Collaboration with Automation: Machines as Teammates
Permanent URI for this collection
Browse
Recent Submissions
1 - 10 of 10
-
ItemWhere is the Bot in our Team? Toward a Taxonomy of Design Option Combinations for Conversational Agents in Collaborative Work( 2019-01-08)With rapid progress in machine learning, language technologies and artificial intelligence, conversational agents (CAs) gain rising attention in research and practice as potential non-human teammates, facilitators or experts in collaborative work. However, designers of CAs in collaboration still struggle with a lack of comprehensive understanding of the vast variety of design options in the dynamic field. We address this gap with a taxonomy to help researchers and designers understand the design space and the interrelations of different design options and recognize useful design option combinations for their CAs. We present the iterative development of a taxonomy for the design of CAs grounded in state of the art literature and validated with domain experts. We identify recurring design option combinations and white spots from the classified objects that will inform further research and development efforts.
-
ItemThe Future of Human-AI Collaboration: A Taxonomy of Design Knowledge for Hybrid Intelligence Systems( 2019-01-08)Recent technological advances, especially in the field of machine learning, provide astonishing progress on the road towards artificial general intelligence. However, tasks in current real-world business applications cannot yet be solved by machines alone. We, therefore, identify the need for developing socio-technological ensembles of humans and machines. Such systems possess the ability to accomplish complex goals by combining human and artificial intelligence to collectively achieve superior results and continuously improve by learning from each other. Thus, the need for structured design knowledge for those systems arises. Following a taxonomy development method, this article provides three main contributions: First, we present a structured overview of interdisciplinary research on the role of humans in the machine learning pipeline. Second, we envision hybrid intelligence systems and conceptualize the relevant dimensions for system design for the first time. Finally, we offer useful guidance for system developers during the implementation of such applications.
-
Item
-
ItemUnderstanding the Role of Trust in Human-Autonomy Teaming( 2019-01-08)This study aims to better understand trust in human-autonomy teams, finding that trust is related to team performance. A wizard of oz methodology was used in an experiment to simulate an autonomous agent as a team member in a remotely piloted aircraft system environment. Specific focuses of the study were team performance and team social behaviors (specifically trust) of human-autonomy teams. Results indicate 1) that there are lower levels of trust in the autonomous agent in low performing teams than both medium and high performing teams, 2) there is a loss of trust in the autonomous agent across low, medium, and high performing teams over time, and 3) that in addition to the human team members indicating low levels of trust in the autonomous agent, both low and medium performing teams also indicated lower levels of trust in their human team members.
-
ItemTrusting Robots in Teams: Examining the Impacts of Trusting Robots on Team Performance and Satisfaction( 2019-01-08)Despite the widespread use of robots in teams, there is still much to learn about what facilitates better performance in these teams working with robots. Although trust has been shown to be a strong predictor of performance in all-human teams, we do not fully know if trust plays the same critical role in teams working with robots. This study examines how to facilitate trust and its importance on the performance of teams working with robots. A 2 (robot identification vs. no robot identification) × 2 (team identification vs. no team identification) between-subjects experiment with 54 teams working with robots was conducted. Results indicate that robot identification increased trust in robots and team identification increased trust in one’s teammates. Trust in robots increased team performance while trust in teammates increased satisfaction.
-
ItemThe Effect of Stress on Reliance Decisions( 2019-01-08)Appropriate reliance on automation is critical in high-risk/high-stress contexts such as military operations. The current study examines how stress affects the decision to rely on automation. Reliance will be examined using a decision making framework, taking into account the cognitive processes that are being affected by stress. Additionally, the role of feedback is essential in updating information to be able to make a more informed decision in the future. Reliability of automation will be manipulated so that actual reliability will be lower than expected reliability in one condition, and will be the same as expected in the other. Participants’ ability to incorporate feedback into subsequent reliance decisions will be assessed between stress conditions. Finally, since motivation can encourage more deliberate thinking, motivation will be manipulated between subjects. A better understanding of how reliance decisions are made under stress can inform the design of systems for better human-automation collaboration.
-
ItemDesigning Automated Facilitation for Design Thinking: A Chatbot for Supporting Teams in the Empathy Map Method( 2019-01-08)The Empathy Map Method (EMM) in the Design Thinking approach is a powerful tool for user centered design but relies on the methodological skills and experience of rare facilitation experts to guide the team. In a collaboration engineering effort, we aim to make this expertise available to teams without constant access to a professional facilitator by packaging facilitation knowledge into structured process support and state-of-the art technology. Based on requirements from scientific and practitioners’ literature, we introduce the concept of a conversational agent in the form of a chatbot to take over the role of the facilitator of the EMM. We present an initial wizard of oz evaluation to derive insights and implications for improvements and the software implementation towards the ambitious goal of automated, non-human facilitation of EMM.
-
ItemTowards a Technique for Modeling New Forms of Collaborative Work Practices – The Facilitation Process Model 2.0( 2019-01-08)Collaboration Engineering (CE) is an approach for the design and deployment of repeatable collaborative work practices that can be executed by practitioners themselves without the ongoing support of external collaboration professionals. A key design activity in CE concerns modeling current and future collaborative work practices. CE researchers and practitioners have used the Facilitation Process Model (FPM) technique. However, this modeling technique suffers from a number of shortcomings to model contemporary collaborative work practices. We use a design science approach to identify the main challenges with the original FPM technique, derive requirements and design a revised modeling technique that is based on the current technique enriched by BPMN 2.0 elements. This paper contributes to the CE literature by offering a revised FPM technique that assists CE-designers to capture new forms of collaborative work practices.
-
ItemExploring Automated Leadership and Agent Interaction Modalities( 2019-01-08)Advances in computer technology and research in the field of artificial intelligence have enabled computers to take on roles traditionally held by humans. Insights from leadership research have identified behaviors that, when applied strategically and systematically, can improve individual and team performance. We propose that some aspects of leadership are candidates for automation. This paper briefly reviews relevant leadership literature and describes three leadership behaviors that may be possibly automated: goal setting, performance monitoring, and performance consequences. The paper also explores the relationship of different embodiments of the artificial leaders, the impact of these embodiments in conveying social presence and the impact of this presence on performance and satisfaction outcomes. We conducted an experiment to investigate the effect of automated leadership on follower attitudes and behavior. Initial results suggest that automated leadership may positively influence performance and accuracy for individuals engaged in a clerical task.
-
Item