Perception of Artificial Agents and Utterance Friendliness in Dialogue (original) (raw)

How People Talk to Computers, Robots, and Other Artificial Communication Partners

2006

End-to-end evaluations of more conversational dialogue systems with naive users have uncovered severe usability problems that, among other things, result in low task completion rates. First analyses suggest that these problems are related to the system’s dialogue management and turn-taking behavior. This paper starts with a presentation of experimental results, which shed some light on the effects of that behavior. Based on these findings, some criteria which lie orthogonal to dialogue quality are spelled out. As such, they nevertheless constitute an integral part of a more comprehensive view on dialogue felicity as a function of dialogue quality and efficiency. Since the work on spoken and multimodal dialogue systems presented and discussed herein is aimed at more conversational and adaptive systems, we also show that in certain dialogical situations it is important for such systems to align linguistically towards the users. After describing the corresponding empirical experiments ...

How Do People Talk with a Robot? An Analysis of Human-Robot Dialogues in the Real World

This paper reports the preliminary results of a human- robot dialogue analysis in the real world with the goal of understanding users’ interaction patterns. We analyzed the dialogue log data of Roboceptionist, a robotic receptionist located in a high-traffic area in an academic building [2][3]. The results show that (i) the occupation and background (persona) of the robot help people establish common ground with the robot, and (ii) there is great variability in the extent that users follow social norms of human-human dialogues in human-robot dialogues. Based on these results, we describe implications for designing the dialogue of a social robot.

Characterising dimensions of use for designing adaptive dialogues for human-robot communication

2007

In this paper we provide a possible characterisation of user behaviour based on an analysis of a corpus of human-robot communication, collected by using the Wizard-of-Oz technique to elicit communicative behaviour. We distinguish between three general types of user behaviour: uniform user behaviour, idiosyncratic user behaviour and distinguishing user behaviour. We also present an analysis of user behaviour that can be characterized in terms of overall task organisation (i.e., interaction episodes) and behaviour that is intimately connected to communicative behaviour. We also discuss to what extent manipulation of objects to prepare the environment can be used to group users along the dimensions: task-vs. interaction-orientation and control vs. monitoring. Using this typology we discuss categorisation into four dimensions of use: Directors, Monitors, Pointers and Players. To support these use dimensions we propose a set of adaptation foci (Focus on Feedback or Action, and on Proactive or Reactive behaviour)

More than just words: Building a chatty robot

IWSDS, 2012

Speech meditates human interactions in all areas of life. Some conversations have a clear purpose such as communicating important information (warning) or creating change (giving an order), while, in others, the goal of the exchange is not so much to transfer linguistic information as to cement social bonds.

Receptionist or Information Kiosk: How Do People Talk With a Robot?

The mental structures that people apply towards other people have been shown to influence the way people cooperate with others. These mental structures or schemas evoke behavioral scripts. In this paper, we explore two different scripts, receptionist and information kiosk, that we propose channeled visitors’ interactions with an interactive robot. We analyzed visitors’ typed verbal responses to a receptionist robot in a university building. Half of the visitors greeted the robot (e.g., “hello”) prior to interacting with it. Greeting the robot significantly predicted a more social script: more relational conversational strategies such as sociable interaction and politeness, attention to the robot’s narrated stories, self-disclosure, and less negative/rude behaviors. The findings suggest people’s first words in interaction can predict their schematic orientation to an agent, making it possible to design agents that adapt to individuals during interaction. We propose designs for interactive computational agents that can elicit people’s cooperation.

A Review on The Development and Effect of Conversational Agents and Social Robots

HCI addresses to the concept of Human-Machine communication through including but not limited to AI-enabled embodied conversational software agents. The emergence of these agents has changed the history of computing and robotics once and for all. One of the most prominent social and intellectual qualities in humans is the ability to have conversations. Typically, a conversation takes place between people through verbal and non-verbal mediums. Languages play a vital role in these communications and conversations. Humanness and human-like interaction qualities are found to be in the core of the human-computer interface designs from the beginning of this research doctrine [1]. Programming languages has enabled computer scientists to establish a connection between humans and machines that enables the machine to understand the instructions given. However, the widespread use of cell phones, computers and other smart gadgets has clearly made it a demand of time that the machines used today can understand the commands given in natural languages (i.e. English, German, Spanish, etc.) as the user set is not limited to the computer scientists anymore[1]. Hence, robotics, natural language processing, machine learning, artificial intelligence, etc. has combined force to bridge the communication gap between the machines and the users. From ELIZA [3], Rea [4] to Siri, Amazon Alexa or Google assistant, the software interfaces has come a long way through a lengthy development process. They have proven to have enough influence to change the social, economic and political outcomes through their intelligent behavior [2]. The boundary between human-like and bot-like behavior is greyer then it is black and white [2]. The software interfaces has changed their appearance over the time by stripping down from the ideals of face-to-face conversations. The chatbots (i.e. Twitter bots) found online has developed different social media ecosystems [2] where humans and robots interact with each other in the same plane. To have a conversation or interaction with the machines humans are being trained to accept and use a new set of vocabularies [1]. In this paper, I would like to discuss how these conversational agents and social robots are shaping our social media ecosystems. I will revisit the interrelation between humans and machines while focusing on the socio-cultural impact of these robots into our IoT –enabled smart homes and online virtual spaces.

Evaluation of Conversational Agents: Mariá and ET

Informatica Educativa …, 2007

This paper describes ET and Mariá, two conversational agents with different characteristics. Mariá presents a more realistic character; however she is not an interactive agent, since she only provides short movies with predefined questions and animations. ET is an interactive agent, which can recognize keywords (in Portuguese) and visual react emotionally and textually. We describe the development of those agents as well as results obtained with interaction with subjects. We also present details about MECA, the system we developed to Model Embodied Conversational Agents. Our contribution is the comparison, from the educational point of view, of learning aspects in subjects which interact with two communicative agents. Results show that interactivity has great impact in people learning process.

Developing and Evaluating Conversational Agents

2000

Conversation agents present a challenging agenda for research and application. We describe the development, evaluation, and application of Baldi, a computer animated talking head. Baldi's existence is justified by the important contribution of the face in spoken dialog. His actions are evaluated and modified to mimic natural actions as much as possible. Baldi has the potential to enrich human-machine interactions and serve as a tutor in a wide variety of educational domains. We describe one current application of language tutoring with children with hearing loss.