Let's Think Together! Assessing Shared Mental Models, Performance, and Trust in Human-Agent Teams

An emerging research agenda in Computer-Supported Cooperative Work focuses on human-agent teaming and AI agent's roles and effects in modern teamwork. In particular, one understudied key question centers around the construct of team cognition within human-agent teams. This study explores the unique nature of team dynamics in human-agent teams compared to human-human teams and the impact of team composition on perceived team cognition, team performance, and trust. In doing so, a mixed-method approach, including three team composition conditions (all human, human-human-agent, human-agent-agent), completed the team simulation NeoCITIES and completed shared mental model, trust, and perception measures. Results found that human-agent teams are similar to human-only teams in the iterative development of team cognition and the importance of communication to accelerating its development; however, human-agent teams are different in that action-related communication and explicitly shared goals are beneficial to developing team cognition. Additionally, human-agent teams trusted agent teammates less when working with only agents and no other humans, perceived less team cognition with agent teammates than human ones, and had significantly inconsistent levels of team mental model similarity when compared to human-only teams. This study contributes to Computer-Supported Cooperative Work in three significant ways: 1) advancing the existing research on human-agent teaming by shedding light on the relationship between humans and agents operating in collaborative environments, 2) characterizing team cognition development in human-agent teams; and 3) advancing real-world design recommendations that promote human-centered teaming agents and better integrate the two.

Does the Correspondence Bias Apply to Social Robots?: Dispositional and Situational Attributions of Human Versus Robot Behavior

Increasingly, people interact with embodied machine communicators and are challenged to understand their natures and behaviors. The Fundamental Attribution Error (FAE, sometimes referred to as the correspondence bias) is the tendency for individuals to over-emphasize personality-based or dispositional explanations for other people’s behavior while under-emphasizing situational explanations. This effect has been thoroughly examined with humans, but do people make the same causal inferences when interpreting the actions of a robot? As compared to people, social robots are less autonomous and agentic because their behavior is wholly determined by humans in the loop, programming, and design choices. Nonetheless, people do assign robots agency, intentionality, personality, and blame. Results of an experiment showed that participants made correspondent inferences when evaluating both human and robot speakers, attributing their behavior to underlying attitudes even when it was clearly coerced. However, they committed a stronger correspondence bias in the case of the robot–an effect driven by the greater dispositional culpability assigned to robots committing unpopular behavior–and they were more confident in their attitudinal judgments of robots than humans. Results demonstrated some differences in the global impressions of humans and robots based on behavior valence and choice. Judges formed more generous impressions of the robot agent when its unpopular behavior was coerced versus chosen; a tendency not displayed when forming impressions of the human agent. Implications of attributing robot behavior to disposition, or conflating robot actors with their actions, are addressed.

Robot Process Automation (RPA) and Its Future

Many software automation techniques have been developed in the last decade to cut down cost, improve customer satisfaction, and reduce errors. Robotic process automation (RPA) has become increasingly popular recently. RPA offers software robots (bots) that can mimic human behavior. Attended robots work in tandem with humans and can operate while the human agent is active on the computer. On the other hand, unattended robots operate behind locked screens and are designed to execute automations that don't require any human intervention. RPA robots are equipped with artificial intelligence engines such as computer vision and machine learning, and both robot types can learn automations by recording human actions.

THE DELETION OF THE HUMAN AGENT IN ENVIRONMENTAL SCIENCE DISCOURSE: AN ECOLINGUISTIC STUDY

This study explores the use of the strategy of erasure in environmental science discourses to explore the deletion of the agent. Three environmental science textbooks have been chosen for analysis. Stibbe’s (2015) framework of erasure has been used as a model for analyzing the data. He asserts that the natural world is marginalized in texts through the use of certain linguistic strategies; these strategies run throughout the whole discourse to construct the erasure of the ecosystem. The researchers aim to identify erasure at the level of void, which is the complete erasure or deletion of the agent from these discourses. Stibbe mentions nine linguistic strategies for the construction of erasure in environmental discourses. These strategies are passive voice, nominalization, co-hyponymy, hyponymy, metaphor, metonymy, construction of noun phrases, transitivity patterns and massification. For the construction of void, the researchers have analyzed the strategies of passivization and nominalization. It has been found that these strategies are pervasive in the discourses, thereby deleting the agent and constructing void. The study suggests a new way to look at the language of ecological discourses and proposes further studies on how euphemistic language in these discourses can negatively influence readers. Keywords: erasure, mask, void, environmental discourse

Interleaving Automatic Segmentation and Expert Opinion for Retinal Conditions

Optical coherence tomography (OCT) has become the leading diagnostic tool in modern ophthalmology. We are interested here in developing a support tool for the segmentation of retina layers. The proposed method relies on graph theory and geodesic distance. As each retina layer is characterised by different features, the proposed method interleaves various gradients during detection, such as horizontal and vertical gradients or open-closed gradients. The method was tested on a dataset of 750 OCT B-Scan Spectralis provided by the Ophthalmology Department of the County Emergency Hospital Cluj-Napoca. The method has smaller signed error on layers B1, B7 and B8, with the highest value of 0.43 pixels. The average value of signed error on all layers is −1.99 ± 1.14 px. The average value for mean absolute error is 2.60 ± 0.95 px. Since the target is a support tool for the human agent, the ophthalmologist can intervene after each automatic step. Human intervention includes validation or fine tuning of the automatic segmentation. In line with design criteria advocated by explainable artificial intelligence (XAI) and human-centered AI, this approach gives more control and transparency as well as more of a global perspective on the segmentation process.

Being Guilty

What can guilt, the painful sting of the bad conscience, tell us about who we are as human beings? This book seeks to answer this question through an examination of the views of Immanuel Kant, Friedrich Wilhelm von Schelling, Arthur Schopenhauer, Paul Rée, Friedrich Nietzsche, and Martin Heidegger on guilt, freedom, responsibility, and conscience. The concept of guilt has not received sufficient attention from scholars of the history of German philosophy. The book addresses this lacuna and shows how the philosophers’ arguments can be more deeply grasped once read in their historical context. A main claim of the book is that this history could be read as proceeding dialectically. Thus, in Kant, Schelling, and Schopenhauer, there are variations on the idea that guilt is justified because the human agent is a free cause of his or her own being—a causa sui—and thus responsible for his or her “ontological guilt.” In contrast, in Rée and Nietzsche, these ideas are rejected, and the conclusion is reached that guilt is not justified but is explainable psychologically. Finally, in Heidegger, we find a synthesis of sorts, where the idea of causa sui is rejected, but ontological guilt is retained and guilt is seen as possible, because for Heidegger, a condition of possibility of guilt is that we are ontologically guilty yet not causa sui. In the process of unfolding this trajectory, the various philosophers’ views on these and many other issues are examined in detail.

MT Evaluation in the Context of Language Complexity

The paper focuses on investigating the impact of artificial agent (machine translator) on human agent (posteditor) using a proposed methodology, which is based on language complexity measures, POS tags, frequent tagsets, association rules, and their summarization. We examine this impact from the point of view of language complexity in terms of word and sentence structure. By the proposed methodology, we analyzed 24 733 tags of English to Slovak translations of technical texts, corresponding to the output of two MT systems (Google Translate and the European Commission’s MT tool). We used both manual (adequacy and fluency) and semiautomatic (HTER metric) MT evaluation measures as the criteria for validity. We show that the proposed methodology is valid based on the evaluation of frequent tagsets and rules of MT outputs produced by Google Translate or of the European Commission’s MT tool, and both postedited MT (PEMT) outputs using baseline methods. Our results have also shown that PEMT output produced by Google Translate is characterized by more frequent tagsets such as verbs in the infinitive with modal verbs compared to its MT output, which is characterized by masculine, inanimate nouns in locative of singular. In the MT output, produced by the European Commission’s MT tool, the most frequent tagset was verbs in the infinitive compared to its postedited MT output, where verbs in imperative and the second person of plural occurred. These findings are also obtained from the use of the proposed methodology for MT evaluation. The contribution of the proposed methodology is an identification of systematic not random errors. Additionally, the study can also serve as information for optimizing the translation process using postediting.

A Novel Training and Collaboration Integrated Framework for Human–Agent Teleoperation

Human operators have the trend of increasing physical and mental workloads when performing teleoperation tasks in uncertain and dynamic environments. In addition, their performances are influenced by subjective factors, potentially leading to operational errors or task failure. Although agent-based methods offer a promising solution to the above problems, the human experience and intelligence are necessary for teleoperation scenarios. In this paper, a truncated quantile critics reinforcement learning-based integrated framework is proposed for human–agent teleoperation that encompasses training, assessment and agent-based arbitration. The proposed framework allows for an expert training agent, a bilateral training and cooperation process to realize the co-optimization of agent and human. It can provide efficient and quantifiable training feedback. Experiments have been conducted to train subjects with the developed algorithm. The performances of human–human and human–agent cooperation modes are also compared. The results have shown that subjects can complete the tasks of reaching and picking and placing with the assistance of an agent in a shorter operational time, with a higher success rate and less workload than human–human cooperation.