Content uploaded by Sung Park
Author content
All content in this area was uploaded by Sung Park on Nov 22, 2024
Content may be subject to copyright.
Framework for Virtual Agent Interaction to Foster Empathy
in Early School-Age Children
Jung Soyoung1, Kim Hansol2, Jeong Seo Yeon3, Lee JungMin2, Li Xuejiao3,
Yoo Young Jo3 and Sung Park4*
1Department of Family Welfare, Sangmyung University, Seoul, Republic of Korea
2Department of Human-Centered Artificial Intelligence, Sangmyung University, Seoul, Republic of Korea
3Department of Emotion Engineering, Sangmyung University, Seoul, Republic of Korea
4School of Data Science and Artificial Intelligence, Taejae University, Seoul, Republic of Korea
E-mail: 202231018@sangmyung.kr; sjp@taejae.ac.kr
Abstract—The negative socio-emotional development
of early school-age children (ages 4-7) resulting from
indiscriminate media use can be mitigated through
empathy development. This study designed an
educational framework to enhance empathy by
facilitating interactions between children and Virtual
Agents (VAs) using animations that children commonly
encounter. To assess the feasibility of this approach,
participatory design interviews were conducted with
parents (N = 5). The final design prototype offers an
effective method for empathy development by
combining emotional interaction through affect
recognition with empathy education.
Keywords—Empathy Development, Affective Virtual
Agents, Early Childhood Education, Social-Emotional
Learning, Participatory Design
I. INTRODUCTION
Advancements in media technology have detrimental
effects on the cognitive, emotional, and social aspects of
early school-age children, a pivotal period for empathy
development. Specifically, there is a decline in social-
emotional development, including diminished prosocial
behaviors and increased aggression. To address this, we
have reviewed empathy enhancement education, in which
parents traditionally perform a crucial role. Our findings
indicated that factors such as parental education level [1],
employment status [2] and emotional interaction style [3]
can make it difficult for parents to provide consistent
empathy education, adding to their caregiving burden.
Therefore, we designed an educational framework to
support children’s empathy development as an alternative
to the parental role.
II. METHODS
We applied the Double Diamond design methodology to
effectively emotional Virtual Agents (VA) as a
complement to parents. To provide effective empathy
education through the VA, we needed to understand how
children convey empathy to others through parental
influence (see Figure 1). Each relationship involves a
relational bond, with attachment forming the core rapport
between parent and child. Building on this foundation,
Children mimic their parents’ emotional and inductive
empathy education, enabling them to convey empathy to
others through role-taking and emotional communication.
.
Fig. 1. Model of empathy education process in relationship
among three parties.
A. Competitive Analysis
To establish a differentiated educational approach, we
analyzed existing empathy enhancement services [4], [5]
and identified two key insights regarding the necessity of
multi-directional communication and the relevance of
ecological validity in education. We adopted these insights
as central goals for our framework.
B. Target Users
We defined the target children users based on
developmental traits. The users are those capable of
mimicry and role-taking, essential for empathy [6], who
can express themselves through verbal and non-verbal
channels [7], [8] and who are at an age suitable for
applying inductive discipline, a method used to provide
empathy education [9]. This aligns with children aged 4-7
in the late preoperational stage, referred to as ‘early school-
age’.
C. Design and Evaluation
We developed detailed design mockups for early school-
age children, allowing them to learn empathy through
discussions and role-playing activities with VA, using
animations commonly encountered in daily life. To
evaluate the feasibility of this concept, we conducted
participatory design interviews with parents (N = 5) of
children. Direct interviews with the children were not
conducted due to cognitive and linguistic limitations,
attention span issues, and challenges in articulating precise
opinions [7]. Instead, we employed proxy interviews,
where parents provided observations of their children. The
interviews included structured questions covering the
study’s overall features and lasted approximately 1 to 1.5
hours. Except for interviewees with distance constraints,
all interviews were conducted face to face (see Table 1).
Table 1. Interviewee Characteristics
No
Interviewee
Child
Age
Interview
Method
1
Mother
7
Video
2
Mother
6
Face to Face
3
Mother
6
Face to Face
4
Father
7
Video
5
Mother
5
Face to Face
We refined the VA’s persona and interaction during the
children’s video viewing, and other relevant features.
Additionally, we used the child-specific Empathy Quotient
for Children (EQ-C) scale, which comprehensively
measures empathy skills, to assess empathy development
[10].
III. RESULTS
We designed an educational framework, which includes
a design mockup to enhance empathy development. The
purpose of this framework is to improve empathy in early
school-age children by using animations they encounter
daily, combined with other relevant empathy education
activities, including interaction with the VA.
A. Relationship between VA-Parents-Child
We established the relationships within the framework
between the VA, parents, and children as follows (see
Figure 2). The primary relationship between parents and
children involves parents providing media mediation.
Between the VA and parents lies in the VA’s role in
reducing caregiving burdens and providing educational
reports on the child’s educational progress. Finally, the key
relationship between the VA and the child is centered on
the VA offering emotional interaction and serving as a
positive role model for empathy development.
Fig. 2 Relationship between VA-Parents-Child
B. User Experience Flow
The user experience flow in the final application is as
follows (see Figure 3). First, users access the service and
choose either the default character (Teacher persona) or a
customizable character (Older Brother persona). They then
select a video for the empathy education session, which
they will view alongside the VA, who displays emotional
expressions aligned with animation’s context. After
watching, a discussion period follows, providing empathy
education through questions based on the EQ-C,
categorized into understanding others, emotional response,
and social relationship skills, and adapted according to the
questions. Following the discussion, used of each cycle,
the child’s empathy skills are assessed, with results either
resetting the cycle or producing a summary report for
parents.
Fig. 3. The Final User experience flow provided by
Framework
IV. CONCLUSION
This educational framework is significant in that it aims
to prevent socio-emotional decline through empathy skills,
enables emotional interactions beyond conventional smart
learning devices, and introduces VA to support parents
who face the burden of traditional caregiving roles. In
addition to enhancing empathy, it offers objective
assessments through EQ-C based questions. The final
design prototype provides emotional interaction and
empathy education, allowing children to gain indirect
experiences and engage in various interactions through
everyday animations. Future iterations could integrate
additional educational components, such as creativity and
character development, to implement a more
comprehensive developmental approach.
ACKNOWLEDGEMENT
This work was supported by the Basic Science Research
Program through the National Research Foundation of
Korea (NRF) funded by the Ministry of Education
(NRF2022R1I1A1A01066657).
REFERENCES
1. S. Chalasani, “The changing relationship between parents’
education and their time with children,” International
Journal of Time Use Research, vol. 4, no. 1, pp. 93–117,
2007.
2. M. A. Milkie, K. Nomaguchi, and S. Schieman, “Time
deficits with children: The link to parents’ mental and
physical health,” Soc Ment Health, vol. 9, no. 3, pp. 277–
295, 2019.
3. J. M. Gottman, L. F. Katz, and C. Hooven, Meta-
emotion: How families communicate emotionally.
Hillsdale, NJ, US: Lawrence Erlbaum Associates, Inc,
1997.
4. R. Muñoz, T. Barcelos, R. Noël, and S. Kreisel,
“Development of software that supports the improvement
of the empathy in children with autism spectrum disorder,”
in 2012 31st International Conference of the Chilean
Computer Science Society, IEEE, 2012, pp. 223–228.
5. E. Kim, Y. Lee, and S. Lee, “Development of a Tailored
Mobile Application to Improve Elementary School
Students’ Competencies of Emotional Awareness and
Empathy,” The Journal of Korean Association of Computer
Education, vol. 17, no. 3, pp. 85–99, 2014.
6. M. L. Hoffman, “Developmental synthesis of affect and
cognition and its implications for altruistic motivation.,”
Dev Psychol, vol. 11, no. 5, p. 607, 1975.
7. J. Piaget, “The theory of stages in cognitive development.,”
in Measurement and Piaget., New York, NY, US:
McGraw-Hill, 1971, pp. ix, 283–ix, 283.
8. L. S. VYGOTSKY, Mind in Society. Harvard University
Press, 1978. doi: 10.2307/j.ctvjf9vz4.
9. M. L. Hoffman, Empathy and Moral Development:
Implications for Caring and Justice. Cambridge:
Cambridge University Press, 2000. doi: DOI:
10.1017/CBO9780511805851.
10. B. Auyeung, S. Wheelwright, C. Allison, M. Atkinson, N.
Samarawickrema, and S. Baron-Cohen, “The children’s
empathy quotient and systemizing quotient: Sex differences
in typical development and in autism spectrum conditions,”
J Autism Dev Disord, vol. 39, pp. 1509–1521, 2009.
