How Parents and Children Interact with Digital Assistants in the Home: An Exploratory Study

• In Event: Poster Session 08
  In Poster Session: PS 08 Topic - Technology, Media & Child Development

Abstract

Currently, 46% of adults in the U.S.A report using a digital assistant (DA) (Pew, 2017) and Amazon alone reports selling over 100 million smart speakers. Despite this level of use, there is little information to guide the study of how DAs are used in homes. As such, we conducted an exploratory study to identify how parents and children use voice-enabled DAs in the home (i.e., Amazon Alexa, Google Assistant, Apple Siri.). The goal is to create a list of common parental and child DA interactions to guide research.
Method. Participants completed an online survey during the Covid-19 lockdowns, which may have increased the frequency of digital assistant use. Closed-ended questions addressed DA ownership. Open-ended questions addressed how parents and children use DAs (e.g., knowledge search, media/entertainment), the specific questions parents and children asked of DAs (e.g., nature, sports), and the nature of the relationship children have with DAs (e.g., how they interact with it).
Participants were parents (n = 50, M = 34.2 years, SD = 5.1), with at least one child age 4-8 (M = 5 years, SD = 2.0), from 10 countries (61% UK). Majority of DAs owned were Amazon Alexa devices (47%), followed by Google Assistants (23%), and finally Apple’s Siri (19%). For ownership duration, over half of our respondents (56%) owned their devices for more than a year, 28% +2 years. A majority of participants used DAs on a daily basis, multiple times a day (42%) or at least 1-2 times a day (34%).
Results. Parents and children use DAs for similar tasks, but in different amounts (see Table 1). Parents most frequently used DAs to conduct knowledge search, play different kinds of media, and check the weather, X2 (11, 226) = 229.79, p < .001. In contrast, children primarily used digital assistants for media and play, X2 (8, 174) = 145.45, p < .001. Parents’ and children’s DA questions were more differentiated, parents most frequently asked DAs about Media/Entertainment, Science, Currency conversion, and Product reviews, X2 (13, 69) = 73.72, p < .001. In contrast, children asked questions about nature (e.g., order of the colors in the rainbow) and idiosyncratic general knowledge questions (how many bricks in my house), X2 (11,77) = 58.77, p < .001.
The results also provided insights into children’s interactions with digital assistants. Nearly half of parents reported that children treated the DA in a transactional manner but a third reported their child treating the DA as a Friend, X2 (8, 174) = 145.45, p < .001 (see table 2). Further, parents reported children take specific tones with DAs, ranging from polite (12.5%) to shouting (18.8%), with many noting their child’s tone with the DA mimicked their own (25%).
These are some of the first insights into how families use and interact with DAs, devices that are increasingly common and that children learn from and even form connections. The behaviors identified here will serve as a starting point for a larger study of children’s Theory of Artificial Minds (ToAM).

Authors

• Nandini Asavari Bharadwaj, McGill University
  Presenting Author

• Adam Kenneth Dubé, McGill University
  Non-Presenting Author

• Victoria Talwar, McGill University
  Non-Presenting Author

• Elizabeth Patitsas, McGill University
  Non-Presenting Author