interactions.

Classification accuracy of whole-body gestures can be improved by selecting gestures that have few conflicts (i.e., confusions or misclassifications). To identify such gestures, an understanding of the nuances of how users articulate whole-body gestures can help, especially when conflicts may be due to confusion among seemingly dissimilar gestures. To the best of our knowledge, such an understanding is currently missing in the literature. As a first step to enable this understanding, we designed a method that facilitates investigation of variations in how users move their body parts as they perform a motion. This method, which we call filterJoint, selects the key body parts that are actively moving during the performance of a motion. The paths along which these body parts move in space over time can then be analyzed to make inferences about how users articulate whole-body gestures. We present two case studies to show how the filterJoint method enables a deeper understanding of whole-body gesture articulation, and we highlight implications for the selection of whole-body gesture sets as a result of these insights.

It is well established that children's touch and gesture interactions on touchscreen devices are different from those of adults, with much prior work showing that children's input is recognized more poorly than adults? input. In addition, researchers have shown that recognition of touchscreen input is poorest for young children and improves for older children when simply considering their age; however, individual differences in cognitive and motor development could also affect children's input. An understanding of how cognitive and motor skill influence touchscreen interactions, as opposed to only coarser measurements like age and grade level, could help in developing personalized and tailored touchscreen interfaces for each child. To investigate how cognitive and motor development may be related to children's touchscreen interactions, we conducted a study of 28 participants ages 4 to 7 that included validated assessments of the children's motor and cognitive skills as well as typical touchscreen target acquisition and gesture tasks. We correlated participants? touchscreen behaviors to their cognitive development level, including both fine motor skills and executive function. We compare our analysis of touchscreen interactions based on cognitive and motor development to prior work based on children's age. We show that all four factors (age, grade level, motor skill, and executive function) show similar correlations with target miss rates and gesture recognition rates. Thus, we conclude that age and grade level are sufficiently sensitive when considering children's touchscreen behaviors.

The future of smart environments is likely to involve both passive and active interactions on the part of users. Depending on what sensors are available in the space, users may make use of multimodal interaction modalities such as hand gestures or voice commands. There is a shortage of robust yet controlled multimodal interaction datasets for smart environment applications. One application domain of interest based on current state-of-the-art is authentication for sensitive or private tasks, such as banking and email. We present a novel, large multimodal dataset for authentication interactions in both gesture and voice, collected from 106 volunteers who each performed 10 examples of each of a set of hand gesture and spoken voice commands chosen from prior literature (10,600 gesture samples and 13,780 voice samples). We present the data collection method, raw data and common features extracted, and a case study illustrating how this dataset could be useful to researchers. Our goal is to provide a benchmark dataset for testing future multimodal authentication solutions, enabling comparison across approaches.

In college-level introductory computer science courses, students traditionally learn to program using text-based languages which are common in industry and research. This approach means that learners must concurrently master both syntax and semantics. Blocks-based programming environments have become commonplace in introductory computing courses in K-12 schools and some colleges in part to simplify syntax challenges. However, there is evidence that students may face difficulty moving to text-based programming environments when starting with blocks-based environments. Bi-directional dual-modality programming environments provide multiple representations of programming language constructs (in both blocks and text) and allow students to transition between them freely. Prior work has shown that some students who use dual-modality environments to transition from blocks to text have more positive views of text programming compared to students who move directly from blocks to text languages, but it is not yet known if there is any impact on learning. To investigate the impact on learning, we conducted a study at a large public university across two semesters in a CS1 course (N=673). We found that students performed better on typical course exams when they were taught using dual-modality representations in lecture and were provided dual-modality tools. The results of our work support the conclusion that dual-modality instruction can help students learn computational concepts in early college computer science coursework.

This course will introduce quantitative methods for use in research on child-computer interaction. We will discuss the types of research questions that can be answered with quantitative methods. Experiment design, data logging, data analysis, and simple statistical techniques will be covered. We will also cover important considerations for conductive quantitative work with young children, especially attentional issues that may affect data quality.

HCI researchers have established a number of evidence-based design recommendations for children's touchscreen interfaces based on developmental appropriateness. Yet, these recommendations are scattered within the academic literature and lack a cohesive framework that makes them accessible to app designers. We created a framework of actionable Touchscreen Interaction Design Recommendations for Children (TIDRC, "tide-rock") by conducting a comprehensive review of the relevant literature. We used our TIDRC framework as a lens to empirically evaluate whether these evidence-based design recommendations were implemented within 50 popular iPad apps designed for children. We found a significant gap between research and practice. On average, only 63% of these apps followed design recommendations for meeting children's cognitive (51%), physical (67%), and socio-emotional (72%) needs. We characterize the nature of this gap and discuss opportunities for closing it when designing mobile touchscreen interfaces for children.

Credit and debit cards enable financial transactions at unattended "pay-at-the-pump" gas station terminals across North America. Attackers discreetly open these pumps and install skimmers, which copy sensitive card data. While EMV (“chip-and-PIN”) has made substantial inroads in traditional retailers, such systems have virtually no deployment at pay-at-the-pump terminals due to dramatically higher costs and logistical/regulatory constraints, leaving consumers vulnerable in these contexts. In an effort to improve security, station owners have deployed security indicators such as low-cost tamper-evident seals, and technologists have developed skimmer detection apps for mobile phones. Not only do these solutions put the onus on consumers to notice and react to security concerns at the pump, but the efficacy of these solutions has not been measured. In this paper, we evaluate the indicators available to consumers to detect skimmers. We perform a comprehensive teardown of all known skimmer detection apps for iOS and Android devices, and then conduct a forensic analysis of real-world gas pump skimmer hardware recovered by multiple law enforcement agencies. Finally, we analyze anti-skimmer mechanisms deployed by pump owners/operators, and augment this investigation with an analysis of skimmer reports and accompanying security measures collected by the Florida Department of Agriculture and Consumer Services over four years, making this the most comprehensive long-term study of such devices. Our results show that common gas pump security indicators are not only ineffective at empowering consumers to detect tampering, but may be providing a false sense of security. Accordingly, stronger, reliable, inexpensive measures must be developed to protect consumers and merchants from fraud.

Text languages are perceived by many computer science students as difficult, intimidating, and/or tedious in nature. Conversely, blocks-based environments are perceived as approachable, but many students see them as inauthentic. Bidirectional hybrid environments provide textual and blocks-based representations of the same code, thereby offering students the opportunity to seamlessly transition between representations to build a conceptual bridge between blocks and text. However, it is not known how use of hybrid environments impacts perceptions of programming. To investigate, we conducted a study in a public middle school with six classes (n=129). We found that students who used hybrid environments perceived text more positively than those who moved directly from blocks to text. The results of this research suggest that hybrid programming environments can help to transition students from blocks to text-based programming while minimizing negative perceptions of programming.

Prior work has shown that intelligent user interfaces (IUIs) that use modalities such as speech, gesture, and writing pose challenges for children due to their developing cognitive and motor skills. Research has focused on improving recognition and accuracy by accommodating children's specific interaction behaviors. Understanding children's expectations of IUIs is also important to decrease the impact of recognition errors that occur. To understand children's conceptual model of IUIs, we completed four consecutive participatory design sessions on designing IUIs with an emphasis on error detection and correction. We found that, while children think of interactive systems in terms of both user input and behavior and system output and behavior, they also propose ideas that require advanced system intelligence, e.g., context and conversation. Our work contributes new understanding of how children conceptualize IUIs and new methods for error detection and correction, and will inform the design of future IUIs for children to improve their experience.

We introduce $Q, a super-quick, articulation-invariant point-cloud stroke-gesture recognizer for mobile, wearable, and embedded devices with low computing resources. $Q ran up to 142X faster than its predecessor $P in our benchmark evaluations on several mobile CPUs, and executed in less than 3% of $P's computations without any accuracy loss. In our most extreme evaluation demanding over 99% user-independent recognition accuracy, $P required 9.4s to run a single classification, while $Q completed in just 191ms (a 49X speed-up) on a Cortex-A7, one of the most widespread CPUs on the mobile market. $Q was even faster on a low-end 600-MHz processor, on which it executed in only 0.7% of $P's computations (a 142X speed-up), reducing classification time from two minutes to less than one second. $Q is the next major step for the "$-family" of gesture recognizers: articulation-invariant, extremely fast, accurate, and implementable on top of $P with just 30 extra lines of code.

The computing and STEM industries face challenges in attracting people to fill expanding needs. The literature shows that computing preconceptions shape interest in and impact decisions of whether or not to enter computing disciplines, especially for women and underrepresented minorities. In this study, our research questions focused on how perceptions of programming in elementary and middle school students varied based on prior programming experience. We examined the programming constructs they found challenging. Our study was in the context of a week-long summer camp dedicated to Scratch-based game development. We conducted semi-structured interviews at the beginning, middle, and end of the weeklong program with 28 students who agreed to participate. During the interviews, we asked students about their perceptions of programming in general and which programming constructs they found easy and/or hard. We found that all students perceived programming as a means of creating artifacts, but that students with prior programming experience went deeper by associating programming with process and function. We also characterize the specific Scratch programming constructs that beginning versus experienced children perceive as easy and/or hard. These findings will help experts and educators better understand how children think about programming and how experience changes these perceptions over time. These findings also have implications on the design of curricula and instructional resources to address difficulties children face while learning to program.

Prior work has shown that children exhibit negative collaborative behaviors, such as blocking others' access to objects, when collaborating on interactive tabletop computers. We implemented previous design recommendations, namely separate physical territories and activity roles, which had been recommended to decrease these negative collaborative behaviors. We developed a multi-touch "I-Spy" picture searching application with separate territory partitions and activity roles. We conducted a deep qualitative analysis of how six pairs of children, ages 6 to 10, interacted with the application. Our analysis revealed that the collaboration styles differed for each pair, both in regards to the interaction with the task and with each other. Several pairs exhibited negative physical and verbal collaborative behaviors, such as nudging each other out of the way. Based on our analysis, we suggest that it is important for a collaborative task to offer equal opportunities for interaction, but it may not be necessary to strive for complete equity of collaboration. We examine the applicability of prior design guidelines and suggest open questions for future research to inform the design of tabletop applications to support collaboration for children.

Children's touchscreen stroke gestures are poorly recognized by existing recognition algorithms, especially compared to adults' gestures. It seems clear that improved recognition is necessary, but how much is realistic? Human recognition rates may be a good starting point, but no prior work exists establishing an empirical threshold for a target accuracy in recognizing children's gestures based on human recognition. To this end, we present a crowdsourcing study in which naïve adult viewers recruited via Amazon Mechanical Turk were asked to classify gestures produced by 5- to 10-year-old children. We found a significant difference between human (90.60%) and machine (84.14%) recognition accuracy, over all ages. We also found significant differences between human and machine recognition of gestures of different types: humans perform much better than machines do on letters and numbers versus symbols and shapes. We provide an empirical measure of the accuracy that future machine recognition should aim for, as well as a guide for which categories of gestures have the most room for improvement in automated recognition. Our findings will inform future work on recognition of children's gestures and improving applications for children.

The proliferation of smartphones and tablets has increased children’s access to and usage of touchscreen devices. Prior work on smartphones has shown that children’s touch interactions differ from adults’. However, larger screen devices like tablets and tabletops have not been studied at the same granularity for children as smaller devices. We present two studies: one of 13 children using tablets with pen and touch, and one of 18 children using a touchscreen tabletop device. Participants completed target touching and gesture drawing tasks. We found significant differences in performance by modality for tablet: children responded faster and slipped less with touch than pen. In the tabletop study, children responded more accurately to changing target locations (fewer holdovers), and were more accurate touching targets around the screen. Gesture recognition rates were consistent across devices. We provide design guidelines for children’s touchscreen interactions across screen sizes to inform the design of future touchscreen applications for children.

Child characters are widely used in animations and games; however, child motion capture databases are less easily available than those involving adult actors. Previous studies have shown that there is a perceivable difference in adult and child motion based on point light displays, so it may not be appropriate to just use adult motion data on child characters. Due to the costs associated with motion capture of child actors, it would be beneficial if we could create a child motion corpus by translating adult motion into child-like motion. Previous works have proposed dynamic scaling laws to transfer motion from one character to its scaled version. In this paper, we conduct a perception study to understand if this procedure can be applied to translate adult motion into child-like motion. Viewers were shown three types of point light display videos: adult motion, child motion, and dynamically scaled adult motion and asked to identify if the translated motion belongs to a child or an adult. We found that the use of dynamic scaling led to an increase in the number of people identifying the motion as belonging to a child compared to the original adult motion. Our findings suggest that although the dynamic scaling method is not a final solution to translate adult motion into child-like motion, it is nevertheless an intermediate step in the right direction. To better illustrate the original and dynamically scaled motions for the purposes of this paper, we rendered the dynamically scaled motion on an androgynous manikin character.

Research on children's interactions with touchscreen devices has examined small and large screens and compared interaction to adults or among children of different ages. Little work has explicitly compared interaction on different platforms, however. Large touchscreen displays can be deployed flat, as in a table, or vertically, as on a wall. While these two form factors have been studied, it is not known what differences may exist between them. We present a study of visitors to a science museum, including children and their parents, who interacted with Google Earth on either a touch table or a touch wall. We compare the types of gestures and interactions attempted on each device and find several interesting results, including: users of all ages tend to make standard touchscreen gestures on both platforms, but children were more likely than adults to try new gestures. Users were more likely to perform two-handed, multi-touch gestures on the touch wall than on the touch table. Our findings will inform the design of future interactive applications for each platform.

No abstract available.

Most touchscreen devices are not designed specifically with children in mind, and their interfaces often do not optimize interaction for children. Prior work on children and touchscreen interaction has found important patterns, but has only focused on simplified, isolated interactions, whereas most interfaces are more visually complex. We examine how interface complexity might impact children's touchscreen interactions. We collected touch and gesture data from 30 adults and 30 children (ages 5 to 10) to look for similarities, differences, and effects of interface complexity. Interface complexity affected some touch interactions, primarily related to visual salience, and it did not affect gesture recognition. We also report general differences between children and adults. We provide design recommendations that support the design of touchscreen interfaces specifically tailored towards children of this age.

Children’s touchscreen interaction patterns are generally quite different from those of adults. In particular, it has been established that children’s gestures are recognized by existing algorithms with much lower accuracy than are adults’ gestures. Previous work has qualitatively and quantitatively analyzed adults’ gestures to promote improved recognition, but this has not been done for children’s gestures in the same systematic manner. We present an analysis of gestures elicited from 24 children (age 5 to 10 years old) and 27 adults in which we calculate geometric, kinematic, and relative articulation features of the gestures. We examine the effect of user age on 22 different gesture features to better understand how children’s gesturing abilities and behaviors differ between various age groups, and from adults. We discuss the implications of our findings and how they will contribute to creating new gesture recognition algorithms tailored specifically for children.

Artists and animators have observed that children’s movements are quite different from adults performing the same action. Previous computer graphics research on human motion has primarily focused on adult motion. There are open questions as to how different child motion actually is, and whether the differences will actually impact animation and interaction. We report the first explicit study of the perception of child motion (ages 5 to 9 years old), compared to analogous adult motion. We used markerless motion capture to collect an exploratory corpus of child and adult motion, and conducted a perceptual study with point light displays to discover whether naive viewers could identify a motion as belonging to a child or an adult. We find that people are generally successful at this task. This work has implications for creating more engaging and realistic avatars for games, online social media, and animated videos and movies.

Creating a predefined set of touchscreen gestures that caters to all users and age groups is difficult. To inform the design of intuitive and easy to use gestures specifically for children, we adapted a userdefined gesture study by Wobbrock et al. [12] that had been designed for adults. We then compared gestures created on an interactive tabletop by 12 children and 14 adults. Our study indicates that previous touchscreen experience strongly influences the gestures created by both groups; that adults and children create similar gestures; and that the adaptations we made allowed us to successfully elicit userdefined gestures from both children and adults. These findings will aid designers in better supporting touchscreen gestures for children, and provide a basis for further userdefined gesture studies with children.

We introduce gesture heatmaps, a novel gesture analysis technique that employs color maps to visualize the variation of local features along the gesture path. Beyond current gesture analysis practices that characterize gesture articulations with single-value descriptors, e.g., size, path length, or speed, gesture heatmaps are able to show with colorful visualizations how the value of any such descriptors vary along the gesture path. We evaluate gesture heatmaps on three public datasets comprising 15,840 gesture samples of 70 gesture types from 45 participants, on which we demonstrate heatmaps' capabilities to (1) explain causes for recognition errors, (2) characterize users' gesture articulation patterns under various conditions, e.g., finger versus pen gestures, and (3) help understand users' subjective perceptions of gesture commands, such as why some gestures are perceived easier to execute than others. We also introduce chromatic confusion matrices that employ gesture heatmaps to extend the expressiveness of standard confusion matrices to better understand gesture classification performance. We believe that gesture heatmaps will prove useful to researchers and practitioners doing gesture analysis, and consequently, they will inform the design of better gesture sets and development of more accurate recognizers.

Surface gesture interaction styles used on modern mobile touchscreen devices are often dependent on the platform and application. Some applications show a visual trace of gesture input as it is made by the user, whereas others do not. Little work has been done examining the usability of visual feedback for surface gestures, especially for children. In this paper, we present results from an empirical study conducted with children, teens, and adults to explore characteristics of gesture interaction with and without visual feedback. We find that the gestures generated with and without visual feedback by users of different ages diverge significantly in ways that make them difficult to interpret. In addition, users prefer to see visual feedback. Based on these findings, we present several design recommendations for new surface gesture interfaces for children, teens, and adults on mobile touchscreen devices. In general, we recommend providing visual feedback, especially for children, wherever possible.

In this paper, we describe the challenges we encountered and solutions we developed while collecting mobile touch and gesture interaction data in laboratory conditions from children ages 5 to 7 years old. We identify several challenges of conducting empirical studies with young children, including study length, motivation, and environment. We then propose and validate techniques for designing study protocols for this age group, focusing on the use of gamification components to better engage children in laboratory studies. The use of gamification increased our study task completion rates from 73% to 97%. This research contributes a better understanding of how to design study protocols for young children when lab studies are needed or preferred. Research with younger age groups alongside older children, adults, and special populations can lead to more sound guidelines for universal usability of mobile applications.

Most work on the usability of touchscreen interaction for people with motor impairments has focused on lab studies with relatively few participants and small cross-sections of the population. To develop a richer characterization of use, we turned to a previously untapped source of data: YouTube videos. We collected and analyzed 187 non-commercial videos uploaded to YouTube that depicted a person with a physical disability interacting with a mainstream mobile touchscreen device. We coded the videos along a range of dimensions to characterize the interaction, the challenges encountered, and the adaptations being adopted in daily use. To complement the video data, we also invited the video uploaders to complete a survey on their ongoing use of touchscreen technology. Our findings show that, while many people with motor impairments find these devices empowering, accessibility issues still exist. In addition to providing implications for more accessible touchscreen design, we reflect on the application of user-generated content to study user interface design.

Current measures of stroke gesture articulation lack descriptive power because they only capture absolute characteristics about the gesture as a whole, not fine-grained features that reveal subtleties about the gesture articulation path. We present a set of twelve new relative accuracy measures for stroke gesture articulation that characterize the geometric, kinematic, and articulation accuracy of single and multi-stroke gestures. To compute the accuracy measures, we introduce the concept of a gesture task axis. We evaluate our measures on five public datasets comprising 38,245 samples from 107 participants, about which we make new discoveries; e.g., gestures articulated at fast speed are shorter in path length than slow or medium-speed gestures, but their path lengths vary the most, a finding that helps understand recognition performance. This work will enable a better understanding of users' stroke gesture articulation behavior, ultimately leading to better gesture set designs and more accurate recognizers.

This paper describes a Science-Technology-Engineering-Mathematics (STEM) outreach workshop conducted with post-secondary students diagnosed with learning differences, including Learning Disabilities (LD), Attention Deficit / Hyperactivity Disorders (AD/HD), and/or Autism Spectrum Disorders (ASD). In this workshop, students were actively involved in participatory design exercises such as data gathering, identifying accessible design requirements, and evaluating mobile applications and games targeted for diverse users. This hands-on experience broadened students' understanding of STEM areas, provided them with an opportunity to see themselves as computer scientists, and demonstrated how they might succeed in computing careers, especially in human-centered computing and interface design. Lessons learned from the workshop also offer useful insight on conducting participatory design with this unique population.

Rapid prototyping of gesture interaction for emerging touch platforms requires that developers have access to fast, simple, and accurate gesture recognition approaches. The $-family of recognizers ($1, $N) addresses this need, but the current most advanced of these, $N-Protractor, has significant memory and execution costs due to its combinatoric gesture representation approach. We present $P, a new member of the $-family, that remedies this limitation by considering gestures as clouds of points. $P performs similarly to $1 on unistrokes and is superior to $N on multistrokes. Specifically, $P delivers >99% accuracy in user-dependent testing with 5+ training samples per gesture type and stays above 99% for user-independent tests when using data from 10 participants. We provide a pseudocode listing of $P to assist developers in porting it to their specific platform and a "cheat sheet" to aid developers in selecting the best member of the $-family for their specific application needs.

We believe handwriting input may be able to provide significant advantages over typing, especially in the mathematics learning domain. The use of handwriting may result in decreased extraneous cognitive load on students, and it may provide better support for the two-dimensional spatial components of mathematics when compared to existing typing-based tools. Here we report progress towards the application of a handwriting interface for mathematics learning. We introduce a prototype system that allows students to use handwriting input to solve algebraic equations in an intelligent tutor. We discuss strategies to improve the existing handwriting system and apply it to math learning. Although the recognition accuracy of current handwriting engines may not be at a level suitable for use by students, we hypothesize that this may be realistically improved via advance training of the engine on a large corpus, as well as via techniques similar to co-training