IEEE Robotics & Automation Magazine - June 2019 - 69

robot, and each other. In the interaction between P7 and P18,
described in the "Qualitative Results" section, P7 was able to
engage in helping behavior, something she had identified as
an important part of her interactions with other residents.
Her ability to accurately use the wake phrase when assisting
others was notable, because she had previously struggled to
do so when interacting with the social robot. It is possible that
the more active helping or teaching role enabled her to
remember how to use the technology, as Chang and
Šabanović also report in their observations [6]. P7 was able to
reaffirm how to use the social robot and to use it as a tool to
engage with P18. This supports how the technology is accepted by the social network and how it influences the community's perception and use of technology [7].
In participants' descriptions of interactions with each
other, the pleasantries theme extended to the social robot as
well. P3 tried several different types of conversation starters
with the agent: "Hi bot! How are you? How are you today?
Did you do anything bad last night? Smile. Be happy today!"
The researchers regularly observed questions about the agent's
preferences and traits, with very few questions about its function ("what are you like?" versus "what do you do?"). The residents engaged in pleasantries with and addressed Jibo as they
would a community member, demonstrating one way they
were accepting of the robot. Similar observations were made
by Kidd et al. [7].
In accordance with the participatory, human-centered
design lens of this study, the residents were asked their opinions of their interactions with the social robot and its impact
in their community, along with their ideas and needs for
future robot development. Describing their range of relationships, older adults focused on common areas of concern
when designing technologies for their peers, including their
difficulties operating the technology. However, residents
vocalized, with a sense of pride, that they were able to learn
how to use the robot. They expressed the desire for the social
robot to be in a community space for all residents to interact
with it-and that they also might be able to spend personal
time with it. The robot was seen as a communication tool that
impacted community interactions, a companion, and a way to
learn how to communicate better with residents-a finding
also reported in [20]. In addition, we paid attention to any
comments from the residents that negatively reflected the
social robot's presence toward the community's social connectedness. As we had expected, residents shared concerns
around the latest communication technologies, revealing the
group's thoughts surrounding security and privacy, fear, and
collective worry for technology's effects on society (mostly in
the low-openness group).
Direct Experience Matters With Social Robots
Through understanding the participants' social robot design
preferences, the initial/final comparison revealed how experience can change the way users conceptualize a technology.
Each cluster changed over time in preferences for the technology, informed by experiences with the social robot. This was

possible because the robot's presence was prompting residents
to engage with each other more face to face, and they did not
see the value of using the social robot to connect with others
remotely. In this community, the presence of the robot and its
natural verbal interaction modality promoted older adults'
connection with other residents more than with the robot
itself. This finding is interesting and should be further
explored, as it runs counter to the general perception that new
forms of immersive or interactive technology will isolate people from their fellow humans. The older adults' experience
with the robot prompted them to be more concerned about
the technology but also to be empowered to discuss their
design recommendations for the social robot.
Methods Reflections
Because the older-adult population includes people with a
wide range of cognitive and physical abilities, not all humancentered design tools are appropriate for the older adult. The
researchers adopted a flexible approach to senior citizens'
participation to respect their autonomy as well as acknowledge their physical and cognitive needs [15]. The results
demonstrate that experiencing an agent can increase preferences for social robot technology functions, but it also
revealed security and privacy concerns, fear, and collective
worry about social robot technology (mostly in the lowopenness group). To better address varied experiences and
preferences for the technology, researchers can further
explore how to incorporate these concerns into design
approaches to empower participants to critically evaluate
their technology choices. Researchers' presence through an
embedded ethnographic approach provided a deeper context
for the formal data collection and encouragement for participants to share more information.
Implications and Future Work
Long-term studies of robots in the real world are necessary
to understand how these technologies will be incorporated
into society. This work revealed several interesting areas for
future exploration, including the investigation of teaching or
mentoring others in the successful use of new technology
among older adults. The low-, mid-, and high-openness categories could be utilized in future studies to better understand
how each group responds to different approaches to technology implementation; simple assessments would enable customized approaches for each group. Our holistic approach
provided a broad perspective on how the social robot
impacted the community through group interactions and
individual conversations.
Future studies could compare individual relationships with
a robot to robot dynamics in a group setting. Additional studies need to investigate preference trends found with each cluster over time, how best to mediate technology concerns, and
the optimal location of the social robot and its effects on community connectedness.
Throughout this process, the participatory design
approach highlighted the participants as key research
JUNE 2019

IEEE ROBOTICS & AUTOMATION MAGAZINE

IEEE Robotics & Automation Magazine - June 2019

Table of Contents for the Digital Edition of IEEE Robotics & Automation Magazine - June 2019