The purpose of this study is to examine the communicative
relationship between older adults and conversational agents (CA), such as a Google Home
Mini, to understand if and how interaction with AI-based voice technology affects
perceptions, technological adoption, and, ultimately, human-machine communicative behaviors.
Using the Communication Accommodation Theory (CAT) framework (Gallois & Giles, 2015),
and the categorical schema as outlined in the Unified Theory of Adoption and Utilization of
Technology (UTAUT) model (Venkatesh et al., 2003) of technology acceptance, we qualitatively
assess the relationship between expectations for use and ongoing / post-interaction user
attitudes. CAT focuses on the adjustments we make in our perceptions of and engagement in
communicative behaviors. In other words, we enter into communicative situations with
intentions and motivations derived from antecedent socio-historical context in mind. This
squares with what the UTAUT model details as influencers of technological adoption and use:
performance expectancy, effort expectancy, social influence, and facilitating conditions
(Venkatesh et al., 2003). We use these constructs as a coding guideline to index data
scraped from the Mini, and collected from surveys, interview transcripts, user journals, and
field notes throughout a 10-week study. Historically, CAT is applied to human-human
communication exchanges. As the theory posits that interpersonal relationships can and will
influence motivations or intentions for dyadic communication, this makes sense. However, we
argue that as AI-based voice technologies become more sophisticated as voice assistants
enter our intimate spaces, the application of CAT to the human-machine communicative
relationship is warranted.