Peggy Wu

A Computational Approach to Etiquette: Operationalizing Brown and Levinson's Politeness Model

A central source of cultural differences is the communication of politeness, which can powerfully affect perception and behavior. Our computational adaptation of a universal theory of human politeness (from Brown and Levinson, 1987) combines culture-specific aspects of social context to generate expectations, score, interpret and recommend polite and impolite behaviors. Tests show this approach could reduce software development costs and increase an intelligent agents behavior repertoire through the incorporation of modular, cross-cultural etiquette libraries.

The Playbook™ Approach to Adaptive Automation

SIFT has pioneered a human-automation integration architecture, called Playbook™, based on a shared model of the tasks in the domain. This shared task model provides a means of human-automation communication about plans, goals, methods and resource usage—a process akin to referencing plays in a sports teams playbook. The Playbook enables human operators to interact with subordinate systems with the same flexibility as with well-trained human subordinates, thus allowing for adaptive automation. We describe this approach and its application in an ongoing project called Playbook-enhanced Variable Autonomy Control System™ (P-VACS).

Maintaining Psycho-Social Health on the Way to Mars and Back

In future long duration Mars exploration missions, network limitations and the lack of real-time communication capabilities will impact various aspects of space crew performance as well as behavioral health. Studies in ground-based analogs of Isolated and Confined Environments (ICE) such as Antarctica have identified sensory and social monotony as threats to crew psychological well-being. Given the importance of behavioral health to mission success and the extreme conditions of space travel, new methods of maintaining psycho-social health and social connections to support systems are critical. We describe ANSIBLE -- A Network of Social Interactions for Bilateral Life Enhancement. ANSIBLE leverages Virtual Environments (VEs) to deliver evidence based wellness promoting strategies and socially intelligent Virtual Agents (VAs) as tools to facilitate asynchronous human-human communication, and counteract behavioral health challenges associated with prolonged isolation and deep space exploration.

Politeness in Teams: Implications for Directive Compliance Behavior and Associated Attitudes

This article reports an experiment in which college students and professional combat air controllers performed a simulated team interaction task designed to explore the effects of the degree of politeness used by a directive giver and the degree of “social distance” (roughly, team affiliation and affinity between the directive giver and the recipient), on directive compliance behaviors and attitudes. The design and experimental approach was informed by the functional theory of politeness in social interactions developed by Brown and Levinson, although hypotheses are advanced that extend this essentially perceptual model to effects on behaviors and attitudes. Results showed that increased politeness in a directive significantly improved attitudes toward the directive giver. Social “nearness” operated similarly and influenced the degree of politeness perceived even when the request itself was unchanged. Both effects operate similarly for novices and experts. Compliance rates (and one portion of reaction time) were similarly affected by the politeness of the directive giver but, interestingly, were affected differently for novices and experts. The politeness of the directive giver increased compliance for novices but decreased it for experts. This result suggests that politeness perceptions are an important influence on work performance but that their interpretation can be influenced through training and/or work “culture.”

Interactive Phrasebook - Embedding Human Social Dynamics in Language Training

Today, many individuals must interact with others who not only use different languages, but have fundamentally different belief structures and concepts of trust and collaboration. While proficiency in a language is important, it is only one aspect of effective communication. Knowledge about culturally influenced social norms, or etiquette, directly impacts interaction outcomes. Computer-based training systems are a promising and cost effective solution for providing simulated social interactions for a learner to both practice language use and observe cultural nuances in the dynamics of an interaction. A computerized actor which displays social characteristics consistent with its cultural background can provide low cost training and rehearsal opportunities. Such a system would allow humans to train with realistically-behaving simulations in high fidelity situational contexts, and interactions with virtual agents that reflect the types of people they will be required to work with. The authors have adapted a well grounded socio-linguistic model of human-human interaction, and applied it in a language and culture training application. Interactive Phrasebook leverages work in a scalable, customizable, and computationally tractable model of a universal code of conduct that defines how social dynamics such as power and social distance manifests into the expression and interpretation of interaction behaviors.

Non-intrusive detection of psycho-social dimensions using sociolinguistics

Long duration space flights such as a two and a half year mission to Mars present many unique challenges to the behavioral health of astronauts. Factors such as social monotony, workload, a confined environment, sensory deprivation, and limited access to family and psychosocial support can affect crew welfare and task performance. NASA recognizes a “risk of performance decrements due to inadequate cooperation, coordination, communication, and psychosocial adaptation within a team;” reports from Mir revealed that conflicts between crew members have resulted in early termination of missions. Currently, flight crews and support staff have real time voice and video communications capabilities on the International Space Station to keep astronauts connected, and allow operations staff to monitor the crews well-being. However, communications for long duration missions will likely be limited and disrupted by time latencies. Crew workload may also prohibit crew members from providing the extensive self-reports that the Earth-bound support team needs to accurately access the crews psychological health. Further, the metrics of interest are difficult to obtain because some are inherently qualitative, while others may not be amendable to self-reports. We first describe an extensive review of psycho-social dimensions relevant to long duration space flight, their manifestations, and possible detection methods. We then describe a novel method of non-intrusive detection developed initially for application in the Empire Challenge military exercise in 2010. This system, called ADMIRE for Assessment of Discourse Media Indicators of Relative Esteem, leverages prior work in cultural and socio-linguistic theory to develop standardized, non-intrusive methods for data collection and knowledge extraction about factors salient to group psychosocial dynamics. Finally, we describe our approach to follow-up work applying ADMIRE to historical space flight data, as well as in ongoing studies in space analog environments to identify potential changes in individual and team psycho-social factors before they lead to deficits in health and task performance.

The Current Bottleneck for Computer-Based Culture Training – Who Cares about Etiquette?

Computerized training systems may be the only viable solution to accommodate the quickly evolving learning needs, the short lead time, and the large number of students that must prepare for cross-cultural interactions. However, current systems place emphasis on language training, which is only part of the equation in effective communication. Knowledge about culturally dependent social norms directly impacts the interaction outcome. A dynamic, customizable, computationally tractable model for the code of conduct, or etiquette, can alert us of the specific communication protocols of groups of interest, and can prevent over-generalizations which can be detrimental to the end goal. We may not consciously care, or even be aware of etiquette, but it impacts our trust and affect for individuals, and ultimately shapes our subsequent actions. We have adapted a socio-linguistic model of human-human interaction for use in language and culture training. This computational model increases the social intelligence of virtual actors by allowing them to act and react to human students in ways that are more aligned with the humans they simulate. Capabilities for virtual agents to reason about and select actions not only enhances interaction fidelity, but can also result in time and cost savings in training scenario generation and maintenance.

ATHENA – A Zero-Intrusion No Contact Method for Workload Detection Using Linguistics, Keyboard Dynamics, and Computer Vision

We describe preliminary evaluation data for ATHENA (Appraisal of Task Health and Effort through Non-intrusive Assessments), a completely no contact, zero-intrusion workload measurement method which harnesses multimodal metrics (e.g. linguistic markers, keyboard dynamics and computer vision). Preliminary results reflect the existence of different types of workload, with our zero-intrusion metrics demonstrating respectable classification accuracies when the variable causing workload (e.g. time) is matched with the type of workload assessed (e.g. temporal). By not requiring extra equipment or interrupting workflow, ATHENA represents a valuable step forward in providing automated workload support tools as well as a tool for understanding the workload concept.

Team PsychoSocial Assessment via Discourse Analysis: Power and Comfort/Routine

We describe the theory, implementation and initial validation of a tool to assess interpersonal relationships and team states through non-intrusive discourse analysis. ADMIRE was developed to assess power/leadership relationships through scoring politeness behaviors in textual dialog to compute a graph of power asserted vs. afforded between individuals—an org chart—trackable over time. ADMIRE was formally tested in a military exercise where it proved 100% successful at deriving power relationships in 3 military chat rooms. Subsequent work extended ADMIRE’s core approach with novel linguistic behaviors to identify “Team Comfort/Routine” (C/R) indicating when a team is performing in a well-understood, relaxed task context. Initial validation is provided by discriminating the disastrous Apollo 13 mission from others.

ANSIBLE: social connectedness through a virtual world in an isolated Mars simulation mission

Long duration Mars exploration missions will face unprecedented sensory and social monotony that will impact various aspects of space crew performance as well as behavioral health. We describe ANSIBLE --- A Network of Social Interactions for Bilateral Life Enhancement. ANSIBLE is a virtual world ecosystem that implements evidence based strategies to promote social connectedness and psychological well-being, while being extendable and able to accommodate technical and environmental limitations of long duration flight. ANSIBLE was deployed in a twelve month isolation study to evaluate its capacity to maintain the social connection between participants in an isolated Mars simulation analog environment and their friends and family (FF) living outside the habitat. The Mars simulation exercise takes place at the Hawaii Space Exploration Analog and Simulation (HI-SEAS) facilities in Waimea, HI, where subjects undergo adjustments in physiology and lifestyle that are analogous to some of the challenges in a confined environment, with Mars exploration conditions (e.g. with communication latencies and blackouts, in close quarters, under restricted water and energy use). We present preliminary results suggesting that ANSIBLE can increase feelings of closeness as well as satisfaction with FF relationships over traditional methods, as well as promoting continued interactions of FF with crew members, closeness to humanity, decreased stress, and the perception that time is moving fast.