384
V. Vouloutsi et al.
Experiment 3 - The Selfie Task with the Robot. The final goal is to
deploy this setup in public. In this pilot study, we conducted the experiments
in a controlled environment (the XIM) to assess the interaction dynamics with
the minimum amount of participants required to form a group (two individuals).
The experiment began when participants entered the XIM where the robot was
located in groups of two. We did not inform them about the robot’s motivation
for the interaction, nor we gave them specific instructions on how to interact
with the robot, aside from informing them that the interaction would be nonverbal. Each pair was randomly assigned to either the extroverted or introverted
robot.
We considered a successful selfie-taking when participants placed themselves
behind the robot, in the camera’s frame. The interaction was over after a successful selfie-taking followed by a salutation by the robot, or after 10 min had
passed. Upon completion of the task, we asked participants to complete a survey
on a tablet that contained the following information: demographics, the participant’s personality and the robot’s perceived personality (10-item Big Five questionnaire), the full Godspeed questionnaire, information about the participant’s
selfie-taking habits and preferences, and questions that evaluated the interaction
and taking a selfie with the robot. Finally, participants were debriefed and also
given the option to receive a copy of their selfie taken by the robot.
The selfie-taking habits were measured by “How often do you take selfies?”
(Response: Never, Daily, Weekly, Monthly) and “I like taking selfies” on a 5point Likert scale. To evaluate the interaction, participants also evaluated in a
5-point Likert scale the following statements: “Taking a selfie with the robot
was fun”, “Taking a selfie with the robot was engaging”, “Taking a selfie with
the robot was easy”, “Taking a selfie with the robot was annoying”, “Taking a
selfie with the robot was natural”, “I liked taking a selfie with the robot”, “I
was uncomfortable taking a selfie with the robot”. Additionally, we logged the
length of each selfie interaction and the total amount of successful selfie-taking
per condition and recorded the interactions. Examples of individual and group
selfies can be found in Fig. 2. The experiments were conducted in English.
3 Results
3.1 Experiment 1 - Selfie Taking in Humans
Initially, we asked random participants to take a selfie with a group of people
that was unknown to them. From an early stage, it became apparent that the
design of the experimental task created scepticism and distrust when applied to
HHI, due to the rarity of this occurrence in actual social contexts. As it was
difficult to find participants who would agree to take a selfie, we modulated the
experiment to facilitate data collection by approaching a group of people and
taking a member of the group aside to be the “selfie-taker”. The task began
when the “selfie-taker” returned to the group to take the selfie with the other
members. In total, we conducted 6 selfie-taking interactions, two out of which
were with a group of strangers and four with a group of friends (with a total of
V. Vouloutsi et al.
Experiment 3 - The Selfie Task with the Robot. The final goal is to
deploy this setup in public. In this pilot study, we conducted the experiments
in a controlled environment (the XIM) to assess the interaction dynamics with
the minimum amount of participants required to form a group (two individuals).
The experiment began when participants entered the XIM where the robot was
located in groups of two. We did not inform them about the robot’s motivation
for the interaction, nor we gave them specific instructions on how to interact
with the robot, aside from informing them that the interaction would be nonverbal. Each pair was randomly assigned to either the extroverted or introverted
robot.
We considered a successful selfie-taking when participants placed themselves
behind the robot, in the camera’s frame. The interaction was over after a successful selfie-taking followed by a salutation by the robot, or after 10 min had
passed. Upon completion of the task, we asked participants to complete a survey
on a tablet that contained the following information: demographics, the participant’s personality and the robot’s perceived personality (10-item Big Five questionnaire), the full Godspeed questionnaire, information about the participant’s
selfie-taking habits and preferences, and questions that evaluated the interaction
and taking a selfie with the robot. Finally, participants were debriefed and also
given the option to receive a copy of their selfie taken by the robot.
The selfie-taking habits were measured by “How often do you take selfies?”
(Response: Never, Daily, Weekly, Monthly) and “I like taking selfies” on a 5point Likert scale. To evaluate the interaction, participants also evaluated in a
5-point Likert scale the following statements: “Taking a selfie with the robot
was fun”, “Taking a selfie with the robot was engaging”, “Taking a selfie with
the robot was easy”, “Taking a selfie with the robot was annoying”, “Taking a
selfie with the robot was natural”, “I liked taking a selfie with the robot”, “I
was uncomfortable taking a selfie with the robot”. Additionally, we logged the
length of each selfie interaction and the total amount of successful selfie-taking
per condition and recorded the interactions. Examples of individual and group
selfies can be found in Fig. 2. The experiments were conducted in English.
3 Results
3.1 Experiment 1 - Selfie Taking in Humans
Initially, we asked random participants to take a selfie with a group of people
that was unknown to them. From an early stage, it became apparent that the
design of the experimental task created scepticism and distrust when applied to
HHI, due to the rarity of this occurrence in actual social contexts. As it was
difficult to find participants who would agree to take a selfie, we modulated the
experiment to facilitate data collection by approaching a group of people and
taking a member of the group aside to be the “selfie-taker”. The task began
when the “selfie-taker” returned to the group to take the selfie with the other
members. In total, we conducted 6 selfie-taking interactions, two out of which
were with a group of strangers and four with a group of friends (with a total of