Candace L. Sidner

Recognizing engagement in human-robot interaction

Based on a study of the engagement process between humans, we have developed and implemented an initial computational model for recognizing engagement between a human and a humanoid robot. Our model contains recognizers for four types of connection events involving gesture and speech: directed gaze, mutual facial gaze, conversational adjacency pairs and backchannels. To facilitate integrating and experimenting with our model in a broad range of robot architectures, we have packaged it as a node in the open-source Robot Operating System (ROS) framework. We have conducted a preliminary validation of our computational model and implementation in a simple human-robot pointing game.

Designing relational agents as long term social companions for older adults

Older adults with strong social connections are at a reduced risk for health problems and mortality. We describe two field studies to inform the development of a virtual agent designed to provide long-term, continuous social support to isolated older adults. Findings include the topics that older adults would like to discuss with a companion agent, in addition to overall reactions to interacting with a remote-controlled companion agent installed in their home for a week. Results indicate a generally positive attitude towards companion agents and a rich research agenda for virtual companion agents.

A reusable framework for health counseling dialogue systems based on a behavioral medicine ontology

Automated approaches to promoting health behavior change, such as exercise, diet, and medication adherence promotion, have the potential for significant positive impact on society. We describe a theory-driven computational model of dialogue that simulates a human health counselor who is helping his or her clients to change via a series of conversations over time. Applications built using this model can be used to change the health behavior of patients and consumers at low cost over a wide range of media including the web and the phone. The model is implemented using an OWL ontology of health behavior change concepts and a public standard task modeling language (ANSI/CEA-2018). We demonstrate the power of modeling dialogue using an ontology and task model by showing how an exercise promotion system developed in the framework was re-purposed for diet promotion with 98% reuse of the abstract models. Evaluations of these two systems are presented, demonstrating high levels of fidelity to best practices in health behavior change counseling.

Automated interventions for multiple health behaviors using conversational agents

OBJECTIVEnAn automated health counselor agent was designed to promote both physical activity and fruit and vegetable consumption through a series of simulated conversations with users on their home computers.nnnMETHODSnThe agent was evaluated in a 4-arm randomized trial of a two-month daily contact intervention comparing: (a) physical activity; (b) fruit and vegetable consumption; (c) both interventions; and (d) a non-intervention control. Physical activity was assessed using daily pedometer steps. Daily servings of fruit and vegetables were assessed using the NIH/NCI self-report Fruit and Vegetable Scan.nnnRESULTSnParticipants in the physical activity intervention increased their walking on average compared to the control group, while those in the fruit and vegetable intervention and combined intervention decreased walking. Participants in the fruit and vegetable intervention group consumed significantly more servings per day compared to those in the control group, and those in the combined intervention reported consuming more compared to those in the control group.nnnCONCLUSIONnAutomated health intervention software designed for efficient re-use is effective at changing health behavior.nnnPRACTICE IMPLICATIONSnAutomated health behavior change interventions can be designed to facilitate translation and adaptation across multiple behaviors.

Using collaborative discourse theory to partially automate dialogue tree authoring

We have developed a novel methodology combining hierarchical task networks with traditional dialogue trees that both partially automates dialogue authoring and improves the degree of dialogue structure reuse. The key to this methodology is a lightweight utterance semantics derived from collaborative discourse theory, making it a step towards dialogue generation based on cognitive models rather than manual authoring. We have implemented an open-source tool, called Disco for Games (D4g), to support the methodology and present a fully worked example of using this tool to generate a dialogue about baseball.

Interactive Hierarchical Task Learning from a Single Demonstration

We have developed learning and interaction algorithms to support a human teaching hierarchical task models to a robot using a single demonstration in the context of a mixedinitiative interaction with bi-directional communication. In particular, we have identified and implemented two important heuristics for suggesting task groupings based on the physical structure of the manipulated artifact and on the data flow between tasks. We have evaluated our algorithms with users in a simulated environment and shown both that the overall approach is usable and that the grouping suggestions significantly improve the learning and interaction. Categories and Subject Descriptors I.2.9 [Artificial Intelligence]: Robotics

Generating connection events for human-robot collaboration

We have developed and tested a reusable Robot Operating System (ROS) module that supports engagement between a human and a humanoid robot by generating appropriate directed gaze, mutual facial gaze, adjacency pair and backchannel connection events. The module implements policies for adding gaze and pointing gestures to referring phrases (including deictic and anaphoric references), performing end-of-turn gazes, responding to human-initiated connection events and maintaining engagement. The module also provides an abstract interface for receiving information from a collaboration manager using the Behavior Markup Language (BML) and exchanges information with our previously developed engagement recognition module.

A design methodology for expressing emotion on robot faces

We have developed a systematic methodology for designing emotional facial expressions for humanoid robots, especially those with limited degrees of freedom. The methodology is firmly grounded in the psychological literature on human static and dynamic emotional facial expressions. We demonstrate the methodology by applying it to a recent humanoid robot and evaluate the results, confirming that the observed confusion matrix agrees qualitatively with the predictions of the methodology. We also investigate how robot facial emotion recognition compares for dynamic versus static expressions.

A Robotic Companion for Social Support of Isolated Older Adults

We demonstrate interaction with a relational agent, embodied as a robot, to provide social support for isolated older adults. Our robot supports multiple activities, including discussing the weather, playing cards and checkers socially, maintaining a calendar, talking about family and friends, discussing nutrition, recording life stories, exercise coaching and making video calls.

On the Sociability of a Game-Playing Agent: A Software Framework and Empirical Study

We report on the results of evaluating a virtual agent that plays games with automatically generated social comments and social gaze. The agent played either a card game (rummy) or a board game (checkers) with each of 31 participants. Based on objective and subjective measures, the agent using social comments and gaze was preferred to both a version of the agent using only social gaze and to playing the game interactively, but without a virtual agent. We have also developed a generic software framework for authoring social comments for any game based on the semantics of the game.