Voice User Interfaces (VUIs) are becoming ubiquitously available, being embedded both into everyday mobility via smartphones, and into the life of the home via 'assistant' devices. Yet, exactly how users of such devices practically thread that use into their everyday social interactions remains underexplored. By collecting and studying audio data from month-long deployments of the Amazon Echo in participants' homes-informed by ethnomethodology and conversation analysis-our study documents the methodical practices of VUI users, and how that use is accomplished in the complex social life of the home. Data we present shows how the device is made accountable to and embedded into conversational settings like family dinners where various simultaneous activities are being achieved. We discuss how the VUI is finely coordinated with the sequential organisation of talk. Finally, we locate implications for the accountability of VUI interaction, request and response design, and raise conceptual challenges to the notion of designing 'conversational' interfaces.

We present the findings from a study of how people interleave mobile phone use with conversation in pubs. Our findings, informed by ethnomethodology and conversation analysis, unpack the interactional methods through which groups of people in pubs occasioned, sustained, and disengaged from mobile device use during conversation with friends. Fundamentally, the work that is done consists of various methods of accounting for mobile device use, and displaying involvement in social interaction while the device is used. We highlight multiple examples of the nuanced ways in which interleaving is problematic in interaction, and relate our findings to the CSCW and HCI literature on collocated interaction. We conclude by considering avenues for future research, and discuss how we may support or disrupt interleaving practices through design to overcome the highlighted interactional troubles.

In this paper we unpack the use of conversational agents, or so-called intelligent personal assistants (IPAs), in multiparty conversation amongst a group of friends while they are socialising in a café. IPAs such as Siri or Google Now can be found on a large proportion of personal smartphones and tablets, and are promoted as 'natural language' interfaces. The question we pursue here is how they are actually drawn upon in conversational practice? In our work we examine the use of these IPAs in a mundane and commonplace setting and employ an ethnomethodological perspective to draw out the character of the IPA-use in conversation. Additionally, we highlight a number of nuanced practicalities of their use in multi-party settings. By providing a depiction of the nature and methodical practice of their use, we are able to contribute our findings to the design of IPAs.

The Think Aloud Protocol (TAP) is a verbalisation technique widely employed in HCI user studies to give insight into user experience, yet little work has explored the impact that TAPs have on participants during user studies. This paper utilises a brain sensing technique, fNIRS, to observe the e↵ect that TAPs have on participants. Functional Near-Infrared Spectroscopy (fNIRS) is a brain sensing technology that o↵ers the potential to provide continuous, detailed insight into brain activity, enabling an objective view of cognitive processes during complex tasks. Participants were asked to perform a mathematical task under 4 conditions: nonsense verbalisations, passive concurrent think aloud protocol, invasive concurrent think aloud protocol, and a baseline of silence. Subjective ratings and performance measures were collected during the study. Our results provide a novel view into the e↵ect that di↵erent forms of verbalisation have on workload during tasks. Further, the results provide a means for estimating the e↵ect of spoken artefacts when measuring workload, which is another step towards our goal of proactively involving fNIRS analysis in ecologically valid user studies.

Drawing from Conversation Analysis (CA), we examine how the orientation towards progressivity in talk-keeping things movingmight help us better understand and design for voice interactions. We introduce progressivity by surveying its explication in CA, and then look at how a strong preference for progressivity in conversation works out practically in sequences of voice interaction recorded in people's homes. Following Stivers and Robinson's work on progressivity, we find our data shows: how non-answer responses impede progress; how accounts offered for non-answer responses can lead to recovery; how participants work to receive answers; and how, ultimately, moving the interaction forwards does not necessarily involve a fitted answer, but other kinds of responses as well. We discuss the wider potential of applying progressivity to evaluate and understand voice interactions, and consider what designers of voice experiences might do to design for progressivity. Our contribution is a demonstration of the progressivity principle and its interactional features, which also points towards the need for specific kinds of future developments in speech technology. CCS CONCEPTS • Human-centered computing → Natural language interfaces.