While the ultimate goal of natural-language based Human-Robot Interaction (HRI) may be free-form, mixed-initiative dialogue,social robots deployed in the near future will likely primarily engage in wakeword-driven interaction, in which users’ commands are prefaced by a wakeword such as “Hey, Robot.” This style of interaction helps to allay user privacy concerns, as the robot’s full speech recognition module need not be employed until the target wakeword is used. Unfortunately, there are a number of concerns in the popular media surrounding this style of interaction, with consumers fearing that it is training users (in particular,children) to be rude towards technology, and by extension, rude towards other humans. In this paper, we present a study that demonstrates how an alternate style of wakeword, i.e., “Excuse me, Robot” may allay this concern, by priming users to phrase commands as Indirect Speech Acts
more »
« less
Fresh Start: Encouraging Politeness in Wakeword-Driven Human-Robot Interaction
Deployed social robots are increasingly relying on wakeword-based interaction, where interactions are human-initiated by a wakeword like “Hey Jibo”. While wakewords help to increase speech recognition accuracy and ensure privacy, there is concern that wakeword-driven interaction could encourage impolite behavior because wakeword-driven speech is typically phrased as commands. To address these concerns, companies have sought to use wake- word design to encourage interactant politeness, through wakewords like “⟨Name⟩, please”. But while this solution is intended to encourage people to use more “polite words”, researchers have found that these wakeword designs actually decrease interactant politeness in text-based communication, and that other wakeword designs could better encourage politeness by priming users to use Indirect Speech Acts. Yet there has been no previous research to directly compare these wakewords designs in in-person, voice-based human-robot interaction experiments, and previous in-person HRI studies could not effectively study carryover of wakeword-driven politeness and impoliteness into human-human interactions. In this work, we conceptually reproduced these previous studies (n=69) to assess how the wakewords “Hey ⟨Name⟩”, “Excuse me ⟨Name⟩”, and “⟨Name⟩, please” impact robot-directed and human-directed politeness. Our results demonstrate the ways that different types of linguistic priming interact in nuanced ways to induce different types of robot-directed and human-directed politeness.
more »
« less
- NSF-PAR ID:
- 10403579
- Date Published:
- Journal Name:
- ACM/IEEE International Conference on Human-Robot Interaction
- Page Range / eLocation ID:
- 112 to 121
- Format(s):
- Medium: X
- Sponsoring Org:
- National Science Foundation
More Like this
-
-
For enhanced performance and privacy, companies deploying voice-activated technologies such as virtual assistants and robots are increasingly tending toward designs in which technologies only begin attending to speech once a specified wakeword is heard. Due to concerns that interactions with such technologies could lead users, especially children, to develop impolite habits, some companies have begun to develop use modes in which interactants are required to use ostensibly polite wakewords such as " Please''. In this paper, we argue that these ``please-centering'' wakewords are likely to backfire and actually discourage polite interactions due to the particular types of lexical and syntactic priming induced by those wakewords. We then present the results of a human-subject experiment (n=90) that validates those claims.more » « less
-
Principles from human-human physical interaction may be necessary to design more intuitive and seamless robotic devices to aid human movement. Previous studies have shown that light touch can aid balance and that haptic communication can improve performance of physical tasks, but the effects of touch between two humans on walking balance has not been previously characterized. This study examines physical interaction between two persons when one person aids another in performing a beam-walking task. 12 pairs of healthy young adults held a force sensor with one hand while one person walked on a narrow balance beam (2 cm wide x 3.7 m long) and the other person walked overground by their side. We compare balance performance during partnered vs. solo beam-walking to examine the effects of haptic interaction, and we compare hand interaction mechanics during partnered beam-walking vs. overground walking to examine how the interaction aided balance. While holding the hand of a partner, participants were able to walk further on the beam without falling, reduce lateral sway, and decrease angular momentum in the frontal plane. We measured small hand force magnitudes (mean of 2.2 N laterally and 3.4 N vertically) that created opposing torque components about the beam axis and calculated the interaction torque, the overlapping opposing torque that does not contribute to motion of the beam-walker’s body. We found higher interaction torque magnitudes during partnered beam-walking vs . partnered overground walking, and correlation between interaction torque magnitude and reductions in lateral sway. To gain insight into feasible controller designs to emulate human-human physical interactions for aiding walking balance, we modeled the relationship between each torque component and motion of the beam-walker’s body as a mass-spring-damper system. Our model results show opposite types of mechanical elements (active vs . passive) for the two torque components. Our results demonstrate that hand interactions aid balance during partnered beam-walking by creating opposing torques that primarily serve haptic communication, and our model of the torques suggest control parameters for implementing human-human balance aid in human-robot interactions.more » « less
-
null (Ed.)This work has developed an iteratively refined understanding of participants’ natural perceptions and responses to unmanned aerial vehicle (UAV) flight paths, or gestures. This includes both what they believe the UAV is trying to communicate to them, in addition to how they expect to respond through physical action. Previous work in this area has focused on eliciting gestures from participants to communicate specific states, or leveraging gestures that are observed in the world rather than on understanding what the participants believe is being communicated and how they would respond. This work investigates previous gestures either created or categorized by participants to understand the perceived content of their communication or expected response, through categories created by participant free responses and confirmed through forced choice testing. The human-robot interaction community can leverage this work to better understand how people perceive UAV flight paths, inform future designs for non-anthropomorphic robot communications, and apply lessons learned to elicit informative labels from people who may or may not be operating the vehicle. We found that the Negative Attitudes towards Robots Scale (NARS) can be a good indicator of how we can expect a person to react to a robot. Recommendations are also provided to use motion approaching/retreating from a person to encourage following, perpendicular to their field of view for blocking, and to use either no motion or large altitude changes to encourage viewing.more » « less
-
The Global Biodiversity Information Facility (GBIF 2022a) has indexed more than 2 billion occurrence records from 70,147 datasets. These datasets often include "hidden" biotic interaction data because biodiversity communities use the Darwin Core standard (DwC, Wieczorek et al. 2012) in different ways to document biotic interactions. In this study, we extracted biotic interactions from GBIF data using an approach similar to that employed in the Global Biotic Interactions (GloBI; Poelen et al. 2014) and summarized the results. Here we aim to present an estimation of the interaction data available in GBIF, showing that biotic interaction claims can be automatically found and extracted from GBIF. Our results suggest that much can be gained by an increased focus on development of tools that help to index and curate biotic interaction data in existing datasets. Combined with data standardization and best practices for sharing biotic interactions, such as the initiative on plant-pollinators interaction (Salim 2022), this approach can rapidly contribute to and meet open data principles (Wilkinson 2016). We used Preston (Elliott et al. 2020), open-source software that versions biodiversity datasets, to copy all GBIF-indexed datasets. The biodiversity data graph version (Poelen 2020) of the GBIF-indexed datasets used during this study contains 58,504 datasets in Darwin Core Archive (DwC-A) format, totaling 574,715,196 records. After retrieval and verification, the datasets were processed using Elton. Elton extracts biotic interaction data and supports 20+ existing file formats, including various types of data elements in DwC records. Elton also helps align interaction claims (e.g., host of, parasite of, associated with) to the Relations Ontology (RO, Mungall 2022), making it easier to discover datasets across a heterogeneous collection of datasets. Using specific mapping between interaction claims found in the DwC records to the terms in RO*1, Elton found 30,167,984 potential records (with non-empty values for the scanned DwC terms) and 15,248,478 records with recognized interaction types. Taxonomic name validation was performed using Nomer, which maps input names to names found in a variety of taxonomic catalogs. We only considered an interaction record valid where the interaction type could be mapped to a term in RO and where Nomer found a valid name for source and target taxa. Based on the workflow described in Fig. 1, we found 7,947,822 interaction records (52% of the potential interactions). Most of them were generic interactions ( interacts_ with , 87.5%), but the remaining 12.5% (993,477 records) included host-parasite and plant-animal interactions. The majority of the interactions records found involved plants (78%), animals (14%) and fungi (6%). In conclusion, there are many biotic interactions embedded in existing datasets registered in large biodiversity data indexers and aggregators like iDigBio, GBIF, and BioCASE. We exposed these biotic interaction claims using the combined functionality of biodiversity data tools Elton (for interaction data extraction), Preston (for reliable dataset tracking) and Nomer (for taxonomic name alignment). Nonetheless, the development of new vocabularies, standards and best practice guides would facilitate aggregation of interaction data, including the diversification of the GBIF data model (GBIF 2022b) for sharing biodiversity data beyond occurrences data. That is the aim of the TDWG Interest Group on Biological Interactions Data (TDWG 2022).more » « less
- Free Publicly Accessible Full Text
-
Accepted Manuscript1.0
- Conference Paper:
- https://doi.org/10.1145/3568162.3576959
