Content uploaded by Julita Vassileva
Author content
All content in this area was uploaded by Julita Vassileva on Dec 11, 2024
Content may be subject to copyright.
Engagement Strategies in a Peer-quizzing Game: Investigating
Student Interactions and Powergaming
Nafisul Kiron, Mehnuma Tabassum Omar and Julita Vassileva
University of Saskatchewan, Saskatoon, Canada
ni.kiron@usask.ca
mehnuma.omar@usask.ca
julita.vassileva@usask.ca
Abstract: Educational games are a popular means of engaging students in learning activities. However, students in game-
based learning environments often engage in powergaming to reap undeserved rewards and spoil the experience of their
peers. This may happen when the students pretend to engage in game activities or collude with other students, exploiting
the game settings and rules to maximize their points. In this explorative study, we investigate powergaming in three settings
of an asynchronous online multiplayer peer-quizzing game in a blended learning setting – a first-year programming class in
a Canadian university. We designed three experimental settings with three versions of an education game which allowed
different levels of powergaming. The between-subject study involved three experimental groups: Group 1 used a game
version where they received weekly performance feedback and tips on how to improve their performance, Group 2 used a
game version with access to an existing resource bank, allowing them to maximize the number if their activities in the game,
and Group 3 served as a control group with no additional interventions. The research aimed to investigate 1) the association
between the power gamers' activity, their grades and their preference to work alone or in a group, 2) the types of activities
(type of quiz questions) most used by power gamers, and 3) which game setting was the most conducive to power gaming.
The results show that better grades are not associated with higher activity levels in the game. Students who engaged in more
diverse types of game activities had better learning outcomes The control Group 3 had the highest average grades. As we
expected, Group 2 engaged in the highest number of activities, esp. in creating questions, a form of powergaming. The
qualitative results showed that contrary to our assumption the powergamers in Group 2 was not harmful, because it created
a rich set of resources in the game and fostered student engagement in the game.
Keywords: Serious Games, Game-based Learning, Peer-quizzing, Engagement
1. Introduction
Game-based learning (GBL) has emerged as a promising pedagogical approach that harnesses the engaging and
motivating power of games to facilitate learning (Marques and Pombo, 2021; Riet, 2019). By incorporating game
elements, mechanics, and principles into educational contexts, GBL aims to create immersive, interactive, and
enjoyable learning experiences that promote active participation, problem-solving, and knowledge acquisition
(Bösche and Kattner, 2011). The growing interest in GBL in blended learning is driven by the recognition that
traditional, passive lecture-based learning practices often fail to capture and maintain learners' attention,
resulting in poor retention rates (Calle et al., 2019; Soo and Lee, 2022).
In contrast, well-designed educational games have the potential to tap into learners' intrinsic motivations,
reduce lapses in attention, and maximize knowledge retention and prevent learners from blanking out during
exams (Calle et al., 2019; Iverson, 2019). The success of GBL depends on various factors, including the visual and
audio impact of the game, the appropriate blend of challenges, rewards, learning content, and assessment units,
and the overall narrative structure that binds these elements together (Stefan et al., 2022; Wang et al., 2016).
In this paper, we study the gameplay behaviour of students playing three versions of an online asynchronous
peer-quizzing game in a blended learning environment during one academic term. The game developed in our
lab is called ToQ (Towers of Questions) (Kiron et al., 2019; Kiron and Vassileva, 2018). The game offers three
types of quiz questions: multiple-choice questions (MCQs), true/false questions, and short answer questions.
Students can create quiz questions, answer questions created by their peers, and review the questions they have
already solved. The game rewards are points that can be earned by creating quiz questions and answering them
correctly.
We wanted to explore if variations of the game setup that could make it easier to power-game would lead to
increase in powergaming. We also wanted to explore if there is any relationship between the preferred studying
mode (alone or group) of students and their activities in the game, including their preferred question types. The
rationale for exploring this relationship is constructing the three types of question (MCQ, T/F, and Short answer)
demands distinct thinking processes. For instance, in a multiple-choice question, the student needs to create
Nafisul Kiron, Mehnuma Tabassum Omar and Julita Vassileva
effective distractors. Students need to create T/F questions that are not easily apparent to their classmates.
Short answer questions require more effort to evaluate the possible answers.
The participants were 1st year students in an Introduction to Programming course, who were randomly assigned
to three groups each using a different version of the game. The first two groups, Groups 1 and 2, played versions
of the game that included additional support: the version played by Group 1 included instructions and tips on
creating good quiz questions and the version used by Group 2 provided a question bank from where students
could choose and posting questions as their own. Group 3 played the game without additional supports.
The study has three primary objectives. The first objective is to explore which experimental group shows signs
of hyperactivity or powergaming. The second objective is to analyse the relationship between the students’
preference for learning (alone or in group), their in-game actions, and academic performance, with particular
attention to the most active students who could be powergamers. The in-game actions include asking or
answering questions and browsing the solved questions and the question bank (for Group 2). The third objective
is to identify the most frequent categories of quiz questions (Multiple Choice Questions, True/False, and Short
Answers) used by power gamers. Understanding the preferred question types will shed light on power gamers'
learning strategies and preferences. By addressing these three objectives, this paper aims to contribute to the
understanding of powergamers' behaviour, how effectively they learn and if they harm or benefit their peers in
the context of a peer-quizzing game. The findings could inform the design of educational games.
2. Background
Game-based learning (GBL) has gained significant attention in recent years as an innovative pedagogical
approach in higher education. By incorporating game elements and principles into educational contexts, GBL
aims to create engaging, interactive, and motivating learning experiences for students (Guerrero-Quiñonez et
al., 2023; Islam, 2017). The use of gamification techniques, such as challenges, rewards, competitions, and
immediate feedback, can enhance learning effectiveness and promote active participation, collaboration, and
problem-solving skills (Ceccarini and Prandi, 2022; Islam, 2017).
One of the key components of GBL is the use of rewards to motivate and reinforce desired behaviours and
learning outcomes (Nipo et al., 2023; Perbawa and Rapiyanta, 2024). Rewards in educational games can take
various forms, such as points, badges, leaderboards, virtual currencies, or unlockable content (Dong, 2023;
Featherstone, 2022). However, the use of rewards in GBL is not without its potential drawbacks and negative
consequences. An overemphasis on extrinsic rewards can undermine learners' intrinsic motivation, shifting their
focus from the natural value of learning to the pursuit of external incentives (Kaldarova et al., 2023; Klit et al.,
2020). This can lead to a superficial engagement with the educational content, where learners prioritize the
acquisition of rewards over deep understanding and meaningful learning (Yu and Tsuei, 2022).
Another concern in GBL is the potential for students to exploit the game mechanics and reward systems in ways
that subvert the intended learning objectives (Dong, 2023). Some learners may find ways to "game the system"
by discovering loopholes, cheats, or strategies that allow them to earn rewards without genuinely engaging with
the educational content. The exploitation of GBL by some students can have negative ripple effects on the
learning experiences of their peers (Klit et al., 2020). When a significant portion of the learner population
engages in exploitative behaviours, it can disrupt the collaborative and supportive learning environment that
GBL aims to foster (Kaldarova et al., 2023). Other students may feel pressured to adopt similar strategies to keep
up with their classmates, leading to a breakdown in the social dynamics and a loss of trust and cooperation
within the learning community.
To mitigate these issues, educators must carefully design and implement GBL activities, striking a balance
between intrinsic and extrinsic motivation (Fu and Yu, 2006). This can involve using a variety of reward types,
such as social recognition and meaningful feedback, alongside tangible rewards (Caeiro-Rodriguez et al., 2021).
Additionally, educators should establish clear guidelines and expectations for student behaviour, emphasizing
the importance of academic integrity and genuine engagement with the learning material.
Game-based learning has the potential to revolutionize higher education by creating engaging and interactive
learning experiences. The use of rewards can be a powerful motivational tool. Still, it is essential to consider the
potential negative consequences of an overreliance on extrinsic rewards and the exploitation of the system by
some students. By carefully designing GBL activities and fostering a culture of intrinsic motivation and academic
integrity, educators can harness the benefits of game-based learning while minimizing its drawbacks, such as
powergaming.
Nafisul Kiron, Mehnuma Tabassum Omar and Julita Vassileva
3. Experiment Design
This section provides an overview of our research methodology, including details on how we recruited the
participants, designed the experiment, and collected the data. This study was granted approval by the
behavioural ethics board of our university, identified as BEH ID-101. Students were recruited from a first-year
programming language course. Participation in the study was voluntary, and as an incentive, students who
actively participated in the game had the opportunity to earn up to five percent of their final grade as
participation marks in the class. Students who decided not to be a part of the study had the five percent grade
added to their final exam’s weight.
After giving their informed consent, the students were directed to a pre-survey, which collected their
information about their study preference. After removing those who opted out of the study or dropped the
class, n=128 students participated in the study. The students were divided into three groups at random, with
each group playing a different game version. The first group, Group1 (n=41), received feedback and suggestions
from moderators. The second group (n=47) had access to a pre-existing question bank, and the third group,
Group 3 (n=40), served as the control group, playing the game without support. The uneven number of students
in the groups is a result of some students dropping out of the game or the course during the study.
The students were given pseudonyms to participate anonymously in a peer-quizzing game themed as a tower
defence game. The game's theme revolves around students creating quiz questions (towers) based on the course
material they are studying, answering the questions (attacking the towers) created by their peers and browsing
the already answered questions (conquered towers). The game rewards are points that can be earned by
creating quiz questions and by answering questions correctly. The students played the game alone and
asynchronously using their computer's web browser. They logged in to the game using the credentials provided
to them after signing up the consent form. There was no chat, or another communication channel provided for
them in the game environment. The game interface featured segments tailored for the experimental groups.
When students logged in to the game, they saw their version of the dashboard. For Group 1, weekly feedback
was shown to the students including tips for creating good questions. A link to an existing question bank was
shown only to Group 2. Group 3 was shown the game’s interface without any extra support.
To gather comprehensive data for the study, we collected information from various sources. In addition to the
pre- and post-study survey responses, we obtained gameplay data for each student. This data included the
number of questions each student created and answered throughout the game. Furthermore, we collected a
detailed log of all the events that occurred within the game. This log contained the specific questions students
created, answered, and viewed, along with precise timestamps indicating when they signed in and out of the
game.
4. Data Analysis and Results
The students' game performance, preferences for working alone or in a group, and class performance are
presented in Tables 1 - 3. The activities performed by each experimental group in the game are summarized in
Table 1. Group 2, with 47 students, had the highest levels of activity across all three metrics - questions created
(45.31%), questions answered (33.38%), and questions browsed (39.35%). Group 1, with 41 students, had the
lowest activity in answering (24.64%), and browsing (27.77%). Group 3, with 40 students, created fewer
questions (26.42%) than the other two groups, but answered the highest number of questions (41.98%). This
may suggest that group size alone does not determine activity levels, and there may be other factors influencing
student engagement with the game, such as the different game supports provided to each group and the
dynamics created within each group. Group 2 with the access to a question bank was the most active in creating
new questions (which was easier, having access to a question bank) and engaged in browsing solved questions.
To determine if there are significant variations between groups, we need a breakdown of solo vs. group work
preferences. Table 2 displays each group's preference for working individually or in a group. Interestingly, most
students in Groups 2 and 3 preferred working alone (55.32% and 62.50%, respectively), while the students in
Group 1 were evenly split between group and solo work preferences.
The average percentage grades for each group on assignments, lab tasks, and their final grade are presented in
Table 3. Group 3 had the highest average scores across all three measures, with an 88% on assignments, 90% on
labs, and an 81.46% final grade. The scores of Group 1 and Group 2 were similar, though Group 2 slightly
outperformed Group 1 in the average final grade (75.38% vs 74.82%).
Nafisul Kiron, Mehnuma Tabassum Omar and Julita Vassileva
Table 1: Summary of game activities by experimental groups.
Group 1 (n=41)
Created
229
28.27%
Answered
189
24.64%
Browsed
1394
27.77%
Group 2 (n=47)
Created
367
45.31%
Answered
256
33.38%
Browsed
1976
39.35%
Group 3 (n=40)
Created
214
26.42%
Answered
322
41.98%
Browsed
1651
32.88%
Table 2: Students' preference for working alone or in a group by experimental groups.
Group 1 (n=41)
Group
20
48.78%
Alone
21
51.22%
Group 2 (n=47)
Group
21
44.68%
Alone
26
55.32%
Group 3 (n=40)
Group
15
37.50%
Alone
25
62.50%
Table 3: Average percentage of grades of students in experimental groups on assignments, lab tasks, and
final grade.
Group 1 (n=41)
Assignments
84%
Lab
90%
Final grade
74.82%
Group 2 (n=47)
Assignments
85%
Lab
87%
Final grade
75.38%
Group 3 (n=40)
Assignments
88%
Lab
90%
Final grade
81.46%
The type of questions asked by the students could impact game dynamics and engagement levels. Table 4 shows
the breakdown of the types of questions - multiple choice, true/false, and short answer - used by each group in
the game. Group 1 created the most multiple-choice questions (153), and much fewer true/false (45) and short
answer (31) questions. Group 2 had a more even split between multiple choice (142), true/false (129) and short
answer (96). Group 3 created multiple-choice the least (94) with a fairly even number of true/false (58) and short
answer (62) questions. This provides insight into the question style preferences of each group. Group 1 favoured
multiple choice, while Group 3 used it the least. Group 2 utilized all three question types more evenly.
0.00%
10.00%
20.00%
30.00%
40.00%
50.00%
Created
Answered
Browsed
Created
Answered
Browsed
Created
Answered
Browsed
Group 1 (n=41) Group 2 (n=47) Group 3 (n=40)
Summary of Activities
0%
20%
40%
60%
80%
100%
Assignments
Lab
Final grade
Assignments
Lab
Final grade
Assignments
Lab
Final grade
Group 1 (n=41) Group 2 (n=47) Group 3 (n=40)
Grades
Nafisul Kiron, Mehnuma Tabassum Omar and Julita Vassileva
Table 4: The types of questions created by each experimental group.
Group 1 (n=41)
MCQ
153
True/False
45
Short answer
31
Group 2 (n=47)
MCQ
142
True/False
129
Short answer
96
Group 3 (n=40)
MCQ
94
True/False
58
Short answer
62
The correlation coefficients for the different factors in each experimental group can be found in Table 5. We
calculated the correlation coefficient using various factors including students' preferences for working alone or
in a group, the number of questions asked, answered, and browsed, and the grades obtained (assignment
grades, coding lab grades, and their final grade).
Table 5: The correlation coefficients for each of the experimental groups.
Group 1
Preference
Group/Alon
e
Asked
Answere
d
Browsed
#Assignme
nts
#Codela
b
#Final
Grade
Group/Alone
1.000
Asked
0.070
1.000
Answered
0.261
0.298
1.000
Browsed
0.228
0.443
0.747
1.000
#Assignments
-0.052
0.243
0.240
0.310
1.000
#Codelab
0.051
0.143
0.154
0.179
0.555
1.000
#Final Grade
-0.138
0.020
0.110
0.220
0.786
0.579
1.000
Group 2
Preference
Group/Alon
e
Asked
Answere
d
Browsed
#Assignme
nts
#Codela
b
#Final
Grade
Group/Alone
1.000
Asked
-0.103
1.000
Answered
-0.179
0.577
1.000
Browsed
-0.249
0.733
0.897
1.000
#Assignments
-0.110
0.246
0.205
0.291
1.000
#Codelab
-0.085
0.273
0.189
0.281
0.776
1.000
#Final Grade
-0.227
0.096
0.234
0.265
0.855
0.701
1.000
0
20
40
60
80
100
120
140
160
180
MCQ
True/False
Short answer
MCQ
True/False
Short answer
MCQ
True/False
Short answer
Group 1 (n=41) Group 2 (n=47) Group 3 (n=40)
Types of Questions Created
Nafisul Kiron, Mehnuma Tabassum Omar and Julita Vassileva
Group 3
Preference
Group/Alon
e
Asked
Answere
d
Browsed
#Assignme
nts
#Codela
b
#Final
Grade
Group/Alone
1.000
Asked
0.044
1.000
Answered
-0.011
0.757
1.000
Browsed
0.039
0.826
0.900
1.000
#Assignments
-0.219
0.280
0.182
0.254
1.000
#Codelab
-0.212
0.242
0.189
0.240
0.808
1.000
#Final Grade
-0.368
0.178
0.160
0.237
0.852
0.820
1.000
Across all three groups, strong positive correlations can be seen between the game activity metrics, especially
questions browsed and answered. The likelihood of questions being answered increases with higher levels of
browsing questions. Assignment grades, lab scores, and final grades also tend to be strongly correlated, which
is logical as assignments and labs are components of the final grade. Interestingly, game activities have only
weak to moderate positive correlations (varying between and 0.020 to 0.291) with assignment, lab and final
grades performance. The highest correlations can be found in Group 2, with correlations of 0.291, 0.281 and
0.265 between questions browsed and assignments, labs, and final grades, respectively. It is interesting to note
that unlike in Group 1 and 2, in Group 3 the correlations between the grades and the number of questions asked
are higher than those for questions browsed. This could suggest that creating new questions without any support
has a stronger effect on learning performance. Yet, the differences are small. Overall, it seems that in-game
activities do not readily translate to higher academic performance in assignments and exams.
In Group 1, there is a negative correlation (-0.138) between students' preference for working alone or in a group
and their final grade, indicating that those who prefer group work tend to have slightly lower grades. The
preference for working alone or in a group in Group 2 and Group 3 is negatively correlated with variables such
as answered, browsed, and final grade. This indicates that students who prefer group work tend to have lower
engagement in these activities and possibly lower final grades.
Figure 1 shows the social network of interactions between students in each experimental group. The directed
edges represent interactions (solving a question created by the other student). The size of each node represents
it centrality (i.e. the number of interactions with different peers).
Group 1
Nafisul Kiron, Mehnuma Tabassum Omar and Julita Vassileva
Group 2
Group 3
Figure 1: Network graph of the experimental groups. The nodes are the students, and the edges show their
interactions with other students. The size of the node corresponds to the number of questions created.
When comparing the activity of students in terms of question creation, which could be a sign of powergaming,
Group 1 (Density: 0.043) has two visibly large nodes (A and B), Group 2 (Density: 0.045) has one large Node (C),
and Group 3 (Density: 0.048) has three large nodes (D, E, and F). The thickness of an edge represents repeated
interactions between two nodes that may potentially signify a collusion (another forum of gaming). We observe
that in Group 1 there are 5 to 6 pairs of students who engage in repeated interactions, while in Group 2 and 3
there are only 2-3 such pairs.
5. Discussion and Limitations of the Study
The findings from this exploratory study offer several points for discussion regarding student behaviour and
performance in educational games or online learning platforms. One key area of interest is the relationship
between activity levels and grades. Although one might assume that increased engagement leads to higher
scores, these findings indicate a more complex relationship. The activity levels of Group 2 were the highest
across two out of the three metrics in Table 1, and they also had the second densest network of student
interactions in Figure 1. However, this group had average grades that were neither high nor low compared to
the other two groups.
In contrast, Group 3 had the lowest activity in two out of the three metrices according to Table 1, and highest
activity levels and the most connected network, and they obtained the highest average grades. This brings up
uncertainties regarding the form and level of engagement that promotes optimal learning. It is possible that
Group 3's approach of selective, focused interactions and a preference for solo work allowed for deeper
processing of the material, leading to better outcomes. The results of our study are consistent with Tejada-
Simon's research (Tejada-Simon, 2024). They reported there was no statistically significant connection between
student engagement with the games and their course grades.
Another notable point is the varying use of question types across groups. Group 1's heavy dependence on
multiple-choice questions and their avoidance of short answer questions, coupled with their lower average
grade, indicates that the choice of question format could affect the learning process. Short-answer questions
Nafisul Kiron, Mehnuma Tabassum Omar and Julita Vassileva
might encourage deeper understanding compared to multiple-choice questions. Utilizing a variety of question
types may have helped Group 3 perform better. Another factor is whether someone likes to work alone or in a
group. Group 3, which had significantly more students who preferred working alone than students who
preferred working groups, performed better than the other two groups where the student preferences were
more balanced. However, the correlations in Table 5 don't show a clear, consistent relationship between group
work preference and performance metrics.
Finally, we noticed some evidence of powergaming by a few students in all groups. However, we did not observe
any negative effects of powergaming on the other students. On the contrary, it seems that the contributions by
the powergamers were beneficial to other gamers. First, these students created a positive example through
their gameplay behaviour. Even in Group 2, where they had a source of questions they could use, they did not
copy directly questions from the database, but tried to improve them. By creating many questions, they made
sure that there were always questions available to other students that they could answer. Second, powergamers
can generate imperfect questions that may stimulate other players to adapt and improve them. This
involvement can result in deeper exploration of subjects, exposure to diverse viewpoints, and a more enriching
learning experience for all. Third, powergamers may introduce topics not taught in the class, adding unique
questions to the question bank of the game. The question bank can become a supplementary learning resource.
Finally, the presence of powergamers raises the overall game interaction and energy in the class, fostering a
more dynamic and stimulating learning environment that benefits everyone.
It's important to consider the limitations of the study. The sample sizes of each group are relatively small, with
40-47 students each. Larger samples would provide greater confidence in the patterns observed. The specific
design of the educational game or platform could also skew behaviours and outcomes. Despite these limitations,
this experiment offers valuable insights into how different interaction patterns and preferences among students
relate to learning outcomes. It challenges assumptions that more activity and interaction are always better and
highlights the need to consider the type and quality of engagement. Future research will explore these dynamics
in more depth, with larger samples and a wider range of academic contexts.
6. Conclusion
The study explores the complex connections among student behaviours, preferences, and performance in an
educational game. A comparison of the engagement and performance of three groups of students under
different game conditions, considering activity levels, question type usage, solo/group work preferences, and
interaction patterns, uncovers several important findings. Higher activity levels do not necessarily translate to
better grades. Although Group 2 had the highest levels of creating, answering, and browsing quiz questions,
their average grades were only moderate. Meanwhile, Group 3 had lower activity but the highest grades. This
suggests that the quality and type of engagement may be more important than sheer quantity.
The type of questions created by students may influence learning outcomes. The lowest average grades were
observed in Group 1, which heavily utilized MCQs and avoided true/false and short answer questions. In
contrast, Group 3's more balanced use of all question types aligned with their higher performance. This indicates
that a mix of question types, including those that require articulating understanding, may be beneficial. The
highest preference for solo work was observed in Group 3, which had the highest average grades. Correlations
between group work preference and performance metrics were inconsistent across groups. This suggests that
while these preferences shape interaction patterns, they don't necessarily dictate outcomes.
Student interaction patterns, as visualized in the network graphs, show some alignment with activity levels and
grades. Group 2 performed well in activity levels and network density but had average grades. Despite having
the lowest activity compared to Group 2, Group 3 achieved the most connected network, and higher average
grades. Within groups, browsing and answering more questions correlated with better assignment and lab
scores, which in turn correlated with higher final grades. Active engagement with course content, such as
exploring and attempting questions, could enhance individual learning. Furthermore, the bigger nodes belong
to the most active students who asked the most questions. Despite their tendency to be overactive, which could
be considered as powergaming, they may have played a beneficial role in the game.
Future research could build on these findings by exploring similar questions with larger, more diverse samples
and across different academic subjects and platforms. Qualitative data, such as quality analysis of the questions
created (using experts or trained large language models), student interviews and content analysis of the
interactions (attempted answers to questions, time to first correct answer, etc.), could provide deeper insight
into the factors driving these patterns. This study emphasizes the importance of carefully designing and
Nafisul Kiron, Mehnuma Tabassum Omar and Julita Vassileva
evaluating these platforms, considering activity levels, question types, interaction patterns, and individual
preferences. Furthermore, while powergamers are usually seen as troublemakers in game-environments, our
findings suggest that may be beneficial depending on the type of games. While their overactivity might be a
concern for us, the overactive users were the ones responsible for providing questions for the other students to
answer.
References
An, Y., 2020. Designing Effective Gamified Learning Experiences. International Journal of Technology in Education 3, 62.
Bösche, W., Kattner, F., 2011. Fear of (Serious) Digital Games and Game-Based Learning?: Causes, Consequences and a
Possible Countermeasure. International Journal of Games Based Learning 1, 1–15.
Caeiro-Rodriguez, M., Manso-Vazquez, M., Llamas-Nistal, M., Mikic Fonte, F.A., Fernandez-Iglesias, M.J., Tsalapata, H.,
Heidmann, O., Vaz-De-Carvalho, C., Jesmin, T., Terasmaa, J., Tolstrup, L., 2021. A collaborative city-based game to
support soft skills development in engineering and economics. International Symposium on Computers in Education.
Calle, S., Bonfante, E., Riascos, R., 2019. Introduction of the Game- Based Learning Platform, Kahoot, as a Tool in Radiology
Resident Training. Academic Medicine.
Ceccarini, C., Prandi, C., 2022. EscapeCampus: exploiting a Game-based Learning tool to increase the sustainability
knowledge of students. Conference on Information Technology for Social Good 390–396.
Dong, X., 2023. Research on Joint Application of Digital Media Interaction Technology and Points-Based Reward System in
Digital Game-based Learning Take Shanghai Eagle Childrens Museum as an Example. Lecture Notes in Education
Psychology and Public Media 2, 813–821.
Faria, A.J., Windsor, Whiteley, T., 2002. COGNITIVE AND BEHAVIORAL CONSISTENCY IN A COMPUTER-BASED MARKETING-
SIMULATION-GAME ENVIRONMENT: AN EMPIRICAL INVESTIGATION OF THE DECISION-MAKING PROCESS.
Featherstone, M., 2022. Optimising gamification using constructive competition and videogames.
Flores, N.H., Pinto, R., 2023. Quest-based Gamification In A Software Development Lab Course: A Case Study. 9th
International Conference on Higher Education Advances (HEAd’23) 541–548.
Fu, F., Yu, S.C., 2006. The Games in E Learning Improve the Performance. International Conference on Information
Technology Based Higher Education and Training 732–738.
Guerrero-Quiñonez, A.J., Guagua, O.Q., Barrera-Proaño, R.G., 2023. Gamified flipped classroom as a pedagogical strategy in
higher education: From a systematic vision. Ibero-American Journal of Education & Society Research 3, 238–243.
Islam, A., 2017. CROSS-MODAL COMPUTER GAMES AS AN INTERACTIVE LEARNING MEDIUM.
Iverson, K., 2019. Gamification of the Classroom: Seeking to Improve Student Learning and Engagement.
Josiek, S., Schleier, S., Steindorf, T., Wittrin, R., Heinzig, M., Roschke, C., Tolkmitt, V., Ritter, M., 2020. Game-Based Learning
Using the Example of Finanzmars. Colloquium in Information Science and Technology 2020-June, 7–14.
Kaldarova, B., Omarov, B., Zhaidakbayeva, L., Tursynbayev, A., Beissenova, G., Kurmanbayev, B., Anarbayev, A., 2023.
Applying game-based learning to a primary school class in computer science terminology learning. Front Educ
(Lausanne) 8.
Kiili, K., Siuko, J., Cloude, E., Dindar, M., 2023. Demystifying the Relations of Motivation and Emotions in Game-Based
Learning: Insights from Co-Occurrence Network Analysis. Int. J. Serious Games 10, 93–112.
Kiron, N., Adaji, I., Long, J., Vassileva, J., 2019. Tower of questions (TOQ): A serious game for peer learning. In: Lecture
Notes in Computer Science (Including Subseries Lecture Notes in Artificial Intelligence and Lecture Notes in
Bioinformatics). Springer, pp. 276–286.
Kiron, N., Vassileva, J., 2018. Tower of Questions: Gamified Testing to Engage Students in Peer Evaluation. In: ITS 2018
Workshop Proceedings. p. 113.
Klit, K.J.M., Nielsen, C.K., Stege, H., 2020. Iterative Development of a Digital Game-Based Learning Concept: Introduction of
Veterinary Herd Health Management in a Virtual Pig Herd. J Vet Med Educ 47, 523–531.
Malegiannaki, I., Daradoumis, T., Retalis, S., 2021. Using a Story-Driven Board Game to Engage Students and Adults With
Cultural Heritage. International Journal of Games Based Learning 11, 1–19.
Marques, M.M., Pombo, L., 2021. Current Trends in Game-Based Learning—Introduction to a Special Collection of
Research. Educ Sci (Basel) 11.
Mavridis, A., Tsiatsos, T., 2014. Improving Collaboration between Students Exploiting a 3D Game. 2014 IEEE 14th
International Conference on Advanced Learning Technologies 671–675.
Muszy´nski, R., Muszy´nski, M., Wang, J., 2017. Happiness Pursuit: Personality Learning in a Society of Agents.
Nipo, D., Gadelha, D., Silva, M. da, Lopes, A., 2023. Game-Based Learning: Possibilities of an Instrumental Approach to the
FEZ Game for the Teaching of the Orthographic Drawings System Concepts. Journal of Interactive Systems 14, 231–
243.
Perbawa, D.S., Rapiyanta, P.T., 2024. Computers Introduction For Elementary School Students Using Game-Based Learning.
International Journal of Business, Law, and Education 5, 635–643.
Riet, M. ter, 2019. Serious Games: Games That Facilitate Learning : The effects of freedom and rewards in game based
learning.
Silveri, L., 2022. THE GAME OF LEARNING! APPROACHING ECOSYSTEMS THROUGH BOARD GAME DESIGN. Education and
New Developments 2022 – Volume 2 172–176.
Nafisul Kiron, Mehnuma Tabassum Omar and Julita Vassileva
Soo, C., Lee, J.A.C., 2022. The Psychology of Rewards in Digital Game-Based Learning: A Comprehensive Review. Journal of
Cognitive Sciences and Human Development 8, 68–88.
Stefan, I.A., Gheorghe, A.F., Stefan, A., Piki, A., Tsalapata, H., Heidmann, O., 2022. Constructing Seamless Learning Through
Game-Based Learning Experiences. International Journal of Mobile and Blended Learning 14, 1–12.
Tejada-Simon, M. V., 2024. Exploring the Impact of Game-Based Learning and Creative Active-Learning Activities on
Student Engagement and Academic Performance: A Case Study in the basic sciences for Pharmacy Education.
Physiology 39.
Wang, Y., Rajan, P., Sankar, C.S., Raju, P.K., 2016. Let Them Play: The Impact of Mechanics and Dynamics of a Serious Game
on Student Perceptions of Learning Engagement. IEEE Transactions on Learning Technologies 10, 514–525.
Yu, Y.T., Tsuei, M., 2022. The effects of digital game-based learning on children’s Chinese language learning, attention and
self-efficacy. Interactive Learning Environments 31, 6113–6132.
