Agents In Virtual Environments Research Articles

External and Internal Attribution in Human-Agent Interaction: Insights from Neuroscience and Virtual Reality

Agents are designed in the image of humans, both internally and externally. The internal systems of agents imitate the human brain, both at the levels of hardware (i.e., neuromorphic computing) and software (i.e., neural networks). Furthermore, the external appearance and behaviors of agents are designed by people and based on human data. Sometimes, these humanlike qualities of agents are purposely selected to increase their social influence over human users, and sometimes the human factors that influence perceptions of agents are hidden. Inspired by Blascovich’s “threshold of social influence’, a model designed to explain the effects of different methods of anthropomorphizing embodied agents in virtual environments, we propose a novel framework for understanding how humans’ attributions of human qualities to agents affects their social influence in human-agent interaction. The External and Internal Attributions model of social influence (EIA) builds on previous work on agent-avatars in immersive virtual reality and provides a framework to link previous social science theories to neuroscience. EIA connects external and internal attributions of agents to two brain networks related to social influence. the external perception system, and the mentalizing system. Focusing human-agent interaction research along each of the attributional dimensions of the EIA model, or at the functional integration of the two, may lead to a better understanding of the thresholds of social influence necessary for optimal human-agent interaction.

Empathic embarrassment towards non-human agents in virtual environments

Humans feel empathic embarrassment by witnessing others go through embarrassing situations. We examined whether we feel such empathic embarrassment even with robot avatars. Participants observed a human avatar and a robot avatar face a series of embarrassing and non-embarrassing scenarios. We collected data for their empathic embarrassment and the cognitive empathy on a 7-point Likert scale. Both empathic embarrassment and cognitive empathy were significantly higher in the embarrassed condition compared to the non-embarrassed condition with both avatars, and the cognitive empathy was significantly higher with the human avatar. There was a tendency of participants showing a higher level of skin conductance while watching the human avatar go through embarrassing situations compared to the robot avatar. A following experiment showed that the average plausibility of the embarrassed condition was significantly higher with the human avatar compared to the robot avatar. However, plausibility scores for emotion were not significantly different among the conditions. These results suggest that humans can feel empathic embarrassment as well as cognitive empathy for robot avatars while cognitive empathy for robot avatars is comparatively lower, and that part of the empathic difference between human and robot avatars might be due to the difference of their plausibility.

Optimization of Neuroprosthetic Vision via End-to-End Deep Reinforcement Learning.

Visual neuroprostheses are a promising approach to restore basic sight in visually impaired people. A major challenge is to condense the sensory information contained in a complex environment into meaningful stimulation patterns at low spatial and temporal resolution. Previous approaches considered task-agnostic feature extractors such as edge detectors or semantic segmentation, which are likely suboptimal for specific tasks in complex dynamic environments. As an alternative approach, we propose to optimize stimulation patterns by end-to-end training of a feature extractor using deep reinforcement learning agents in virtual environments. We present a task-oriented evaluation framework to compare different stimulus generation mechanisms, such as static edge-based and adaptive end-to-end approaches like the one introduced here. Our experiments in Atari games show that stimulation patterns obtained via task-dependent end-to-end optimized reinforcement learning result in equivalent or improved performance compared to fixed feature extractors on high difficulty levels. These findings signify the relevance of adaptive reinforcement learning for neuroprosthetic vision in complex environments.

Synchronizing navigation algorithms for crowd simulation via topological strategies

We present a novel topology-driven method for enhancing the navigation behavior of agents in virtual environments and crowds. In agent-based crowd simulations, each agent combines multiple navigation algorithms for path planning, collision avoidance, and more. This may lead to undesired motion whenever the algorithms disagree on how an agent should pass an obstacle or another agent.In this paper, we argue that all navigation algorithms yield a strategy: a set of decisions to pass obstacles and agents along the left or right. We show how to extract such a strategy from a (global) path and from a (local) velocity. Next, we propose a general way for an agent to resolve conflicts between the strategies of its algorithms. For example, an agent may re-plan its global path when collision avoidance suggests a detour. As such, we bridge conceptual gaps between algorithms, and we synchronize their results in a fundamentally new way. Experiments with an example implementation show that our strategy concept can improve the behavior of agents while preserving real-time performance. It can be applied to many agent-based simulations, regardless of their specific navigation algorithms. The concept is also suitable for explicitly sending agents in particular directions, e.g. to simulate signage.

Analytical Competences of Teachers in Big Data in the Era of Digitalized Learning

The progress of technology in the field of education has led to the development of new professional skills in the teaching community, such as digital competence. Likewise, it has propagated the appearance of new learning environments mediated by technological resources that favor the generation of a large volume of data, known as Big Data, derived from the interactions of educational agents in virtual environments. This study aims to determine the level of digital competence in the informational area of teachers and to determine the projection of Big Data in them. For this study, a descriptive and correlational quantitative method was followed. The study sample consists of 832 Spanish teachers, with data collected through a questionnaire. The results reveal that the teaching staff have a medium level of digital competence in the informational area. The teachers obtained better valuations in the dimensions concerning storage and recovery, postulating themselves as connoisseurs of the Big Data concept to a large extent. In some instances, the teachers affirmed that sometimes they access and interpret the data generated by the digital interactions, transforming it into useful information to improve their teaching processes. Finally, our research shows that the level of digital competence in the informational area influences the degree of knowledge about Big Data.

Architecture and interaction protocol for pedagogical-empathic agents in 3D virtual learning environments

This paper proposes an interaction design architecture and an interaction protocol for the construction of pedagogical agents acting in distributed 3D learning environments. Agents designed based on the proposed architecture are able to interact verbally, non-verbally or both with the students, combining empathic and pedagogical behavioral parameters in order to support students during online educational activities. The representation of the agent results from both pedagogical and emotional factors and is related with the learning environment and its technology. The agent logic is based on three types of the learning environment's events: (a) emotional events, (b) pedagogical events and (c) events that are triggered by the environment itself. The proposed architecture is validated through the implementation of an autonomous pedagogical-empathic agent in the virtual environment of OpenSim. The agent observes students' anxiety during the learning process and reacts when their anxiety level is considered high.

Interactive Teachable Cognitive Agents: Smart Building Blocks for Multiagent Systems

Developing a complex intelligent system by abstracting their behaviors, functionalities, and reasoning mechanisms can be tedious and time consuming. In this paper, we present a framework for developing an application or software system based on smart autonomous components that collaborate with the developer or user to realize the entire system. Inspired by teachable approaches and programming-by-demonstration methods in robotics and end-user development, we treat intelligent agents as teachable components that make up the system to be built. Each agent serves different functionalities and may have prebuilt operations to accomplish its own design objectives. However, each agent may also be equipped with in-built social-cognitive traits to interact with the user or other agents in order to adapt its own operations, objectives, and relationships with others. The results of adaptation can be in the form of groups or multiagent systems as new aggregated components. This approach is made to tackle the difficulties in completely programming the entire system by allowing the user to teach the components toward the desired behaviors in the situated context of the application. We exemplify this novel method with cases in the domains of human-like agents in virtual environment and agents for in-house caregiving.

Automated Space Classification for Network Robots in Ubiquitous Environments

Network robots provide services to users in smart spaces while being connected to ubiquitous instruments through wireless networks in ubiquitous environments. For more effective behavior planning of network robots, it is necessary to reduce the state space by recognizing a smart space as a set of spaces. This paper proposes a space classification algorithm based on automatic graph generation and naive Bayes classification. The proposed algorithm first filters spaces in order of priority using automatically generated graphs, thereby minimizing the number of tasks that need to be predefined by a human. The filtered spaces then induce the final space classification result using naive Bayes space classification. The results of experiments conducted using virtual agents in virtual environments indicate that the performance of the proposed algorithm is better than that of conventional naive Bayes space classification.

Developing Speech Input for Virtual Applications

This paper contextualises the position of speech input from a user-centred human factors perspective. It is presented as a position paper so that researchers and designers can consider the underlying and future factors of a people-orientated approach to speech input for virtual applications. A number of key areas are explored including: human factors for speech input; speech input for virtual applications; speech as a spare mode of interaction; user acceptance and uptake; incorporating speech in the development of virtual applications; and speech input as an interaction tool. Given the user-centred perspective of this paper, this paper does not set out to address issues associated with spoken dialogue technologies, dialogue, and dialogue management; recent work on conversational agents in virtual environments; or multimodal interaction. This paper places the focus more fundamentally within human factors by looking at the user first as a basis for developing usable virtual applications incorporating speech input rather than to review the current state of the art in interaction design. A particular point this paper makes, however, is that speech input should be designed and used as another interaction tool that users need to learn to use, rather than assuming it will offer a natural or intuitive interface.

Recognizing Emotion in Virtual Agent, Synthetic Human, and Human Facial Expressions

A growing interest in the HCI community is the design and development of embodied agents in virtual environments. For virtual environments where social interaction is needed, an agent's facial expression may communicate emotive state to users both young and old. However, younger and older adults differ in how they label human facial expressions (Ruffman et al., 2008). Such possible age-related differences in labeling virtual agent expressions may impact the user's social experience in a virtual environment. The purpose of the current research was to investigate age-related differences in emotion recognition of several on-screen characters of varying degrees of human-likeness. Participants performed a recognition task with three characters demonstrating four basic emotions or neutral. The results indicated age-related differences for all character types. Older adults commonly mislabeled the human and synthetic human emotions of anger, fear, sadness, and neutral. For the virtual agent face, older adults commonly mislabeled the emotions of anger, fear, happiness, and neutral.

Ribbon networks for modeling navigable paths of autonomous agents in virtual environments

This paper presents the Environment Description Framework (EDF) for modeling complex networks of intersecting roads and pathways in virtual environments. EDF represents information about the layout of streets and sidewalks, the rules that govern behavior on roads and walkways, and the locations of agents with respect to navigable structures. The framework serves as the substrate on which behavior programs for autonomous vehicles and pedestrians are built. Pathways are modeled as ribbons in space. The ribbon structure provides a natural coordinate frame for defining the local geometry of navigable surfaces. EDF includes a powerful runtime interface supported by robust and efficient code for locating objects on the ribbon network, for mapping between Cartesian and ribbon coordinates, and for determining behavioral constraints imposed by the environment.

An Efficient and Flexible Perception Pipeline for Autonomous Agents

Agents in virtual environments require a combination of perception and action to behave in an autonomous way. We extend a software architecture for the management of actions blending, called AGENTlib, with a perception mechanism. The perception system provides a uniform interface to various techniques in the field of virtual perception, including synthetic vision, database access and perception persistency. We describe the framework we designed to efficiently filter valuable information from the scene and we address concerns about computation redundancy and data propagation through multiple filtering modules.