Page 61 - DP-2-3

Design+ Design chatbot using activity theory

and culturally sensitive chatbots, thus advancing both the
theoretical and practical understanding of chatbot usability
and UX design.

3. Results
Engati also provides a chatbot template that comes
with well-defined conversation flows, including display
constructs such as messages, carousel cards, and more.
Figure 5 shows part of the design of the conversational flow
Figure 4. Satisfactory level of road sign chatbot
in the road sign chatbot using Engati. The chatbot answers
user queries with regards to Malaysian road signs.
This tension exists because it is not clear what The road sign chatbot is based on textual exchanges
rules or regulations determine who should be between the bot and the user. Figure 6 shows the starting
involved in determining the use of signs. conversation with the road sign chatbot. Note that the
• There are different perceptions of the activity chatbot offers three options to users: information on road
object, which reflects the heterogeneous nature of sign definitions, fun facts, and quizzes.
the subject group and their object (e.g., learning
as knowledge construction or information For quizzes regarding road sign definition, users choose
gathering). This causes secondary contradiction. the definition, and the chatbot will ask the color of road
• There is also tension between community sign that the user selects.
(society) and the object node (learning road Figure 7 (left) shows the fun fact functions that the road
signs). Society needs to have students’ drivers sign chatbot offers. Figure 7 (right) shows the quiz functions.
understand the road signs and use them effectively
whereas student drivers are generally interested 3.1. Evaluation
in passing examinations. This creates secondary The road sign chatbot was evaluated through a User
contribution. Acceptance Test involving 33 participants, all of whom were
(C) Potential quaternary contradictions between the use undergraduate students from Asia Pacific University (APU)
of chatbot and traditional ways of learning the road in Malaysia. The participants were aged between 18 and
signs 25 years and represented a variety of academic disciplines,
• There is also tension between the availability of including information technology, engineering, and
technology and the use of chatbot.
business. Before the evaluation, participants were provided
2.3.7. Contributions of the use of activity theory to with a brief demonstration and a user guide explaining
chatbot design how to interact with the chatbot. Each participant was then
instructed to use the chatbot freely for a minimum of 15 min
This study offers a novel contribution by operationalizing to explore its features, including road sign information,
activity theory not merely as an analytical tool but as a quizzes, and fun facts. Following their interaction,
practical design framework for developing a usable and participants completed an online survey through Google
engaging road sign learning chatbot. Unlike previous Forms, which included questions assessing satisfaction,
research that applies activity theory conceptually, this work user-friendliness, engagement, and interface design. The
systematically incorporates activity theory principles–such evaluation focused on collecting initial user feedback to
as the hierarchy of activities, contradictions, cultural- validate the usability and UX based on the activity theory-
historical context, and tool mediation–into the end-to-end driven design. No control group was utilized in this study,
chatbot design process. It extends prior chatbot usability as the primary objective was exploratory–to assess baseline
research by addressing the socio-cultural, historical, and perceptions of usability and satisfaction. Future work will
context-dependent factors that influence user interaction,
which are often overlooked by traditional usability models consider the inclusion of comparative control groups to
such as Nielsen’s heuristics or Norman’s design principles. further substantiate the findings.
Furthermore, this study proposes specific strategies for The respondents were asked:
handling contradictions in chatbot interactions to improve • Question 1: Are you satisfied with the road sign
UX and validates the approach through an empirical User chatbot?
Acceptance Test (UAT). By doing so, the paper provides • Question 2: Do you feel road sign chatbot is user-
new insights into designing context-aware, goal-directed, friendly for you?

Volume 2 Issue 3 (2025) 11 doi: 10.36922/DP025060009