Classroom VR: a VR game
to improve communication skills

in secondary-school teachers
Classroom VR: un juego en RV para

mejorar las capacidades comunicativas
de profesores de secundaria

Trabajo de fin de grado del Grado en Ingeniería
Informática

Facultad de Informática
Universidad Complutense de Madrid

Autores
Mario Bocos Corredor
Álvaro López García
Alejandro Díaz Nieto

Directores
Borja Manero Iglesias

Alejandro Romero Hernández

Curso 2019-2020





Acknowledgements

This work is dedicated to everyone that has contributed to making it fea-
sible. Firstly, we wanted to thank our directors, Borja and Alejandro, who
have allowed us to get involved in this project, giving us a unique oppor-
tunity to contribute to solving a problem that is present in current society.
Furthermore, they have been patient and thoughtful with us during this
course, helping us to achieve every goal and milestone, as well as to solve
any conflicts or doubts that appeared in the process.

In second place, we feel honored for the fact that the pedagogues from
Barcelona, concretely Ibis, Isabel and Montse, have carried out an extensive
analysis for our project and always have counted on us to discuss every aspect
of it. We are really thankful.

We also thank the Complutense University of Madrid, concretely the
Facultad de Informática, for providing us with the means to conduct the
different parts of our project. Besides, we are grateful to Meriem El Yamri,
a member of Borja and Alejandro’s research team, for helping us with the
integration of virtual reality and with the testing of the tool.

Finally, we appreciate the support given by our families and friends and
the help received by those who contributed to recording the audio clips for
the efficiency analysis and the scenarios.

iii





Resumen

Según los informes de prácticas de capacitación de los futuros maestros
de educación secundaria, existen importantes deficiencias en las aptitudes de
comunicación necesarias para dirigir adecuadamente una clase. A pesar de
la existencia de módulos relacionados con la comunicación en los maestros
de enseñanza, no hay un aprendizaje real en esta área, y los estudiantes (los
futuros profesores) expresan su temor cuando se trata de manejar una clase.

Actualmente, la realidad virtual está presente en diferentes campos, co-
mo la medicina, la ingeniería, las humanidades, etc. Sus características (en
particular la relativa a la creación de un entorno seguro, su ubicuidad y
su alta capacidad de inmersión) la convierten en una tecnología ideal para
la formación de habilidades de comunicación. Paralelamente, en los últimos
diez años, los videojuegos han demostrado su eficacia como herramientas de
aprendizaje, solos o combinados con otras estrategias pedagógicas.

Por estas razones, consideramos la siguiente hipótesis: La realidad virtual
en combinación con los videojuegos es, por un lado, un entorno eficaz para
simular situaciones realistas relacionadas con la enseñanza secundaria y, por
otro, un entorno ideal para practicar y aprender habilidades de comunicación
para el aula.

Teniendo esto en cuenta, hemos creado un entorno virtual que sirve co-
mo herramienta de formación para los futuros profesores de secundaria, para
mejorar sus habilidades de comunicación y acostumbrarse a una clase real.
Simula una clase de secundaria con estudiantes y funciona como un videojue-
go. Los estudiantes virtuales simulan el comportamiento de los estudiantes
reales en el aula.

Los futuros profesores empiezan el juego delante de la clase, como en la
mayoría de las escuelas secundarias. La herramienta establece el fondo del
juego, ofreciendo un contexto completo, acciones o comportamientos previos
(por ejemplo, dos estudiantes han estado peleando toda la semana anterior).
Después, el usuario tiene que reaccionar como en una situación real, mane-
jando situaciones complicadas. Nuestro videojuego analiza en tiempo real el
comportamiento del usuario (tono de voz, distancia entre el usuario y el o
los estudiantes problemáticos y detección de vocabulario asertivo o negati-
vo) y reacciona a través de los propios estudiantes, es decir, ofreciendo un

v



Resumen vi

comportamiento realista a las acciones del usuario.
Somos optimistas sobre las posibilidades de nuestra herramienta en la

formación de los profesores, ya que evita las costosas y complicadas prác-
ticas logísticas en entornos reales. Los expertos consultados coinciden en el
valor añadido que esta herramienta puede ofrecer a las prácticas en escuelas
reales. El videojuego ofrece el entorno perfecto para aprender y comprender
cómo manejar situaciones estresantes a través de una comunicación efectiva
antes de enfrentarse a estudiantes reales.

Palabras clave
Videojuegos, Educación, Realidad virtual, Enseñanza, Juegos serios.



Abstract

According to the reports of training practices from future secondary-
school teachers, there are significant shortcomings in the communication
skills required to properly manage a class. Despite the existence of modules
related to communication in the educational masters of teaching staff, there
is no real learning in this area, and students (the future teachers) express
their fear when it comes to dealing with a class.

Currently, virtual reality is present in different fields, such as medicine,
engineering, humanities, etc. Its characteristics (in particular the one concer-
ning the creation of a safe environment, its ubiquity, and its high immersion
capacity) turn it into an ideal technology for training communication skills.
In parallel, in the last ten years, video games have proven their effectiveness
as learning tools, alone or combined with other pedagogical strategies.

For these reasons, we consider the following hypothesis: Virtual reality in
combination with video games is, on the one hand, an effective environment
for simulating realistic situations related to secondary education, and, on the
other, an ideal environment to practice and learn communication skills for
the classroom.

Taking this into account, we have created a virtual environment that
serves as a training tool for future secondary-school teachers, to improve their
communication skills and get used to a real class. It simulates a secondary-
school class with students and works as a video game. The virtual students
work by simulating behaviors of real students in the classroom.

Future teachers start the game in front of the class as in most secondary
schools. The tool sets the game background, offering the complete context,
actions, or previous behaviors (i.e., two students have been fighting all over
the previous week). After, the user has to react as in a real situation, mana-
ging complicated situations. Our video game analyses in real-time the user’s
behavior (voice tone, the distance between the user and the problematic
student or students and detection of assertive or negative vocabulary) and
reacts through the students themselves, i.e., offering realistic behavior to the
user’s actions.

We are optimistic about the possibilities of our tool on the teachers’training
since it avoids the expensive and logistical complicated real environments

vii



Abstract viii

practices. The experts consulted agree on the added value that this tool can
offer to practices in real schools. The video game offers the perfect environ-
ment to learn and understand how to manage stressful situations through
effective communication before facing real students.

Keywords
Video games, Education, Virtual reality, Teaching, Serious games.





Index

1. Introduction 1
1.1. Objectives . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2

2. State of the art 3
2.1. The problem of current education . . . . . . . . . . . . . . . . 3

2.1.1. Teacher training and conflict management techniques . 4
2.1.2. The Four Pillars of Learning . . . . . . . . . . . . . . 4

2.2. Video games . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5
2.2.1. What is a video game? . . . . . . . . . . . . . . . . . . 6

2.3. Virtual reality . . . . . . . . . . . . . . . . . . . . . . . . . . . 7
2.4. Virtual environments applied to education . . . . . . . . . . . 8

2.4.1. Serious games or educational video games . . . . . . . 8
2.4.1.1. Serious games applications . . . . . . . . . . 8

a. Edutainment . . . . . . . . . . . . . . . . . . 8
b. Educational games . . . . . . . . . . . . . . . 9
c. Games for health . . . . . . . . . . . . . . . . 11
d. Companies, institutions and business corpo-

rations . . . . . . . . . . . . . . . . 11
2.4.1.2. Serious games at Universidad Complutense

de Madrid . . . . . . . . . . . . . . . . . . . 12
a. The ComunicArte (PR2005-174/01) project . 12
b. The Didascalia (RTI2018-096401-A-I00) pro-

ject . . . . . . . . . . . . . . . . . . 13
2.5. Video game creation tools . . . . . . . . . . . . . . . . . . . . 13

2.5.1. Unity . . . . . . . . . . . . . . . . . . . . . . . . . . . 14
2.5.2. Unreal Engine . . . . . . . . . . . . . . . . . . . . . . . 15

3. Development and implementation 16
3.1. Design of scenarios . . . . . . . . . . . . . . . . . . . . . . . . 16

3.1.1. Selection criteria . . . . . . . . . . . . . . . . . . . . . 16
3.1.2. Obtaining the reports . . . . . . . . . . . . . . . . . . 17

x



Index xi

3.1.3. Proposed scenarios . . . . . . . . . . . . . . . . . . . . 19
3.1.3.1. First scenario: Inappropriate comment from

a student . . . . . . . . . . . . . . . . . . . . 20
3.1.3.2. Second scenario: Two students refuse to sit

apart . . . . . . . . . . . . . . . . . . . . . . 21
3.1.3.3. Third scenario: A student is not welcomed by

his or her classmates . . . . . . . . . . . . . . 21
3.1.4. Adaptation of the scenarios to Unity . . . . . . . . . . 22

3.1.4.1. First scenario . . . . . . . . . . . . . . . . . . 22
3.1.4.2. Second scenario . . . . . . . . . . . . . . . . 22
3.1.4.3. Third scenario . . . . . . . . . . . . . . . . . 23

3.2. Implementation . . . . . . . . . . . . . . . . . . . . . . . . . . 24
3.2.1. Student models and animations . . . . . . . . . . . . . 24
3.2.2. First draft of the class . . . . . . . . . . . . . . . . . . 25
3.2.3. Asset pack selection process . . . . . . . . . . . . . . . 26

3.2.3.1. Initial asset pack . . . . . . . . . . . . . . . . 26
3.2.3.2. Final asset pack . . . . . . . . . . . . . . . . 28

3.2.4. Implementation of the menu . . . . . . . . . . . . . . . 29
3.2.5. Tools . . . . . . . . . . . . . . . . . . . . . . . . . . . . 30

3.2.5.1. Controllers . . . . . . . . . . . . . . . . . . . 30
3.2.5.2. Colliders . . . . . . . . . . . . . . . . . . . . 32
3.2.5.3. NavMesh . . . . . . . . . . . . . . . . . . . . 32
3.2.5.4. SoundLoudness . . . . . . . . . . . . . . . . . 33
3.2.5.5. KeywordRecognizer . . . . . . . . . . . . . . 34
3.2.5.6. Vokaturi . . . . . . . . . . . . . . . . . . . . 34

a. Efficiency analisis . . . . . . . . . . . . . . . 34
b. Unity adaptation . . . . . . . . . . . . . . . . 36

3.2.5.7. Virtual reality adaptation . . . . . . . . . . . 36
3.2.6. Audios . . . . . . . . . . . . . . . . . . . . . . . . . . . 37

3.3. Explanation of the scenarios . . . . . . . . . . . . . . . . . . . 38
3.3.1. Common structure . . . . . . . . . . . . . . . . . . . . 38

3.3.1.1. Context screen . . . . . . . . . . . . . . . . . 38
3.3.1.2. TipsPanel . . . . . . . . . . . . . . . . . . . . 39
3.3.1.3. Feedback screen . . . . . . . . . . . . . . . . 39

3.3.2. First scenario: Inappropriate comment from a student 40
3.3.2.1. Context and problem . . . . . . . . . . . . . 40
3.3.2.2. Paths . . . . . . . . . . . . . . . . . . . . . . 40
3.3.2.3. Scenario script . . . . . . . . . . . . . . . . . 42

3.3.3. Second scenario: Two students refuse to sit apart . . . 43
3.3.3.1. Context and problem . . . . . . . . . . . . . 43
3.3.3.2. Paths . . . . . . . . . . . . . . . . . . . . . . 43



Index xii

3.3.3.3. Scenario script . . . . . . . . . . . . . . . . . 46
3.3.4. Third scenario: A student is not welcomed by his class-

mates . . . . . . . . . . . . . . . . . . . . . . . . . . . 47
3.3.4.1. Context and problem . . . . . . . . . . . . . 47
3.3.4.2. Paths . . . . . . . . . . . . . . . . . . . . . . 47
3.3.4.3. Scenario script . . . . . . . . . . . . . . . . . 49

4. Individual contribution 50
4.1. Alejandro Díaz . . . . . . . . . . . . . . . . . . . . . . . . . . 50
4.2. Álvaro López . . . . . . . . . . . . . . . . . . . . . . . . . . . 51
4.3. Mario Bocos . . . . . . . . . . . . . . . . . . . . . . . . . . . . 52
4.4. Collaborative work . . . . . . . . . . . . . . . . . . . . . . . . 53

5. Conclusions and future work 55
5.1. Conclusions . . . . . . . . . . . . . . . . . . . . . . . . . . . . 55

5.1.1. Accomplished objectives . . . . . . . . . . . . . . . . . 56
5.1.2. Main conclusions . . . . . . . . . . . . . . . . . . . . . 56

5.2. Future work . . . . . . . . . . . . . . . . . . . . . . . . . . . . 57

Appendices 59
Acceptance Letter . . . . . . . . . . . . . . . . . . . . . . . . . . . 59
EDULEARN20 Paper . . . . . . . . . . . . . . . . . . . . . . . . . 61

References 73



List of Figures

2.1. Components of our project . . . . . . . . . . . . . . . . . . . . 3
2.2. The Four Pillars of Learning (Vallejo, 2005) . . . . . . . . . . 5
2.3. VR headset from HTC and two controllers (HTC) . . . . . . 7
2.4. Assassin’s Creed Origins: The Discovery Tour (Ubisoft, 2018) 9
2.5. Teatrix (Prada et al., 2000) . . . . . . . . . . . . . . . . . . . 9
2.6. Construct3D (Kaufmann et al., 2000) . . . . . . . . . . . . . . 10
2.7. The motion capture system (Chan et al., 2010) . . . . . . . . 10
2.8. Dancing a treasure . . . . . . . . . . . . . . . . . . . . . . . . 11
2.9. JUST-TALK (Frank et al., 2002) . . . . . . . . . . . . . . . . 12
2.10. The ComunicArte project (ComunicArte) . . . . . . . . . . . 13
2.11. The Unity game engine (Github For Unity) . . . . . . . . . . 14
2.12. Unreal Engine (MasterD) . . . . . . . . . . . . . . . . . . . . 15

3.1. Communication process diagram . . . . . . . . . . . . . . . . 17
3.2. Analysis of the disciplinary records from one of the institutes 18
3.3. Screenshot of the meeting with the educators (02 / 13 / 2020) 19
3.4. A scenario as depicted in the slides from the educators . . . . 20
3.5. First scenario as proposed by the educators . . . . . . . . . . 20
3.6. Second scenario as proposed by the educators . . . . . . . . . 21
3.7. Third scenario as proposed by the educators . . . . . . . . . . 21
3.8. First scenario adapted to Unity . . . . . . . . . . . . . . . . . 22
3.9. Second scenario adapted to Unity . . . . . . . . . . . . . . . . 23
3.10. Third scenario adapted to Unity . . . . . . . . . . . . . . . . 23
3.11. Models created with Adobe Fuse (Wikiwand) . . . . . . . . . 24
3.12. Animating a model in Mixamo . . . . . . . . . . . . . . . . . 25
3.13. The first draft of the class . . . . . . . . . . . . . . . . . . . . 26
3.14. The class with the first asset pack “Snaps Art HD - School” . 27
3.15. Our class compared with a typical Spanish class . . . . . . . . 28
3.16. Students in the final class . . . . . . . . . . . . . . . . . . . . 29
3.17. Main menu . . . . . . . . . . . . . . . . . . . . . . . . . . . . 30
3.18. Submenu . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 30

xiii



List of figures xiv

3.19. Students´ Controller . . . . . . . . . . . . . . . . . . . . . . . 31
3.20. Example of Override Controller . . . . . . . . . . . . . . . . . 31
3.21. Sphere Collider, in green . . . . . . . . . . . . . . . . . . . . . 32
3.22. Sitting down animation being played after a Collider collision 32
3.23. Walkable Surface of the NavMesh, in blue . . . . . . . . . . . 33
3.24. Original and modified graph model . . . . . . . . . . . . . . . 36
3.25. Member of the research team testing the tool . . . . . . . . . 37
3.26. Freesound . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 37
3.27. Context screen in the second scenario . . . . . . . . . . . . . . 38
3.28. Tips panel before and after the player’s choice . . . . . . . . . 39
3.29. Feedback screen in the second scenario . . . . . . . . . . . . . 39
3.30. Students getting distracted after authoritative answer . . . . 40
3.31. The teacher talking to the student calmly . . . . . . . . . . . 41
3.32. Students getting distracted after the teacher ignores the problem 41
3.33. Conflictive student animation . . . . . . . . . . . . . . . . . . 42
3.34. Student getting distracted after the assertive petition . . . . . 44
3.35. Students making noise after the teacher ignores the conflict . 44
3.36. Students talking with the teacher . . . . . . . . . . . . . . . . 45
3.37. Student walking to another desk . . . . . . . . . . . . . . . . 46
3.38. Student sitting back down in his desk . . . . . . . . . . . . . . 48
3.39. Student walking to the back of the class . . . . . . . . . . . . 48
3.40. Student explaining the problem to the teacher . . . . . . . . . 49

4.1. Results of the brainstorming session . . . . . . . . . . . . . . 53



List of Tables

3.1. Correlation matrix for the English audios . . . . . . . . . . . 35
3.2. Correlation matrix for the Spanish audios . . . . . . . . . . . 35

xv





Chapter 1

Introduction

The lack of communication skills training in many developed countries is
a fact. European Union has warned of the need to take action to enhance Eu-
ropeans oral communication skills. Teachers are not an exception. (European
Comission, 2020) Around the world, teachers are not adequately trained, and
their lack of communication skills is notorious. (Hoffman et al., 2004) As a
result, many of them leave their jobs in their first year due to frustration and
other mental health problems (Fukuda et al., 2017). Particularly in Spain,
there is no systematic treatment of communication issues, and most teachers
do not receive specific training in oral expression. (García Aróstegui, 2008)

Several approaches are trying to solve this aspect inside the master’s de-
grees in teacher training (Universidad Complutense de Madrid) (Universitat
de Barcelona). However, the results obtained show that the development of
these skills has significant shortcomings (Segovia et al., 2010) as there is no
specific subject to improve communication skills. Instead, basic notions are
taught in other subjects, which are insufficient, and practical sessions on this
topic are short and given by teachers who don’t have full knowledge of the
issue.

On the other hand, advances in technology have much contributed and
in different ways to society, ranging from the treatment of mental health pro-
blems such as post-traumatic stress disorder (Rothbaum et al., 1999), social
anxiety (Anderson et al., 2003) or phobias (Rothbaum et al., 1996), to the
training of workers (Frank et al., 2002), in addition to new ways of teaching
students in different disciplines such as theatre (Prada et al., 2000), dance
(Chan et al., 2010) or geometry (Kaufmann et al., 2000). One of the advances
that have contributed most to making this possible is virtual reality, which
makes it possible to create very immersive environments. (Rubio-Tamayo et
al., 2017)

As part of the Didascalia group, we were encouraged to find a way to help
teachers face a classroom full of students who would behave unexpectedly.
Our proposal is a virtual reality-based video game. By developing a simulator

1



1.1. Objectives 2

with a virtual classroom full of students whose behavior can be scripted, we
could provide teachers with different situations to see how they would react
and give them feedback based on their actions.

To develop such a video game, we would need examples of real-life si-
tuations that happen inside a class. For this purpose we got in contact with
our project partners, who were professors from the University of Barcelona
(UAB). They provided us with detailed scenarios that took place in real
classes. We proceeded to integrate said situations into our video game, with
students who would react according to what the player did to solve the pro-
blem.

This approach would help teachers to give an actual lecture and know
which actions would have a better outcome for the class, dealing with con-
flictive situations which will be helpful for their professional career. So, our
video game/simulator is addressed to future teachers who would want to
experience how teaching a class is like.

We had the chance to write a paper on the resulting video game and send
it to the EDULEARN20 (Edulearn) International Education Conference,
with the name “A VR GAME TO IMPROVE COMMUNICATION SKILLS
IN SECONDARY-SCHOOL TEACHERS”, which was accepted. Due to the
sanitary situation, the conferences could not be held in person, so we made
a presentation video to be submitted instead. (Video Presentation)

1.1. Objectives

The main goal of our project was to create a tool capable of helping
future teachers to face conflicts in a classroom and reduce their stress level
in real-life situations.

Other objectives were the following:

To carry out a study of the existing virtual reality tools related to
education and teaching.

To simulate a secondary education classroom with virtual students
with realistic behaviours.

To develop a video game which accurately reflects a real-life classroom
with students who behave in different ways.

To give detailed feedback to the teachers that may use this tool, infor-
ming them about their performance.

To allow future researchers to gather data on what would be the reac-
tions from the teachers to the different situations presented to them.



Chapter 2

State of the art

In this chapter we will cover the problem of current education, the charac-
teristics of video games and virtual reality, the application of virtual reality
to education and different tools used for creating video games.

Figure 2.1: Components of our project

2.1. The problem of current education

Conflicts are one of the main concerns of teachers, tutors, parents and
educational institutions (Jurado-de-los Santos y Tejada-Fernández, 2019)
(MAHVAR et al., 2018). Respecting and maintaining order in the class-
rooms are activities to which a great deal of effort is devoted on a daily basis
without, in many cases, achieving the expected results.

According to the ANPE syndicate, conflict in the classrooms occurs at
younger and younger ages. (Europa Press, 2019) Moreover, the reports of
training practices from future secondary-school teachers say that there are
significant shortcomings in the communication skills required to properly
manage a class. Despite the existence of modules related to communication

3



2.1. The problem of current education 4

in the educational masters of teaching staff, there is no real learning in this
area, and students (the future teachers) express their fear when it comes to
dealing with a class. (Flores y Fernández-Castro, 2004)

2.1.1. Teacher training and conflict management techniques

The current training process of future teachers in Spain is as follows
(Universidad Complutense de Madrid):

From September to February, teachers have more practical subjects and
then more theoretical ones. Then, starting in February they have a practicum
that lasts 3 months being assistant teachers in institutes, with a master
teacher who guides them. However, they also must teach some lectures. After
that the training is over, so it’s not enough time to get used to teaching in a
real classroom. Because of this, they are still very anxious, and 52% of the
future teachers think they are not sufficiently prepared. (ABC, 2019)

Regarding conflict in classrooms, the teacher must know how to use it
skillfully, regulate it and treat it positively. For this purpose, They must use
mediation and negotiation. (MAHVAR et al., 2018)

Both are complementary ways of managing conflicts, with mediation
being one way of carrying out the negotiation. These terms are especially
important for mitigating the factors that cause violence in education.

There are two basic types of negotiation: collaborative and competitive.
(Thomas, 2013) In collaborative negotiations, both parts look for an adequa-
te solution to their demands. This is the system of negotiation that should
prevail in schools, since there is an affective bond and equality.

Conversely, competitive negotiation is based in personal positioning wit-
hout putting oneself in the place of the other part. This approach should
be used with caution because it can promote more serious conflicts than the
initial ones.

With our tool, we are going to try to identify how this negotiation is
approached and give feedback to the users in order to improve their skills
when it comes to dealing with conflicts.

On the other hand, mediation derives directly from negotiation and in-
volves the physical presence of a person who tries to neutralize the resulting
behaviors in a conflict. The teacher is the one who usually acts as a mediator
in schools, being the person who helps to create a good environment in the
classroom. (Vallejo, 2005)

2.1.2. The Four Pillars of Learning

According to the report for UNESCO on Higher Education by Jacques
Delors, education should be structured around four pillars of knowledge.
(Vallejo, 2005) In any educational system, each of these pillars should be



2.2. Video games 5

given equivalent attention in order to make education a comprehensive and
lifelong experience at the cognitive and practical levels.

These four pillars are:

Learning to know: acquiring the tools that allow the subject to unders-
tand the world around him.

Learning to do: the capacity to influence one’s environment, putting
into practice the knowledge learned.

Learning to live with others: participating and cooperating with ot-
hers in the various human activities. Respecting the values of mutual
understanding and peace.

Learning to be: through education, people should possess autonomous
and critical thinking, as well as their own judgement about things.
Aspects such as imagination and creativity are very relevant in this
pillar.

Encouraging teamwork is the best way to act in conflict prevention and
resolution. Sometimes, group activities are proposed, but teachers do not
dive into interpersonal relationships that occur among the students.

Figure 2.2: The Four Pillars of Learning (Vallejo, 2005)

2.2. Video games

The video game industry has experienced a huge increase in interest from
the population in recent years, especially among young people. By 2018, the
United States video game industry had matched that of the United States
film industry in terms of revenue. (Minotti, 2019) (Robb, 2018)



2.2. Video games 6

According to Microsoft, there are more than two billion gamers around
the world, including everyone from those playing free games on their phone
to those using a computer fitted with the latest hardware. (Webb, 2019)

Video games have become a part of many people’s daily lives. However,
we should not think of them only as an entertainment, since they can be
extended to a wide variety of functions. But before we go any further, it is
essential to define what a video game is for the purpose of understanding
this project.

2.2.1. What is a video game?

It is difficult to find a single definition for this term, since video games
are present in many aspects of human beings today.

Every day new types of video games are created that can have very
diverse applications, from simple entertainment to the education of students,
including physical exercise.

However, according to Nicolas Esposito (Esposito, 2005), this definition
could work:

A video game is a game which we play thanks to an audiovisual apparatus
and which can be based on a story.

Following this definition, we must also clarify what exactly is a game. Eric
Zimmerman describes it as a “voluntary interactive activity, in which one or
more players follow rules that constrain their behavior, enacting an artificial
conflict that ends in a quantifiable outcome”. (Esposito, 2005)

Playing would be the free space of movement within a more rigid struc-
ture. A voluntary activity or occupation executed within certain fixed limits
of time and place, according to rules freely accepted but absolutely binding.

Playing has some benefits such as the pleasures of competition, accom-
plishment, system mastering, narrative enjoyment and audiovisual experien-
ce. However, these benefits are also useful in many other aspects beyond
entertainment, as we will see later.

Talking about the audiovisual apparatus, this refers to the electronic
system with computing capabilities, input devices and output devices. This
means that there are human-computer interactions and that video games
can be seen as user interfaces. In this way, the interactivity is the factor that
makes the major difference between video games and games.

In the case of virtual reality, this interactivity is enhanced in a much
more immersive way for the player, who can experience a simulation of a real
environment. Finally, the story is the element in which a game can be based,
but not in all cases. There are many ways to insert narrative elements in a
video game, such as back-stories, cut-scenes between levels or conversations
between characters.



2.3. Virtual reality 7

However, according to Celia Pearce, “games have a play-centric context.
They center in play, unlike literature and film, which center in story. In
games, everything revolves around play and the player experience.” (Pearce,
2004)

This is why there are video games like Tetris. They don’t always need to
be based on a story.

2.3. Virtual reality

Virtual reality, or VR for short, is a computer-generated simulation which
places the user inside a 3D artificial world in which he/she can interact with
the environment as if it was real. This allows for the creation of realistic
scenarios that can be used for various purposes, such as education, enter-
tainment or social science.

Most of the current virtual reality systems use a virtual reality headset, a
device which the player places in his/her head and displays the environment
through two different images, one for each eye. (Oculus) It also provides
sound and tracks the movements of the player’s head, so that the view can
be changed according to where the player looks physically.

As well as that, some VR systems also contain controllers which are
especially useful for video games, since they help the player move around
the virtual environment without walking around in the real world. (VIVE)
They can also be used to point to a specific object, like a mouse, among
other things.

Figure 2.3: VR headset from HTC and two controllers (HTC)

Video games are one of the most common applications of VR. Since VR
allows the player to get inside a virtual world, it is the most realistic depiction
of an environment that can be made, and thus provides the perfect setting
for a video game that is intended to provide an immersive experience.

Despite all this, VR is not the most common technology for developing
nor playing video games. This is due to its high cost, the lack of VR video



2.4. Virtual environments applied to education 8

games and the convenience that a small console offers over a bigger system
which requires space around the player to work. (VALVE)

2.4. Virtual environments applied to education

2.4.1. Serious games or educational video games

Serious games are usually defined as games used for purposes other than
mere entertainment. (Susi et al., 2007)

The purpose of this type of games is creating learning environments that
allow us to experiment with real world problems. In this way, serious games
help us explore different solutions to problems posed in real situations and
discover the information and knowledge that would help to be able to inter-
vene without fear of error. The lack of real-life consequences facilitates the
decision-making process, teaching us the optimal choices in those situations
and helping the development of different skills. (López Raventós, 2016)

This is an extremely powerful tool that makes possible the creation of
virtual environments that would be impossible in the real world for reasons
of safety, cost, time, etc. (Susi et al., 2007)

2.4.1.1. Serious games applications

Some markets where serious games are used can include military games,
educational games, healthcare games, etc. (Susi et al., 2007) In this section
we are going to see some examples in order to have a context for the better
understanding of this project.

a. Edutainment

One of the most common terms used for the classification of this type
of games is edutainment, coming from education and entertainment. (Okan,
2003) These games try to teach players by presenting information in a more
fun way, using animations or multimedia tools or by including puzzles or
memory games.

Major companies in the video game entertainment industry have also
created such games in order to bring together the interests of students and
players to a single place.

In 2018, Ubisoft launched Assassin’s Creed Origins: The Discovery Tour
(Ubisoft, 2018), a virtual museum with guided tours and historical sites to
discover that allows players to explore Ancient Egypt. This commercial game
has been used for conducting studies for comparing the learning outcomes of
a group following the game’s tours with those of a group learning about the
subject from a teacher, who used still images from the same tours. (Engadget,
2018)



2.4. Virtual environments applied to education 9

The results showed that those who had the teacher’s guidance improved
their grades up to 55 percent, concluding that a hybrid approach of video
games and traditional instruction could further enhance learning.

Figure 2.4: Assassin’s Creed Origins: The Discovery Tour (Ubisoft, 2018)

Other tools have also been used in schools in order to help children and
their teachers in the learning process. TEATRIX (Prada et al., 2000) was
designed in order to assist in the whole process of collaborative story creation.
It allows children to interact with each other in a distributed 3D environment,
choosing a character which will be controlled by the children or will act
autonomously according to the actions and goals set up by their role in the
story.

Figure 2.5: Teatrix (Prada et al., 2000)

b. Educational games

These games are explicitly created with the intention of teaching, without
trying to prioritize the diffusion of certain contents in the most playful way
possible.



2.4. Virtual environments applied to education 10

Construct3D (Kaufmann et al., 2000) is a three-dimensional geometric
construction tool that uses a stereoscopic head mounted display and a two-
handed 3D interaction tool that simplifies 3D model interaction. Its appli-
cation in mathematics and geometry education is interesting for its use at
high school as well as university level.

Figure 2.6: Construct3D (Kaufmann et al., 2000)

Another example is an educational tool for dance training. (Chan et al.,
2010) This game integrates virtual reality with a motion capture system to
analyze the user’s movement and give feedback. An interesting application
of this tool can be inside a classroom where the students have to follow the
teacher’s movements.

Figure 2.7: The motion capture system (Chan et al., 2010)

Finally, Dancing a Treasure (Romero-Hernandez et al., 2019) is another
educational game concerning Spanish dance, whose purpose is to motivate
young people to learn about this field, showcasing its basic moves and ele-
ments. It was developed for mobile devices, with the collaboration of the



2.4. Virtual environments applied to education 11

Spanish National Ballet company. (BNE) The game is composed of different
screens with several difficulty levels where the player must tap on the screen
to follow the dance moves being performed.

Figure 2.8: Dancing a treasure

c. Games for health

In the field of health, applications have been developed in which video
games serve as a vehicle for a variety of functions, such as training medical
students and general medical personnel. (Dinsmore et al., 1997)

Many of these games are designed as psychological therapy, being useful
for all kinds of patients. From Vietnam combat veterans with posttraumatic
stress disorder (Rothbaum et al., 1999), where patients are exposed to vir-
tual environments such as a Huey helicopter flying over a virtual Vietnam
and a clearing surrounded by jungle; to patients with social phobia and an-
xiety (Anderson et al., 2003), who are presented with a virtual classroom
environment in order to mitigate their speaking fears.

d. Companies, institutions and business corporations

Serious games are also used by companies to train their own employees
in new skills and resources. (Azadegan y ckh Riedel, 2012)

This type of games represents a very important time and money saving
for companies, in addition to reducing the usual rejection of employees to
this type of training. This training mode makes it easier for participants to
think of themselves as active rather than passive agents in that training,
while the trainer becomes a guide rather than a teacher, thus generating
greater dynamism in the process. (López Raventós, 2016)

JUST-TALK (Frank et al., 2002) is an application of responsive virtual
humans to train law enforcement personnel in dealing with subjects that
present symptoms of serious mental illness, providing a computerized virtual



2.4. Virtual environments applied to education 12

person to interact with the student in a role-playing environment.
Students will be able to conversate with the virtual person using spoken

natural language, and see and hear the virtual human’s responses: a com-
bination of facial gesture, body movements, and spoken language, showing
students basic techniques for managing encounters with the mentally ill. It
also teaches them to look for indications of particular forms of mental illness
so that they can adapt their responses appropriately.

Figure 2.9: JUST-TALK (Frank et al., 2002)

2.4.1.2. Serious games at Universidad Complutense de Madrid

a. The ComunicArte (PR2005-174/01) project

ComunicArte (PR2005-174/01) (ComunicArte) was created in the Uni-
versidad Complutense de Madrid (from now on UCM) with the aim of tea-
ching public speaking by using video games in VR environments. For this
aim, a team was created with members who had experience in the fields
of educational video games, computational narratology, affective computing,
art, psychology and VR. This project took place over two years, and ob-
tained financing from the Fundación BBVA, amongst nine other projects in
Spain. (BBVA, 2017)

The result of this research is a method of learning through an immersive
environment in which the player is taught how to deal with a real audience.
(El-Yamri et al., 2019) Information about the player is recorded -biometric
measurements, voice analysis, body movements and speech- and valued in
regards to a model of the virtual public, which allows to give the player advice
to improve his/her communication skills. This would serve as a prelude to
real audiences, which are far more difficult to obtain and where opportunities
to practice speaking skills are more limited.

Our project is closely related to ComunicArte, since it also consists on
the implementation of a virtual environment in which the player is presented



2.5. Video game creation tools 13

with an audience (in our case, a classroom with students) that would react
in a particular way according to what the player says or does, providing
feedback. As well as that, our video game also aims to improve the player’s
communication skills.

Figure 2.10: The ComunicArte project (ComunicArte)

b. The Didascalia (RTI2018-096401-A-I00) project

Didascalia (RTI2018-096401-A-I00) is a project created as an extension
of ComunicArte in the UCM, with the aim of creating and proving the ef-
ficiency of a virtual environment, in VR, that can be used as practice both
for students in teaching masters courses as well as teachers who are already
teaching in secondary school. Part of the members of Didascalia are members
of ComunicArte, including our directors.

Didascalia attempts to make use of the existing advances developed in
ComunicArte to create new scenarios set in a classroom, where teachers will
have to face questions or conflicts that may arise while giving a lecture,
receiving feedback from the system regarding their performance and com-
munication skills to solve said conflicts. However, the parameters to take
into account greatly differ from the previous project, since the new proposal
has a much more specific scope and will focus on teaching communicative
skills to a secondary-school teacher, not an ordinary speaker.

Our project is positioned inside Didascalia, acting as the core of the
development of its first prototype. Thus, the video game we want to develop
will be an initial environment that will promote the project and serve as a
base for the upcoming two years of research.

2.5. Video game creation tools

Video games are created thanks to game engines. A game engine is a set
of programming libraries that allow the design, creation and representation
of a video game.



2.5. Video game creation tools 14

One of the most important aspects when choosing a specific game engine
among all of the ones available in the market are the graphic capabilities,
because they are in charge of showing the 2D and 3D images on screen as
well as calculating some aspects such as the lightning and the textures.

Other characteristics to take into account are the ease of exporting the
game to different platforms or the usability of the engine.

In this case we are going to talk about Unity and Unreal Engine, currently
the two main game engines in the market.

2.5.1. Unity

Unity (Unity) is a game engine that allows the creation of 2D and 3D
video games. It is the most widely used tool for creating video games in the
world, and it is estimated that sixty percent of all virtual reality video games
are developed using this engine. (Bonfiglio, 2018)

Unity is a cross-platform game engine. The Unity Editor is supported
on Windows and macOS (it is still experimental on Linux). The engine sup-
ports creating games for more than 25 platforms that include mobile (iOS,
Android), desktop (Windows, Mac and Linux) or console (PlayStation, Xbox
or Nintendo).

The programming language that Unity uses is C#. Using this language,
the developer can build scripts which can be made for several purposes:

To adapt the game to player input.

To arrange events inside the game.

To set the behavior of the characters in the game.

To create visual effects.

Figure 2.11: The Unity game engine (Github For Unity)



2.5. Video game creation tools 15

Apart from this, Unity offers a service called Unity Asset Store (Unity
Asset Store) which allows developers to install Assets in their projects. Assets
can go from simple functionalities in the Unity IDE to large 3D files for a
video game in 4K definition.

2.5.2. Unreal Engine

Unreal Engine (Epic Games) is one of the most popular game engines.
It’s a complete suite of development tools based on the C++ language that
includes all the necessary elements to build a video game or an interactive
environment in virtual reality for several platforms, such as PC, PS4, iOS
and Android.

Initially, Unreal Engine emerged as an engine for the creation of video
games of the shooter type, but has evolved over time, expanding to different
game modes.

Unreal Engine presents a high degree of portability and is nowadays
widely used by a lot of video game designers. Its most recent version is Unreal
Engine 4, although Unreal Engine 5 is currently in development phase. (Epic
Games)

To create video games with Unreal Engine, the process consists on wor-
king together with the Unreal Editor and with programming, either by pro-
gramming in C++ directly or with Blueprints (Epic Games), which is an
ideal proposal for those who don’t have the knowledge of programming lan-
guages.

The Unreal Editor is an integrated development environment. It serves
as a user interface, where the developer sees the current state of the game
being created and the menus or tools available. With support for multi-user
editing (Epic Games), designers can simultaneously make changes to the
same Unreal Engine project in a safe and reliable way.

Figure 2.12: Unreal Engine (MasterD)



Chapter 3

Development and
implementation

Here we cover the whole process of creating and developing the video
game, following the instructions provided by education professionals.

3.1. Design of scenarios

This section explains the process to obtain the scenarios, from an initial
idea to a full description of the implementation details with Unity.

3.1.1. Selection criteria

The analysis was carried out by professionals from the Universidad Autò-
noma de Barcelona, who selected the scenarios based on a first approximation
to the records of an institute known by one of the experts that participate
in the project.

These reports included the most common infractions committed by stu-
dents of said institute, and its analysis will be explained in the next section.
The knowledge of the context was here a weight factor.

The content of the reports was insufficient for the design of the scena-
rios, since there was a lack of background information that is normally not
recorded on them.

The creation of the scenarios was mainly based on the behaviour of the
students. Among the most frequent incidents that the pedagogues found, we
selected two. One puts the core of the conflict on the teacher’s communication
with a student, while in the other one, the conflict involves more students.
In addition, a third situation was created that also focuses on a conflict
involving several students.

Since it is required to design interactive communicative situations bet-
ween the teacher and the students, complementary contents were proposed

16



3.1. Design of scenarios 17

to design the teacher’s answers.

3.1.2. Obtaining the reports

In order to gather all of the necessary reports, the schools were contacted
directly, through the educational psychologist or counsellor who is linked to
them. Every secondary school has such a figure.

Centers were selected based on two basic criteria: public ownership and
socially and culturally diverse population. Both criteria favour the represen-
tativeness of the population, both students and teachers. The second cri-
terion is fundamental because the socio-cultural composition of the groups
can theoretically be associated with a greater conflict. (Ridenour, 2004) The
communication process to obtain consent for participation was as follows:

First, in a brief interview, the researcher representing the project informs
about the objectives as well as the expectations regarding their collaboration.
Having obtained his or her approval, the school psychopedagogue informs the
director and, if he or she considers it viable, submits the application to the
faculty, since it is the latter who must give their consent to use their discipline
reports as the object of study. Throughout this chain of communication, the
project is presented through a piece of text (informed consent) that is finally
returned signed by the director of the center that agrees to collaborate in
the project.

Figure 3.1: Communication process diagram

There are three centers involved so far, all of them public: one in Bar-
celona and two in Vic (not named for ethical commitment). In two of the
centers the discipline reports were digitalized in an Excel sheet. In the other
one they were on paper and their content was also digitalized to facilitate



3.1. Design of scenarios 18

their analysis. Although they are very short reports, we were able to obtain
essential data to study the communicative situation that must be simula-
ted later in the scenarios: reason for the conflict, subject in which it occurs;
students involved (individual or group incident); time of the class in which
the conflict is detected (at the beginning, during the class, at the end) and
recidivism by the student or students involved in the conflict. In addition,
personal variables were recorded such as sex and course for students and the
experience of teachers (novice or expert).

Based on the results of this preliminary analysis, reports of serious in-
disciplines (e.g. those referring to mistreatment of people or social property)
were discarded and those that appeared more frequently in both boys and
girls and occurred in core subjects were considered representative, as seen in
Figure 3.2. Another criterion for selecting the report to be represented in the
scenario was the individual or group nature of the incident, which in turn
is related to the format of the communicative situation between the teacher
and the students involved in the conflict.

Figure 3.2: Analysis of the disciplinary records from one of the institutes

However, some limitations were recognized in this study:
Firstly, given the rush to achieve a technically viable prototype, the analy-

sis of the reports was not as rigorous as we would have liked. We are currently



3.1. Design of scenarios 19

conducting a more refined analysis to typify the conflicts based on the scien-
tific literature and through Atlas.ti (Atlas.ti), a data analysis software. This
procedure will give us better arguments to decide which situations should be
represented in the scenarios.

Secondly, informed consent for students was not initially considered ne-
cessary. However, we believe that they should also be informed, even if it is
without reference to any particular person. Apart from ethical considerations
that must be respected, in the next phase of the research, students, like tea-
chers, can provide useful information to understand conflicts based on their
experience and assess how they approach the different ways of managing
them.

3.1.3. Proposed scenarios

Three scenarios with a similar scheme were given to us by the educators.
Each of them began with the description of an initial conflict, where the
teacher had to make a choice. Depending on the action taken by the teacher,
the player can follow different paths, each one with positive or negative
consequences for the class. Afterward, feedback would be provided to the
players, informing them about their performance.

These scenarios were explained to us through virtual calls with the edu-
cators, as seen in Figure 3.3.

Figure 3.3: Screenshot of the meeting with the educators (02 / 13 / 2020)

Each of them was thoroughly described through several slides that con-
tained information about the scenario, with descriptive images and sample
discussion on the consequences of each decision that the teacher could make
(Figure 3.4).



3.1. Design of scenarios 20

Figure 3.4: A scenario as depicted in the slides from the educators

Apart from these slides, the situations were described in the form of a
flowchart made by the Barcelona pedagogues, one for each scenario, contai-
ning all the relevant information, as shown in Figures 3.5, 3.6 and 3.7.

3.1.3.1. First scenario: Inappropriate comment from a student

In this scenario a student says something aggressive towards the professor
out loud, so other classmates pay attention to him. He tends to make these
comments in class and is often warned by the professor to stop.

Figure 3.5: First scenario as proposed by the educators



3.1. Design of scenarios 21

3.1.3.2. Second scenario: Two students refuse to sit apart

In this case the problem is that two students are always sitting and
talking together, without paying attention to the class and disturbing their
classmates, who cannot focus on what the teacher says with the noise they
make.

Figure 3.6: Second scenario as proposed by the educators

3.1.3.3. Third scenario: A student is not welcomed by his or her
classmates

In this case the class is divided into groups, who are working together.
In one of the groups, a student is not well received by the rest of the group,
who do not pay any attention to him and laugh at him. He gets fed up and
ends up leaving the group.

Figure 3.7: Third scenario as proposed by the educators



3.1. Design of scenarios 22

3.1.4. Adaptation of the scenarios to Unity

The next step towards building the video game was to adapt these dia-
grams (Figures 3.8, 3.9 and 3.10) to make them suitable for their imple-
mentation. To that aim, we created a new flowchart for each scenario that
contained, for each possible situation, a description of how it would be re-
presented in the implementation.

3.1.4.1. First scenario

An example of the translation process in this scenario can be observed
by looking at the first path. In the original diagram, it says that the teacher
stops the class with a firm voice. Thus, in the adapted chart, we wrote “Detect
a firm voice tone”, something that we could identify with Unity.

Figure 3.8: First scenario adapted to Unity

3.1.4.2. Second scenario

As another example, in the right path from the original diagram of this
scenario, it is specified that the teacher approaches the students. The soft-
ware we used to develop the video game (Unity 3D) contains features that
allow monitoring if an object (in this case, the player) is within a specific
range of another (the students). Thus, in the new diagram, this procedure is
described as “Detect that the teacher approaches the students”, to ease the
implementation phase.

Furthermore, we decided that it could be interesting to detect if the
teacher speaks in a high tone to the students, despite it not being part of
the three original paths. Therefore, our pedagogues advised adding a fourth



3.1. Design of scenarios 23

path that considers this case, marked in green, to provide a better experience
to the player.

Figure 3.9: Second scenario adapted to Unity

3.1.4.3. Third scenario

In this last scenario, specifically in the second path, the teacher tells
the student to sit in the back of the class until the lesson ends. Therefore,
we translated it as detecting the words “end” or “back”, which would be
recognized by a Unity tool called Keyword Recognizer, that will be later
explained.

Figure 3.10: Third scenario adapted to Unity



3.2. Implementation 24

3.2. Implementation

When selecting a game engine to develop this project, we decided to
use Unity. One of the reasons of choosing this program over Unreal Engine
was that there is a bigger community in Unity, having a good number of
tutorials and manuals, as well as third-party content available for its use
(Unity Asset Store). Another advantage is that Unity is easier and faster to
learn, something crucial for us since we did not have any experience with
video game developing engines.

In this part, we detailed the components that were part of the implemen-
tation of the video game, covering from the objects to the scripts.

3.2.1. Student models and animations

We created models of students using Adobe Fuse (Adobe), which is a
graphics software creator that enables the creation of 3D characters in an
intuitive and easy way. This program allows users to create figures from
scratch and customize their body parts, clothing and facial expression. These
characters are then rigged with a model and can be exported to different
platforms.

Figure 3.11: Models created with Adobe Fuse (Wikiwand)

We decided to create a total of eight models. We found that eight students
were high enough to represent all the proposed scenarios, and low enough
so that it was not overwhelming for the teacher. Four of the students were
boys, and four were girls. Since according to the Spanish Ministry of Edu-
cation (Ministerio de Educación y Formación Profesional, 2019), 51% of all
secondary-school students are men, and 49% are women. All of them were
Caucasian, since that is the most common race among European schools,
with just 9% of international students in Spanish classrooms. (Ministerio de
Educación y Formación Profesional, 2019)

After creating all the students, we used Adobe Mixamo (Adobe Mixamo),
a tool used for animating the previous models. Mixamo is an online software
that imports the characters from Fuse and lets the user choose amongst
hundreds of different animations to assign them to the imported models,



3.2. Implementation 25

thus creating a realistic, animated avatar to include in the game. In this
manner, we were able to represent in a precise way disruptive behaviour of
students in the classroom, such as cheering, laughing or getting distracted.

However, some of the animations were not properly configured and shifted
the student’s position in one of its axis. Thus, we edited these animations
inside the Unity animation editor. As well as that, we searched for different
animations for each student, so they would behave slightly differently and
not move in the same way as the rest.

Figure 3.12: Animating a model in Mixamo

3.2.2. First draft of the class

Our idea was to create an initial environment with the eight students,
with some pieces of furniture like chairs and desks that were available in the
free asset packs in the Unity Asset Store (Unity Asset Store), as shown in
Figure 3.13. This would serve us as a prototype to get used to working with
Unity and as a base for developing a more complex scene later on, as it would
already have some functionalities such as the students´ initial animations.

We imported the models fromMixamo into Unity and gave each of them a
name: María, Dani, Alberto, Marta, Laura, Rebeca, Diego and Pablo. Then,
we placed each of them in a chair in front of a desk facing towards the player.

By creating two different scripts, PlayerMovement and MouseController,
as well as the Unity component Character Controller, we were able to create
an object that allowed the player to move around using the WASD keys
and look around using the mouse. Naturally, this would later be adapted for
using Virtual Reality glasses and a controller.

Another idea was to implement a system to make students move around
the classroom, for possible situations that may arise such as simulating a
misbehaving student. Initially, the student would move towards the direction
where the player clicked with the mouse. To make it more realistic, the



3.2. Implementation 26

student would first play the animation of standing up and then another
walking animation while getting to the destination, as well as a standing one
when he or she has arrived.

We used the NavMesh and KeywordRecognizer classes to add a couple
of features to our project. These two components will be explained in detail
further on.

Using the NavMesh object, the students would be able to walk around
the room freely but would not walk into or over chairs or desks.

Otherwise, the KeywordRecognizer class made it possible to integrate a
mechanism by which the video game could recognize certain words that the
player said. For example, when the player addressed a student by their name
or told them to stay quiet, and this way we could set up certain reactions
from the students accordingly. Moreover, we implemented a function that
would find the student and change their animation to make the student
raise their hand when addressed by the teacher.

Figure 3.13: The first draft of the class

3.2.3. Asset pack selection process

3.2.3.1. Initial asset pack

Once we had the first draft of the class done, we needed to adapt it to
real components of a class to represent everything with maximum precision.

First, we started looking for asset packs from the Unity Asset Store trying



3.2. Implementation 27

to find one that was quite complete. We looked at one in particular, which
was called “Snaps Art HD - School”. (Asset Store Originals) This pack used
HDRP (High Definition Render Pipeline) technology. (Unity Blogs)

The SRP (Scriptable Render Pipeline) is a system introduced by Unity
in 2018 which allows the users to customize the rendering of their project
based on their needs through C# scripts. This system is divided in two
main components: The LWRP (Lightweight Render Pipeline) and the HDRP
(High Definition Render Pipeline), both in preview phase. While the LWRP
is focused in optimizing the real-time performance, the HDRP aims for high
visual fidelity. Inside the HDRP you can customize aspects like lightning,
shadows and materials.

When we downloaded the project and included the students fromMixamo
in the class, several problems started to arise. First of all, the project was
very heavy because of the number of elements it had, as the pack included
almost a whole school built. Despite taking out everything but one class,
it still occupied a couple of GBs, making it more difficult to use a version
control system like GitHub.

The assets and textures also took up a lot of space because of their high
quality of definition. But what affected us the most was that Unity gave us
a problem when processing the children’s textures. They appeared without
any colour and we couldn’t solve it. We spent a couple of weeks trying to
figure out a solution to this, but we did not achieve anything.

Figure 3.14: The class with the first asset pack “Snaps Art HD - School”



3.2. Implementation 28

3.2.3.2. Final asset pack

As a result of these problems, we decided that the best thing to do was to
look for another pack that was much simpler and lighter, but above all, that
was compatible with a Unity 3D project without the use of HDRP, which
would allow us to integrate all the elements well.

Moreover, we decided to gather information about the characteristics of
a real classroom, so when designing the setting we took a typical Spanish
secondary-school classroom as a reference. Its main components were the
desks and chairs, the blackboard, and a platform for the teacher with a
lectern. To this end, we decided to use an asset pack from the Unity Asset
Store named “Classroom”, which included all the necessary elements.

Figure 3.15: Our class compared with a typical Spanish class

We decided to include thirty seats for students since that is the maximum
number of students per classroom according to Spanish law. (Ministerio de
Educación, 2010) Another design decision was to place the teacher in a plat-



3.2. Implementation 29

form, higher than the students, a common feature of classrooms that allows
the teacher to see and be seen by all the students. Initially, the teacher is
facing the students, as the scenarios require him/her to be able to observe
their behaviour.

Apart from that, we assigned to each student a label with his/her name
to help the player to be able to address them. Labels are important for the
communication between teachers and students since students can benefit
from perceiving that instructors know their names. (Cooper et al., 2017)

Figure 3.16: Students in the final class

3.2.4. Implementation of the menu

The next step was to build a menu based on a typical design, which
would be friendly and guiding for the player, and through which the different
scenarios could be accessed.

The first thing we did was create a new scene called Menu. Then, inside
this scene, two objects were created representing the main menu and the
submenu. Each one of these objects contained all the components we needed,
which were a canvas object, a background panel, the buttons and the texts
associated with the buttons.

The operation of the buttons of the two menus was as follows:

Main menu

• Jugar: this button led to a submenu that served to choose the
game scenario the player wanted to experience. The transition to
the submenu was done by hiding the main menu component and
showing the submenu component.

• Salir: this button exited the game.



3.2. Implementation 30

Figure 3.17: Main menu

Submenu

• Escenario 1, Escenario 2 and Escenario 3: these buttons loaded
the class scene with the corresponding scenario ready to be played
by means of a script.

• Volver: this button returned to the main menu. This action was
achieved hiding the submenu component and showing the main
menu component again.

Figure 3.18: Submenu

3.2.5. Tools

3.2.5.1. Controllers

With the aim of building a more realistic and immersive game, its cha-
racters would react according to the events of the scenario and the player’s
actions.



3.2. Implementation 31

In order to make this possible, Animator Controllers (Unity Documen-
tation) from Unity were used. These allowed us to arrange and maintain a
set of Animation Clips and associated Animation Transitions between them
using a State Machine, as shown in Figure 3.19.

In this way, the above-mentioned animations from Adobe Mixamo are
attached to each state, being able to move from one state to another by
activating different parameters.

Figure 3.19: Students´ Controller

Each student in Unity would have their own version of this Animation
Controller attached. By using an Override Controller (Unity Documenta-
tion), a custom Animation Controller for each character can be created,
which would maintain the original structure of the State Machine but repla-
ce each specific Animation Clip associated with each state with its own, as
can be seen in Figure 3.20.

Figure 3.20: Example of Override Controller

Therefore, thanks to these Controllers, we will be able to represent the
different reactions of the students during each scenario, when they have to
be distracted or get up from their seat to sit in another one.



3.2. Implementation 32

3.2.5.2. Colliders

The Colliders (Unity Documentation) were helpful in spotting various
events during the simulation. These Unity components surround the object
to which it is attached (this can be seen in Figure 3.21) and allow us to
detect physical collisions between this and other objects. For example, when
the player approaches a student, thus taking one of the paths available in
the scenario.

Figure 3.21: Sphere Collider, in green

Another example would be when a student arrives at the new seat that
has been set as his/her destination, therefore changing the active State in
his/her Controller and playing the sitting down animation, as shown in the
following image.

Figure 3.22: Sitting down animation being played after a Collider collision

3.2.5.3. NavMesh

In order to move the students around the class, we used a Unity compo-
nent called NavMesh. (Unity Documentation)

NavMesh is a class that can be used to do spatial queries, like pathfinding,
helping to represent an area where an agent is able to move. In other words,



3.2. Implementation 33

this component delimits the walkable surface in an area (see Figure 3.23), so
the agent can detect obstacles using artificial intelligence and find a path to
reach a destination.

The NavMesh agent is a component that will be attached to the students,
in order to allow them to navigate the Unity Scene using that NavMesh.

For the creation of the walkable surface, it is necessary to select the scene
geometry that should affect the navigation, i. e. the class floor. In addition,
to avoid that the agent goes through the elements of the environment, we
must specify those objects that we do not want to be walkable, the desks
and the chairs in this case, by adding them a NavMesh Modifier that will
take care of this.

However, since the students must be seated in those chairs at the begin-
ning of the class, the NavMesh agent attached to them must be turned off
at first, otherwise these students would be displaced to the nearest walkable
surface, being seated in the middle of the corridor.

The moment we want a student to get up from his chair and walk, we
must activate the agent and establish his/her destination by coordinates.

In the early development, the student would move towards the direction
where the player clicked with the mouse. Subsequently, the destination is set
manually depending on where we want the students to go on each scenario
(like another seat).

To make it more realistic, the student would first play the animation of
standing up and then a walking animation while getting to the destination,
as well as a sitting animation if the student has to sit somewhere else again.

Figure 3.23: Walkable Surface of the NavMesh, in blue

3.2.5.4. SoundLoudness

For measuring the player’s voice tone, we modified SoundLoudness, a class
implemented by other students from this Faculty in their project (Algaba
Aranda y Hernández Nájera-Alesón, 2019), which uses the Unity API to
measure the volume of a sound source (the microphone). For this purpose,
a sample file is recorded from the microphone and a normalized value is



3.2. Implementation 34

calculated using the root mean square, with the following formula:

rms =

√∑size
i=0 wavedata(i)2

size

Where size is the number of samples in the audio clip, and wavedata is
the collection of values inside the clip that was generated. The value of rms
is stored inside a list so that it can be used for different purposes throughout
the execution of the scenario.

Thanks to this value, we can determine if the player is speaking with a
higher or lower voice tone by comparing it to an adjustable threshold, as it
may change depending on the microphone or the ambient noise.

With this tool we will also be able to check if the player has overreacted
to the different events that occur during the scenario by comparing this noise
reaction to the average of his/her own voice tone.

3.2.5.5. KeywordRecognizer

KeywordRecognizer is a class from the Unity core (Unity Documenta-
tion), which is used for identifying specific words from an audio input. In
our case, it would be the microphone present in the virtual reality goggles
worn by the player.

This class contains a dictionary that can be provided with a word or set of
words along with actions to be performed, like playing different animations
or sounds, when each one is recognized. By using this dictionary, we can
adapt the behaviour of the students to what the player says, which is really
helpful for implementing the different paths inside each scenario.

3.2.5.6. Vokaturi

Due to the medical situation in Spain when developing this project, the
experimental phase was impossible to carry out, as mobility around the
country was restricted. Therefore, we could not travel to Barcelona to test
the tool with secondary-school teachers in institutes. Because of this, we
decided to add a new feature instead called Vokaturi.

Vokaturi (Vokaturi) is a software able to recognize directly from a voice
audio the emotions associated to it. In the open-source version of this tool, it
can detect five different emotions, which are happiness, sadness, neutrality,
anger and fear.

a. Efficiency analisis

According to the Vokaturi website (Vokaturi), on its open-source version,
the emotions recognition has a 66.5% of accuracy. However, this tool is



3.2. Implementation 35

designed for English speech, while we wanted to use it for Spanish teachers.
This could result in some emotions being mistaken for others.

With that in mind, we decided to analyse the accuracy of this tool, by
testing it with hundreds of audio clips in English and Spanish.

Concerning the English audios, we made use of an emotional speech da-
ta set from the University of Toronto (University of Toronto, 2010), which
contained 400 samples of audio clips for each emotion. These audio clips
consisted in an English speaker saying a word with a different intonation for
each of the emotions that Vokaturi could recognize.

For the Spanish audios, we used a multilingual emotional speech corpus
called EmoFilm (EmoFilm, 2018), from which we used the Spanish ones.
These audio clips were extracted from 43 films and were classified in anger,
happiness, sadness and fear. However, as there were no audio clips for neu-
trality, we recorded those remaining ones from native Spanish speakers with
the help of our directors.

For analysing all these audio clips, we modified a testing Python script
included in the Vokaturi open-source version in order to traverse all the clips,
obtaining their individual values and calculating a mean for each set of clips
of a specific emotion.

We created a correlation matrix for each language (Tables 3.1 and 3.2),
where in each row, the values of the recognized emotions are displayed for
a specific type of set of audio clips. The strongest emotion recognized was
marked in green, and the weakest in red:

Angry Sad Happy Fear Neutral
Angry 0,237 0,008 0,430 0,206 0,119
Sad 0,005 0,238 0,326 0,348 0,083

Happy 0,134 0,010 0,634 0,076 0,146
Fear 0,050 0,003 0,673 0,253 0,020

Neutral 0,015 0,126 0,259 0,079 0,521

Table 3.1: Correlation matrix for the English audios

Angry Sad Happy Fear Neutral
Angry 0,480 0,035 0,253 0,196 0,037
Sad 0,059 0,210 0,138 0,247 0,140

Happy 0,256 0,089 0,119 0,261 0,210
Fear 0,188 0,067 0,249 0,410 0,071

Neutral 0,022 0,213 0,066 0,121 0,578

Table 3.2: Correlation matrix for the Spanish audios



3.2. Implementation 36

Even though some of the emotions are acceptably recognized, the accu-
racy of the tool is far from perfect. Furthermore, in our analysis we can see
that with the Spanish audio clips, in three of the five emotions the prevai-
ling emotion is well recognized, while in the English clips, just two of the
five emotions are correct.

Taking all this into account, we decided not to give an important role to
the tool when choosing between the different paths but using it as an addi-
tional feedback feature at the end of the scenario, with just an informative
function.

b. Unity adaptation

As Vokaturi is implemented in Python, we had to adapt it to a Unity
version in C#. For this purpose, we used a Python environment that used the
IronPython library. This adaptation was carried out having another End of
Degree Project as a reference. (Algaba Aranda y Hernández Nájera-Alesón,
2019)

Once we could get Vokaturi working in our project, we measured the
emotions on the player’s speech in real time and stored the values in different
lists, one for each emotion. When the scenario was over, we calculated the
means for each emotion.

We decided to scale those values, normalizing them by dividing each one
by the biggest value. In this way, the predominant emotion would be set to
1 to represent it more clearly in the following graph.

We downloaded this graph model from another Unity Asset pack called
“Radar Chart” (MvPlanet, 2015), adapted it to use the above-mentioned
mean values and added the emoticons for a more visual effect. This graph
was included in the feedback screen at the end of each scenario. In Figure
3.24 we can see the comparison between the original graph model and its
more visual final adaptation.

Figure 3.24: Original and modified graph model

3.2.5.7. Virtual reality adaptation

The virtual reality adaptation process for our project was carried out in
the last place. We followed a guide (Oculus) to include the Oculus integration
package (Oculus) into Unity to adapt the player’s head movements to the



3.2. Implementation 37

camera in the game.
However, since we did not have access to the VR goggles itself due to

the medical situation, other member of the research team who had access
to them tested this integration and made some final adjustments to make it
work properly.

Figure 3.25: Member of the research team testing the tool

3.2.6. Audios

In this section we mention the audios we have used for our project, which
are five.

In order to accurately represent events in the classroom, we needed audios
to contribute to emulate the different scenarios. These audios brought some
realism to our work.

To do this, we searched the Internet for audio banks, such as SoundBi-
ble (SoundBible) or SoundSnap (SoundSnap). We found what we wanted
specifically in Freesound (Freesound), which is a huge collaborative databa-
se of audio snippets, samples, recordings, bleeps... released under Creative
Commons licenses that allow their reuse.

Figure 3.26: Freesound

From this website we downloaded three audios:

An audio that played ambient sound from a classroom, composed of



3.3. Explanation of the scenarios 38

whispering and conversations between students, to simulate they are
distracted.

An audio recording of students laughing, used in the first and third
scenarios.

An audio of an insult to simulate that the student insults the teacher
in the first path.

Lastly, we needed two more audios that were very specific so they couldn’t
be found on the Internet, so we recorded them by ourselves:

An audio of a student insulting and complaining about his classmates,
used in the third scenario.

An audio of a student explaining the reasons of his complaints to the
teacher, used also in the third scenario.

3.3. Explanation of the scenarios

3.3.1. Common structure

Here we describe some of the common screens and features for all of the
scenarios that we implemented.

3.3.1.1. Context screen

Every scenario begins with a context screen (Figure 3.27), which contains
the background information as specified in the flowcharts, to put the player
into perspective. After reading the story, the player will be able to click on
“Accept” to start playing. As two of the three scenarios have a maximum
reaction time, we included a variable to act as a counter for the number of
seconds elapsed since the start of the scenario, which would start increasing
as soon as the player clicked the “Accept” button.

Figure 3.27: Context screen in the second scenario



3.3. Explanation of the scenarios 39

3.3.1.2. TipsPanel

Another common feature of all scenarios was including a tips panel after
the initial reaction from the student, that showed different options for the
player as a guide, corresponding to the different paths that can be followed
in each scenario. Our first approach was not to include this panel and allow
the player to improvise, but after consulting the pedagogues from Barcelona
this was discarded. They recommended us to include this element so that
the teacher can be guided through the scenario. When the teacher makes a
choice, the selected option will be highlighted in green or red (right side of
Figure 3.28), depending if it is the right one or not.

Figure 3.28: Tips panel before and after the player’s choice

3.3.1.3. Feedback screen

All scenarios contained a feedback screen (Figure 3.29), which would ap-
pear when the scenario ends, with information regarding the choice made by
the player according to the flowchart, and whether it was the correct path
or not. The emotions graph implemented with Vokaturi is also shown, and
three buttons which allow the player to exit the game, try again and go back
to the scenario selection menu.

Figure 3.29: Feedback screen in the second scenario



3.3. Explanation of the scenarios 40

3.3.2. First scenario: Inappropriate comment from a student

3.3.2.1. Context and problem

The first scenario exemplifies the situation where a student makes an
inappropriate comment to try to get the class’s attention and cause an inte-
rruption. Specifically, in our game the student insults the teacher out loud.

This problem can result in a disorder in the classroom and, if the teacher
does not know how to solve it correctly, other students may follow the bad
example of their classmate, thus reducing the teacher’s effectiveness as a
consequence and, in consequence, the effectiveness of the teacher is reduced.
Therefore, the teacher has to make one of the following decisions detailed
below.

3.3.2.2. Paths

The first path is where the teacher, in a firm and determined tone,
tells the student that he/she will not tolerate such comments in the
classroom. This leads to a general silence of the class that then leads
to background noises when the teacher resumes the explanation, so the
problem is not solved correctly.

Figure 3.30: Students getting distracted after authoritative answer

In the second path the teacher approaches the student who has said
the inappropriate comment and tells him that when the class is over,
they will talk about it and try to find a solution. With this gesture of
the teacher the student feels attended, and the class continues. This is
considered the correct path.



3.3. Explanation of the scenarios 41

Figure 3.31: The teacher talking to the student calmly

The third path is taken if the teacher decides to ignore the student and
the comment he has made, because he/she thinks it is better not to
give it any importance. However, this is not an appropriate way to deal
with the problem, because students become distracted and background
noise is heard.

Figure 3.32: Students getting distracted after the teacher ignores the problem

We have uploaded a video demonstration of this scenario that shows all
of its different paths. (YouTube Video)



3.3. Explanation of the scenarios 42

3.3.2.3. Scenario script

In order to represent the inappropriate comment of the conflictive stu-
dent, we played an animation of that student shouting (see Figure 3.33) and
the audio itself.

Figure 3.33: Conflictive student animation

To differentiate between firm and calm tone, we used the SoundLoudness
tool (explained above) which measured the intensity of the sound.

In developing the first path our approach was as follows:
At first, we measured the voice tone after the student’s commentary.

However, to make sure it was more realistic, we changed it to measure the
voice tone all the time from the beginning of the scenario and make two
averages: the one of the audios before the comment and the one after the
comment. This way we could compare the two averages and make sure that
the player really does raise the voice tone.

Furthermore, just the moments in which the player really speaks were
included to make the average, in order to avoid the silence between words
or sentences from lowering the player’s voice tone average. We accomplished
this by means of an adjustable threshold that may change depending on the
ambient noise of the room.

Then we also encountered special cases on stage, such as if the teacher
speaks in a firm tone and then approaches the student, in which case we
take the first path, or if he speaks in a low tone and approaches the student,
where we would take the second path.

In addition, at the beginning we put the student’s name on the word
recognizer so that it would recognize whether the teacher was addressing
him, but later we realized that this was not decisive, since the teacher can



3.3. Explanation of the scenarios 43

say words in a strong tone that do not have to be the student’s name (such
as, “Shut up already”).

To check if the second path was taken, we created a Collider around the
student that would detect when the teacher approached him.

The third path would be taken when the player ignores the student; in
other words, if the player has not spoken with a high tone nor approached
the student after a time counter reaches its end.

In the first and third scenarios, animations of distraction would be played
in all the students and a background noise audio as well.

3.3.3. Second scenario: Two students refuse to sit apart

3.3.3.1. Context and problem

In this second scenario the following situation arose: two students when
they were together in previous classes did not work efficiently and talked
to each other. As a result, the teacher decides to give them a warning at
the beginning of the class to separate, but they do not listen to him. This
situation can result in a distraction for the class and it also can damage
the activity that the teacher is planning to do. With this context the scene
begins from where the player must decide which decision to take by following
one of the paths, each with different possibilities.

3.3.3.2. Paths

The first path involves repeating the warning to the students in an
assertive way, using words like “please” or “could you”. The consequence
of this action would be that the students refuse again, and the class
gets distracted and makes noise.



3.3. Explanation of the scenarios 44

Figure 3.34: Student getting distracted after the assertive petition

The second path is taken if the player decides to ignore the two stu-
dents and the problem that exists. By doing this, the situation is not
addressed so they continue to bother the class during the lesson.

Figure 3.35: Students making noise after the teacher ignores the conflict

In the third path the player approaches the two students to calmly and
closely comment on the fact that the previous day it did not work out
well for them to get together and therefore he is considering that it is
not the best thing to do. This path would be the right one, since it



3.3. Explanation of the scenarios 45

deals with the problem without losing any authority and in an effective
way at the same time.

Figure 3.36: Students talking with the teacher

In the first instance these were the three paths that existed, but we
were left wondering what would happen if the teacher addressed the two
students not in an assertive way or by approaching them to talk, but in
an authoritative way. As this was not contemplated in the scheme of the
scenario, we decided to add a fourth path explaining precisely this possibility.

We showed it to the professionals in Barcelona so that they could explain
to us in detail the consequence of carrying out the action of speaking in an
authoritative way to the students. They explained to us that in this case the
students would separate because they would feel intimidated by the teacher’s
way of addressing them, but due to this situation of tension an atmosphere
of noise would be generated in the class.



3.3. Explanation of the scenarios 46

Figure 3.37: Student walking to another desk

We have uploaded a video demonstration of this scenario that shows all
of its different paths. (YouTube Video)

3.3.3.3. Scenario script

For the first path we used the KeywordRecognizer class (explained before)
for identifying assertive words from audio input that, in our case, would be
the microphone present in the virtual reality headset worn by the player. The
action mapped to those assertive words would be the call to the FirstPath()
method, that would play the distraction animations and an ambient noise
from the students.

For the second path, in order to detect if the player ignores the students,
we created a timer which started from zero when the scene began. Once
the timer reached a specific value and the player had not done any of the
other actions, the same distraction animations and ambient noise from the
students would be played.

For the third one, in order to check the approach from the teacher to the
students to speak to them, we detected when the teacher entered a specific
area around them by means of a Collider. Then, the students would not
make any noise for the rest of the scenario.

Finally, in the fourth path, we detected authoritative words in the sa-
me way as in the first path, or a high voice tone. For the latter, we used
SoundLoudness (as in the previous scenario) to measure the volume of the
microphone. This allowed us to detect if the player is speaking in a tone hig-
her than a standard number of decibels. If this happened, one of the students
would move to another seat with a walking animation. However, after a few



3.3. Explanation of the scenarios 47

seconds, the students would start talking again, and the background noise
would continue, with an ambient noise audio clip.

3.3.4. Third scenario: A student is not welcomed by his class-
mates

3.3.4.1. Context and problem

In the third scenario, a student who is working in a group gets up and
confronts his classmates, who have been laughing at him throughout the
duration of the class. He is no longer interested in working nor talking to his
classmates. This poses several problems:

Firstly, the student who abandons the group will not be able to work on
the task that was assigned to him, and therefore may not be able to learn
certain contents of the subject that the group activity was meant to teach.
This will result in him being less prepared for the course and will lower his
productivity in class.

Secondly, the rest of the students in the group who were laughing at him
will continue to do so once he leaves the group, creating a hostile atmosphere
for the student who left the group. In addition to this, students who are
laughing are not focusing in the group activity and thus will not be able to
gain the required knowledge about the subject either.

In the video game, apart from an initial explanation the students are
shown in two different groups of four people each. After a few seconds, one
of the students gets up from his seat and an audio is played in which the
student calls the rest of the group “a bunch of jerks” and refuses to continue
sitting with them, standing up until the teacher decides to act. This audio
was recorded from an actual secondary-school student, in order to provide
a more realistic experience. Another animation is shown for the rest of the
group where they cheer and a laughing audio is played, to represent the
students laughing at the member who left the group.

3.3.4.2. Paths

Once the audio finishes, three different paths open up for the player:

This path is followed if the player tells the student to sit back down.
In this case, the student sits down where he was before, but the rest
of the group keeps laughing at him and he doesn’t work on the group
task since he is not motivated.



3.3. Explanation of the scenarios 48

Figure 3.38: Student sitting back down in his desk

The player tells the student to sit on the back of the class. The student
goes to the back of the class and sits on one of the empty chairs, and
after a few seconds the other students that were in his group continue
mocking him, without paying attention to the group activity.

Figure 3.39: Student walking to the back of the class

The player approaches the student to ask him what’s wrong. This path
is followed if the player gets close to him. Then, the student explains
that “his classmates are not working and are mocking him, and he refu-



3.3. Explanation of the scenarios 49

ses to work with those jerks” The group activity is thus finished. This
is considered the correct path since the teacher listens to the student,
who feels cared for, and the continuous mocking by his classmates stops
when they realize that the teacher is addressing the problem directly.

Figure 3.40: Student explaining the problem to the teacher

We have uploaded a video demonstration of this scenario that shows all
of its different paths. (YouTube Video)

3.3.4.3. Scenario script

For the first path, the KeywordRecognizer would recognize the words like
“sit” or “sit down” and then a sitting down animation would be played on the
student. In addition, animations for his classmates cheering and a laughing
audio clip would be played.

In the second path, the KeywordRecognizer is also used, so when the
player says words like “back” or “end”, the student would head to a seat
in the back of the class by a walking animation and would sit down there
with another one. However, the same animations for his classmates and the
laughing audio clip would be played.

Finally, the third path would be followed if the Collider object attached
to the player touches the student’s one, so an audio clip with the explana-
tion of the student would be played. In this case, no laughing audio clip or
animation from his classmates would be played.



Chapter 4

Individual contribution

4.1. Alejandro Díaz

To begin with, we were required to research different academic articles
concerning virtual and mixed reality technology applications. Each of us del-
ved into a specific area, which in my case was medical applications, including
simulators for the treatment of traumatic experiences and teaching tools for
future surgeons.

Once we chose to make an educative tool, we decided to create a first draft
of the classroom with free assets from the Asset Store and a few students
as a starting point. I was in charge of creating this environment in Unity,
importing the models and animations previously created, and placing them
around a scene to create a classroom. This was useful to get used to the basic
functionalities of the Unity engine.

Moreover, I implemented an initial mechanism for the students to move
around the classroom using the NavMesh technology, which made one of the
students stand up from his chair and move towards the position where the
player clicked, by writing a script to allow the player to do so as well as move
around the scene. For this aim, I had to divide the scene into different layers
to avoid the students from climbing onto the desks and chairs.

For the student to get up from his chair and walk realistically, I added
an Animation Controller that would manage its various states, making him
move his legs when he walks or stands up.

When it came to use the NavMesh technology for moving the students
in the final implementation of the scenarios, I established specific locations
to direct the students to other desks when it was necessary. Also, I placed
Colliders in the places where the student would have to sit down in order to
play the corresponding animation and rotate its model to face the blackboard
with a script.

During the analysis of the accuracy of Vokaturi, I normalized all the audio
clips that we recorded, by cutting them, equaling their volume and reducing

50



4.2. Álvaro López 51

their noise.
When writing the State of the Art section, I collected information on

virtual reality and its most common applications, since our project was aimed
to this technology since its inception.

Regarding the coordination with the pedagogues team from Barcelona,
I discussed the initial flowcharts of the three scenarios with them through
several video calls.

I wrote the introduction part of the paper we sent to EDULEARN20,
summarizing the problem of current education, explaining our approach and
mentioning the main objectives of the project. Also, I recorded myself pre-
senting the tool for the video.

Lastly, I recorded the audio clips for the third scenario, in which a student
gets up from his chair and talks to his classmates and to the teacher, with
the help of my brother who is currently a student in secondary school. This
made the scenario seem more realistic.

4.2. Álvaro López

I searched for information about tools used with educational purposes
during the search of a final topic for our project. I came across interesting
practical applications such as geometric construction tools based on augmen-
ted reality or story creation ones that helped children and their teachers to
collaborate.

During the initial development of the project, on its first draft, I handled
the integration of a word recognizer by using a Unity resource. In this way,
the students would be able to raise their hand when the teacher addressed
them by their names, something that was interesting as a first contact to the
interaction between the teacher and the students.

In the final implementation of the scenarios, I included several words or
short sentences to the KeywordRecognizer dictionaries that were essential to
detect some of the paths that the player could take when he/she talked to
the students.

In terms of the State of the Art, it was essential to clearly define what
video games are and investigate how they have been used as a learning tool
in all kinds of industries, including the most traditional ones, so I searched
for previous approaches on this topic.

Moreover, I used this information and researched more about the pro-
blems of communication in the education field for writing the context section
of the submitted EDULEARN20 paper. For the video presentation, I recor-
ded a demo of our tool for including it and edited the final video itself.

When studying the efficiency of Vokaturi, I modified one of its testing
scripts for conducting a performance analysis with all of the audio clips



4.3. Mario Bocos 52

that were available. To obtain the audios in Spanish, I found a multilingual
emotional speech corpus called EmoFilm, and requested the use of their
audios for our project.

Concerning the communication with the pedagogues, I was in charge of
sending them the translated flowcharts of the scenarios and modifying them
to obtain their final approval.

Finally, I made some modifications for unifying the animations by using
a general controller and overriding it for each student. In addition, I edited
some of those animations that made the models erratically move and searched
for different animations for each student so they would not always move in
the same way and at the same time, to provide a better experience.

4.3. Mario Bocos

I searched for other virtual reality application topics different from me-
dicine and education. I finally found some examples like a dance training
system using motion capture, a virtual storytelling system oriented to mu-
seum contexts or an application of responsive virtual humans to train law
enforcement personnel in dealing with subjects that present symptoms of
serious mental illness.

Furthermore, in the chapter of the State of the Art I researched in diffe-
rent forums and web pages about the different tools that exist today for the
creation of video games, delving into the main features and components of
the two most popular ones: Unity and Unreal Engine.

For the design of the different students who were part of our game with
Adobe Fuse, I first researched as much as I could about the program and how
it worked to explore all its possibilities. Once this was done, I created the
eight models of the children based on common European student profiles,
four girls and four boys. For this task I followed a refinement process until I
was happy with the result. Later, I learned to export these models toMixamo
to be able to animate them and after that integrate them into Unity.

I implemented another of the technologies that were part of our project,
that was SoundLoudness. I adapted the native class in order to use and store
the values obtained with it. By means of the UnityEvents I stored those
values in different places so we could compare them with the ones previously
received from the player or with a fixed value allowing us to know the level
of intensity of his/her voice.

Another task I did was to create the game’s menu. To do this I followed a
YouTube course about creating menus in Unity and I evaluated what would
be the best disposition of a menu for our work. Once I decided this, I created
all the buttons, texts, colors and operations of the menu itself.

In the inclusion of Vokaturi, I added the integration script to Unity and
the emotions chart that was part of the feedback screen, by customizing one



4.4. Collaborative work 53

of the pre-made charts of the asset.
I also got in contact with the team in Barcelona whenever different design

decisions arose, such as adding an extra path in one of the scenarios or
including a panel with the different options for the player, as well as how to
manage the emotion analysis.

Besides, I wrote the conclusions section of the paper explaining the main
results obtained from our project and the future work to be performed. With
respect to the video presentation, I composed the script which Alejandro
would later use to record himself.

4.4. Collaborative work

As a preparation for working with Unity, a game engine which we had
never used before, we took an online course in Udemy, which consisted in
the creation of a First-Person Shooter video game. This proved to be helpful
for us since it showed us the basic notions of using Unity, and some charac-
teristics of first-person video games that we implemented in our game, such
as moving the camera or the player around.

The first step in our project was to decide the topic that we were in-
terested in. To this aim, we had several meetings with our directors during
a “divergent phase”, where we discussed possible applications of a virtual
reality video game. Figure 4.1 shows the results of a brainstorming session
in which we wrote down possible ideas in post-it notes.

Figure 4.1: Results of the brainstorming session

When our project partners from Barcelona sent us the three flowcharts
for the scenarios, we took care of adapting them to something that we could
implement in Unity. Thus, we made a translation between the context and
actions in the scenarios and the components in Unity, coming up with three
new flowcharts that were approved by the educators.

With the help from our directors, we searched for reports about the
current situation of teacher training in Spain, which we reflected in the State



4.4. Collaborative work 54

of the art section.
Concerning the implementation of the game, all of us were involved in

most of the parts such as the context and feedback screens, as well as gat-
hering all the different technologies used to make the scenarios work in a
consistent way.

Apart from that, we made some extra modifications such as adding a
fourth path to the second scenario or showing a tips panel to the user with
every path available, which were implemented after consulting the experts
from Barcelona.

Before adding Vokaturi to our project, when carrying out an analysis
on its accuracy, we needed seventy-five audios in Spanish with a neutral
tone, since they were missing from the corpus we downloaded. To obtain
them, each of us and our directors recorded several male and female audio
clips, which were then edited adjusting their duration and removing any
background noise.

Another task that we performed in common was the addition of Oculus
Integration to our project. We found documentation regarding the integra-
tion of this asset to Unity and its configuration for our type of project. Since
we did not have access to the virtual reality goggles and controllers due to
the closure of our faculty, we asked one of the members of the research team
to test this new functionality by herself.

Regarding the EDULEARN20 paper, we all wrote the abstract, as it was
the first step in the acceptance process of our tool at the congress. Moreover,
we completed the development section and refined the other ones before
sending the paper.



Chapter 5

Conclusions and future work

In this section, we will go through the main conclusions that we have
obtained from our work, as well as cover the main objectives from the be-
ginning of the project and propose possible future work to further enhance
our application.

5.1. Conclusions

For many secondary-school teachers, handling conflictive situations inside
a class proves to be a real challenge, since they usually never receive any
training before confronting a real situation with real students. The teacher’s
behavior regarding oral communication plays a crucial role in developing a
positive relationship with a student (Campbell et al., 2004), since an educator
is a reference in the educational and personal development of the pupils (S
Yoon, 2002), especially at early ages.

The communication process between a teacher and a group of students
can be seen as that of a speaker and his/her audience. There are clear si-
milarities between them, in particular that the speaker must talk about a
certain subject and keep the attention of the audience by giving an interes-
ting speech.

The effect of a virtual audience in a speaker has proved to be a useful
tool in several previous works, helping to alleviate the fear of public speaking.
(El-Yamri et al., 2019) With this in mind, we developed this tool to make it
easier for future teachers to adapt to the events that take place in the real
world.

To do so, we added to the game many features that we extracted from
real situations to build a highly immersive environment, like the set of audios
or the animations used in the scenarios. Furthermore, we allowed a realis-
tic communication system, such as detecting the teacher’s position concer-
ning the students or spotting different emotions depending on the teacher’s
speech.

55



5.1. Conclusions 56

5.1.1. Accomplished objectives

Looking back at the objectives that we were considering in the beginning
of the project, we have accomplished the following:

We have developed a virtual reality video game with several functiona-
lities, such as voice and emotion recognition, that depicts a secondary-
school classroom in a precise way

We have integrated different conflictive behaviors that students can
show inside a real classroom, as well as different responses a teacher
could have and their consequences.

We have created a tool that will be useful for students of master’s
degree in teacher training, and for secondary-school teachers, to help
them deal with the difficult situations that may arise in a classroom,
providing them with feedback based on their actions.

In our view, these objectives meet what was expected from our project and,
as stated in Section 2.4.1.2.b, will serve as a starting point for the Didascalia
(RTI2018-096401-A-I00) project.

5.1.2. Main conclusions

The first conclusion is that realistic scenarios are essential to strengthen
the players´ (preservice teachers) immersion and to minimize the impact
when they arrive in a real classroom. Besides, this tool, as any simulator, will
permit the players to make mistakes within a virtual classroom with no risks,
something essential to improve their communication skills and raise their self-
confidence. In addition, players will be informed about their performance and
will be given a detailed feedback.

We found that the Unity game engine was a helpful software to create
this simulator, with a high level of accuracy and concreteness. It allowed
us to model the students in every way and create very detailed conflictive
situations, with different behaviors for each of them. This makes Unity a
valuable tool for creating virtual environments in the form of a video game,
especially for serious games.

In addition, modelling disruptive behavior of students based on peda-
gogical analysis makes the virtual students very realistic. The study of the
disciplinary reports from a real institute has helped us a great deal when
building the environment. Moreover, the combination of the artificial inte-
lligence (included in the development process) and the immersive features
allow us to create an accurate reflection of the pedagogues model.

Also, working alongside a team of people unfamiliar with computer en-
gineering and software development, pedagogues in this case, showed us the
process of communicating with other specialists and combining knowledges



5.2. Future work 57

of different fields in order to create a final product of high quality. This expe-
rience will serve us in the future as it is something that software developers
have to deal with frequently.

Finally, we find the use of virtual reality and AI, a novel and exciting path
to solving the problem that many educators face when they start teaching.
Besides, a videogame (an interactive approach) makes it more engaging and
accessible to users than traditional methods of learning, as seen in several
cases. (Manero et al., 2015) Furthermore, the pedagogical part of the team
reaction was positive after testing the video game, since they considered that
the scenarios were correctly represented. With that being said, we consider
this project a promising approach to be included in the program of master’s
degrees in teacher training.

Our tool could be helpful for this kind of training programs, as a practi-
cal experience to complement the current methods of teaching. Moreover, it
could be used as a reference for building highly immersive simulators concer-
ning any topic, by reusing some of its technologies such as scripts, NavMesh
or KeywordRecognizer.

5.2. Future work

Due to the impossibility of carrying out the experimental phase in this
project, the next step would be to test the application in a real institute
with actual lecturers, in order to refine the tool and prove its efficiency. By
recording the reactions from teachers who play the video game as well as their
feedback and impressions about the tool, we can obtain data to investigate
the most common obstacles that teachers find when dealing with conflicts in
classrooms.

Also, something that we were not able to finally include as we did not have
access to the goggles, was the adaptation of the virtual reality controllers,
in order to make the player move around the class and navigate through the
menu by using them.

As far as we have implemented only three scenarios, this initial project
will serve as a starting point for a more complex application, which can be
enhanced by other researchers or members of the Didascalia group. Further-
more, it can be a reference for other students who plan to develop similar
projects.

There would be several choices to improve the efficiency of this tool.
First, more scenarios with more elaborate interactions could be developed,
such as having more students involved or several conflicts occurring at the
same time. Secondly, the technologies used in the project such as Vokaturi
or the KeywordRecognizer could be upgraded, for instance by using a more
efficient emotion detection tool or a sentence recognizer instead of a word
one.





Appendices

Acceptance Letter

This is the Acceptance Letter received confirming that our abstract was
accepted at EDULEARN20, so we could follow the process by submitting
the paper.

59



25th of June, 2020

EDULEARN20 Local Organising Committee

ABSTRACT ACCEPTANCE LETTER

This is a confirmation that the abstract entitled:

“A VR GAME TO IMPROVE COMMUNICATION SKILLS IN SECONDARY-SCHOOL
TEACHERS”
Author(s): Mario Bocos-Corredor, Alejandro Diaz-Nieto, Alvaro Lopez-Garcia, Alejandro
Romero-Hernandez, Ibis Alvarez, Borja Manero

has been accepted as VIRTUAL presentation at EDULEARN20.

Name of event: EDULEARN20 (12th annual International Conference on Education and
New Learning Technologies)
Dates: 6th-7th of July, 2020
Organising entity: IATED

IATED Academy. email: edulearn20@iated.org – Tel.: (+34) 96 344 62 37
iated.org



Appendices 61

EDULEARN20 Paper

The following pages show the paper that we submitted to EDULEARN20,
once our abstract was accepted.



 A VR GAME TO IMPROVE COMMUNICATION SKILLS IN 
SECONDARY-SCHOOL TEACHERS 

Mario BOCOS-CORREDOR1, Alejandro DIAZ-NIETO1, Álvaro LOPEZ-GARCIA1, 
Alejandro ROMERO-HERNANDEZ1, Ibis ALVAREZ2, Borja MANERO1 

1Universidad Complutense de Madrid 
2Universitat Autónoma de Barcelona 

Abstract 

According to the reports of training practices from future secondary-school teachers, there are significant 
shortcomings in the communication skills required to properly manage a class. Despite the existence of 
modules related to communication in the educational masters of teaching staff, there is no real learning 
in this area, and students (the future teachers) express their fear when it comes to dealing with a class. 

Currently, virtual reality is present in different fields, such as medicine, engineering, humanities, etc. Its 
characteristics (in particular the one concerning the creation of a safe environment, its ubiquity, and its 
high immersion capacity) turn it into an ideal technology for training communication skills. In parallel, in 
the last ten years, videogames have proven their effectiveness as learning tools, alone or combined 
with other pedagogical strategies. 

For these reasons, we consider the following hypothesis: Virtual reality in combination with videogames 
is, on the one hand, an effective environment for simulating realistic situations related to secondary 
education, and, on the other, an ideal environment to practice and learn communication skills for the 
classroom. 

Taking this into account, we have created a virtual environment that serves as a training tool for future 
secondary-school teachers, to improve their communication skills and get used to a real class. It 
simulates a secondary-school class with students and works as a videogame. The virtual students work 
by simulating real students' behaviors in the classroom. 

Future teachers start the game in front of the class as in most secondary schools. The tool sets the 
game background, offering the complete context, actions, or previous behaviors (i.e., two students have 
been fighting all over the previous week). After, the user has to react as in a real situation, managing 
complicated situations. Our videogame analyses in real-time the user's behavior (voice tone, the 
distance between the user and the problematic student or students and detection of assertive or 
negative vocabulary) and reacts through the students themselves, i.e., offering realistic behavior to the 
user's actions. 

We are optimistic about the possibilities of our tool on the teachers' training since it avoids the expensive 
and logistical complicated real environments practices. The experts consulted agree on the added value 
that this tool can offer to practices in real schools. The videogame offers the perfect environment to 
learn and understand how to manage stressful situations through effective communication before facing 
real students. 

Keywords: Videogames, education, virtual reality, teaching, serious games. 

1 INTRODUCTION 

The lack of communication skills training in many developed countries is a fact. European Union has 
warned of the need to take action to enhance Europeans oral communication skills. Teachers are not 
an exception. [1] Around the world, teachers are not adequately trained, and their lack of communication 
skills is notorious. [2] As a result, many of them leave their jobs in their first year due to frustration and 
other mental health problems [3]. Particularly in Spain, there is no systematic treatment of 
communication issues, and most teachers do not receive specific training in oral expression. [4] 

Several approaches are trying to solve this aspect inside the master’s degrees in teacher training [5] [6]. 
However, the results obtained show that the development of these skills has significant shortcomings 
[7] as there is no specific subject to improve communication skills. Instead, basic notions are taught in 
other subjects, which are insufficient, and practical sessions on this topic are short and given by teachers 
who don’t have full knowledge of the issue. 



On the other hand, advances in technology have much contributed and in different ways to society, 
ranging from the treatment of mental health problems such as post-traumatic stress disorder [8], social 
anxiety [9] or phobias [10], to the training of workers [11], in addition to new ways of teaching students 
in different disciplines such as theatre [12], dance [13] or geometry [14]. One of the advances that have 
contributed most to making this possible is virtual reality, which makes it possible to create very 
immersive environments. [15] 

We wanted to find a way to help teachers face a classroom full of students who would behave 
unexpectedly. Our proposal is a virtual reality-based videogame. By developing a simulator with a virtual 
classroom full of students whose behavior can be scripted, we could provide teachers with different 
situations to see how they would react and give them feedback based on their actions. 

Our approach would help teachers to give an actual lecture and know which actions would have a better 
outcome for the class, as well as let them feel more relaxed and reduce their stress level. So, our 
videogame/simulator is addressed to future teachers who would want to experience how teaching a 
class is like.  

1.1 Objectives 

Our main goals throughout the process were: 

• To simulate a secondary education classroom with virtual students with realistic disruptive 
behaviors. 

• To develop a videogame which accurately reflects a real-life classroom with students who have 
disruptive behaviors in the classroom.  

• To provide teachers with a helpful tool which can help them prepare for managing classroom 
conflicts. 

• To allow future researchers to gather data on what would be the reactions from the teachers to 
the different situations presented to them, analyze this data and come up with a summary of the 
most common reactions to discuss with teachers. 

2 CONTEXT 

In this section we describe some of the most relevant topics in our project, providing a background to 
the motivations and tools used for developing this work. 

2.1 Education 

According to the ANPE (Spanish acronym of Independent union, at the service of public education 
teachers) syndicate, conflict in the classrooms occurs at younger and younger ages. [16] Moreover, the 
reports of training practices from future secondary-school teachers say that there are significant 
shortcomings in the communication skills required to manage a class properly. Despite the existence of 
modules related to communication in the educational masters of teaching staff, there is no real learning 
in this area, and students (the future teachers) express their fear when it comes to dealing with a class. 
[17] 

The teacher must know how to use conflict skilfully, regulate it, and treat it positively. Effective classroom 
management depends on the teacher’s ability in using an appropriate tone and encouraging the learners 
to cooperate in the classroom. For this purpose, mediation and negotiation are strongly advised [18]. 
Both are complementary ways of managing conflicts, with mediation being one way of carrying out the 
negotiation. These terms are especially important for mitigating the factors that cause violence in 
education. 

There are two basic types of negotiation: collaborative and competitive [19]. In collaborative 
negotiations, both parts look for an adequate solution to their demands. This way of negotiation should 
prevail in schools, since there is an affective bond and equality. 

Conversely, competitive negotiation is based on personal positioning without putting oneself in the place 
of the other part. This approach should be used with caution because it can promote more serious 
conflicts than the first ones. 



With our tool, we are going to try to identify how this negotiation is approached and give feedback to the 
users to improve their skills when it comes to dealing with conflicts. 

2.2 Serious games 

Literature usually define serious games as games used for purposes other than mere entertainment. 
The purpose of this type of games is creating learning environments that allow us to experiment with 
real-world problems. In this way, serious games help us explore different solutions to problems posed 
in real situations and discover the information and knowledge that would help to be able to intervene 
without fear of error. The lack of real-life consequences facilitates the decision-making process, teaching 
us the optimal choices in those situations, and helping the development of different skills. [20] Thus, 
videogames are a potent tool that makes possible the creation of virtual environments that would be 
impossible in the real world for reasons of safety, cost, time, etc. [21]  

Some industries realize the effectiveness of serious games (called simulators), including military, 
education, civil aviation or healthcare. [21] For instance, there are several tools companies use to train 
their employees in new skills and resources, such as basic techniques for managing different situations 
and give them feedback so they can adapt their responses appropriately. [11] 

2.3 Virtual reality applied to communication skills 

Virtual reality places the user inside a 3D artificial world in which he/she can interact with the 
environment as if it was real, allowing the creation of realistic scenarios for various purposes, such as 
education, entertainment, or social science. 

These characteristics have made it possible to apply virtual reality to communication skills [22] by 
building virtual environments with an audience that reacts to the player’s verbal and non-verbal 
language. Examples of this include its use for treating the fear of public speaking [23] or as a 
communicative medium between patients and therapists [24]. 

3 DEVELOPMENT PHASE 

Here we cover the whole process of creating and developing the videogame, following the instructions 
provided by education professionals. 

3.1 Design of scenarios 

This section explains the process to obtain the scenarios, from an initial idea to a full description of the 
implementation details with Unity.  

3.1.1 Selection criteria 

The analysis was carried out by professionals from the Universitat Autònoma de Barcelona, who 
selected the scenarios based on a first approximation to the records of an institute known by one of the 
experts that participate in the project. The knowledge of the context was here a weight factor. 

The content of the reports was insufficient for the design of the scenarios. Since there was a lack of 
background information that is not normally recorded on them.  

The creation of the scenarios mainly based on the behavior of the students. Among the most frequent 
incidents that the pedagogues found, we selected two. One puts the core of the conflict on the teacher's 
communication with a student, while in the other one, the conflict involves more students. In addition, a 
third situation was created that also focuses on a conflict involving several students. 

Since it is required to design interactive communicative situations (teacher-student), complementary 
contents were proposed to design teacher's answers. 

3.1.2 Proposed scenarios 

Three scenarios with a similar scheme were given to us by the educators. Each of them began with the 
description of an initial conflict, where the teacher had to make a choice. Depending on the action taken 
by the teacher, the user can follow different paths, each one with positive or negative consequences for 
the class. Afterward, feedback would be provided to the players, informing them about their 
performance. 



As shown in Figure 1, we describe every situation in the form of a flowchart, one for each scenario, 
containing all the relevant information. 

 

Figure 1. Second scenario as proposed. 

3.1.3 Adaptation of the scenarios 

The next step towards building the videogame was to adapt these diagrams to make them suitable for 
their implementation. To that aim, we created a new flowchart for each scenario that contains, for each 
possible situation, a description of how it would be represented in the implementation.   

As example, in the right path from Figure 1, it is specified that the teacher approaches the students. The 
software we used to develop the videogame (Unity 3D) contains features that allow monitoring if an 
object (in this case, the player) is within a specific range of another (the students). Thus, in the new 
diagram, this procedure is described as “Detect that the teacher approaches the students”, to ease the 
implementation phase. 

Furthermore, in the second scenario (see Figure 2), we decided that it could be interesting to detect if 
the teacher speaks in a high tone to the students, despite it not being part of the three original parts. 
Therefore, our pedagogues advised adding a fourth path that considers this case, marked in green, to 
provide a better experience to the player. 



 

Figure 2. Second scenario adapted. 

 

3.2 Implementation 

In this part, we detailed the components that were part of the implementation of the videogame, covering 
from the objects to the scripts. 

3.2.1 Environment 

When designing the setting, we took a typical Spanish secondary-school classroom as a reference. Its 
main components were the desks and chairs, the blackboard, and a platform for the teacher with a 
lectern. To this end, we decided to use an asset pack from the Unity Asset Store named “Classroom”, 
which included all the necessary elements. 

We decided to include thirty seats for students since that is the maximum number of students per 
classroom according to Spanish law [25]. Another design decision was to place the teacher in a platform, 
higher than the students, a common feature of classrooms that allows the teacher to see and be seen 
by all the students. Initially, the teacher is facing the students, as the scenarios require him/her to be 
able to observe their behavior. 



 

Figure 3. Simulated an actual secondary-school classroom. 

3.2.2 Students 

Next, we created models of students using Adobe Fuse [26]. This program allowed the creation of 3D 
objects in a fast and straightforward way, customizing their aspect and characteristics. Apart from that, 
we assigned to each student a label with a name to help the player to be able to address them. Labels 
are important for the communication between teachers and students since students can benefit from 
perceiving that instructors know their names [27]. 

We decided to include a total of eight students. We found that eight students were high enough to 
represent all the proposed scenarios, and low enough so that it was not overwhelming for the teacher. 
Four of the students were boys, and four were girls, since according to the Spanish Ministry of Education, 
51% of all secondary-school students are men, and 49% are women [28]. All of them were Caucasian, 
since that is the most common race among European schools, with just 9% of international students in 
Spanish classrooms [28]. 

 

Figure 4. Students from the teacher’s view. 

3.2.3 Students behaviour 

After creating all of the students, we used Adobe Mixamo [29], a tool used for animating the previous 
model. In this manner, we were able to represent in a precise way common students’ disruptive 
behaviors in the classroom, such as sitting, cheering, laughing, or getting distracted. 



To explain the behavior of the students, we go over the three scenarios, going into detail in the second 
one: 

Each scenario starts with a screen with text describing its context, to put the player in the situation. In 
the second scenario, the teacher told two students to sit apart, but they didn’t listen. When the player 
clicks on the Accept button, the game shows the class from the teachers' point of view. The platform in 
front of the students, with the name of the two conflictive ones highlighted in green. From that point on, 
the teacher can move around the class and speak.   

A timer starts, and the player must now follow one of the displayed options. Each option is the starting 
point for a path and will have consequences for the class. 

In this case, we follow the first path when the teacher speaks using assertive words. For this purpose, 
we used a word recognizer called KeywordRecognizer [30], a class from the Unity core, which is used 
for identifying specific words from audio input, that in our case, would be the microphone present in the 
virtual reality headset worn by the player. Said class contains a dictionary that maps a word or set of 
words to action inside the program. By using this dictionary, we can adapt the students’ behavior to what 
the player says. For instance, we added the word “punish” to the dictionary and the action of playing an 
animation in which one of the students moves to a different seat, out of fear of being grounded. 

In the second path for this scenario, the player ignores the students and continues the class. The game 
detects it when the timer reaches a specific value, and the player has not done any of the other actions. 
Consequently, we trigger the laughing animation and sounds of student laughter. 

In the third one, the player approaches the students to speak to them. We detect when the teacher 
enters a specific area around them. Then, the students feel calmer, and they don’t make any noise for 
the rest of the scenario, allowing for the class to continue. The game considers it the best path possible, 
as the teacher keeps his/her authority and does not let the students interrupt the class. 

Finally, we take the last case if the player speaks using authoritative words, detected in the same way 
as the first path, or with a high voice tone. For the latter, we used SoundLoudness [31], which uses the 
Unity API to measure the volume of a sound source (the microphone). For this purpose, we record a 
sample file from the microphone, and we calculate a normalized value between zero and one using the 
root mean square. This value is compared to a threshold, which may change depending on the 
microphone or the ambient noise. That comparison allows us to detect if the player is speaking in a tone 
higher than a certain number of decibels. As a result, one of the students moves to another seat since 
the teacher’s voice has frightened them. However, after a few seconds, the students start talking again, 
and the background noise continues, with an ambient noise audio clip. 

We also apply this feature in the first scenario, where a student makes an inappropriate comment about 
the teacher. In this case, we record the sound from the teacher before the comment, and then we 
compare to his/her initial reaction after it, to check if she overreacts. 

To move the student around the class, we used a Unity component called NavMesh, which represents 
the area where an agent (an object assigned to the students) can move. In other words, this component 
delimits the walkable surface in an area, so the agent can detect obstacles using AI (desks or other 
students) and find a path to reach a destination. We also use this component in the third scenario, where 
the teacher may tell the student to sit on the back of the class. 

Another of the tools that we implemented was Vokaturi [32], a library that, receiving an audio track, 
analyses and classifies the speech into 5 basic emotions: neutrality, happiness, anger, sadness, and 
fear. These emotions can have a value from zero to ten, indicating their level of intensity.  It helped us 
to analyse the emotions of the player, in addition to the other parameters, to be able to take everything 
into account when choosing the right path. For example, if the emotion of anger is recognized, the game 
takes the path of the authoritative words. 

When the scenario finishes, the players receive feedback, highlighting positive aspects, depending on 
their decisions.  Each scenario has a best path, where the teacher deals with the problem without losing 
his/her authority and without losing control. 

The record of the player's actions (preservice teacher) as well as the content of the feedback received 
will be the essential inputs to further develop teacher training in classroom conflict management. 

We made use of every tool described to adjust the scenarios to the flowcharts that were provided and 
subsequently adapted. 



4 CONCLUSIONS 

For many secondary-school teachers, handling conflictive situations inside a class proves to be a real 
challenge, since they usually never receive any training before confronting a real situation with real 
students. The teacher's behavior regarding oral communication plays a crucial role in developing a 
positive relationship with a student. [33]  

The effect of a virtual audience in a speaker has proved to be a useful tool in previous works, helping to 
alleviate the fear of public speaking. [34] With this in mind, we developed this tool to make it easier for 
future teachers to adapt to the events that take place in the real world. To do so, we added to the game 
many features that we extracted from real situations to build a highly immersive environment. Further, 
we allowed a realistic communication system, such as detecting the teacher's position concerning the 
students, or spotting different emotions depending on the teacher's speech.   

The first conclusion is that realistic scenarios are essential to strengthen the players' (preservice 
teachers) immersion and to minimize the impact when they arrive in a real classroom. Besides, this tool, 
as any simulator, will permit the players to make mistakes (within a virtual classroom with no risks), 
something essential to improve their communication skills. 

Secondly, modelling student's disruptive behavior based on pedagogical analysis makes the virtual 
students very realistic. The combination of the AI (included in the development process) and the 
animations allow us to create an accurate reflection of the pedagogues' model. 

Finally, we find the use of virtual reality and artificial intelligence, a novel and exciting path to solving the 
problem that many educators face when they start teaching. Besides, a videogame (an interactive 
approach) makes it more engaging and more accessible to future teachers than traditional methods of 
learning.   

Furthermore, the pedagogical part of the team reaction was positive after testing the videogame.  We 
consider it a promising approach to be included in the program of master’s degrees in teacher training. 

5 FUTURE WORK 

The next step to take is to experiment with the application, to refine the tool and prove its efficiency. By 
recording the reactions from teachers who play the videogame as well as their feedback and 
impressions about the tool, we can obtain data to investigate the most common obstacles that teachers 
find when dealing with conflicts.  

As far as we have implemented only three scenarios, this initial project will serve as a starting point. We 
should develop more scenarios with more complex interactions to improve the efficiency of this tool. 

ACKNOWLEDGEMENTS 

This project has been partially founded by BBVA foundation (ComunicArte project: PR2005-174/01), 
and by the Ministry of Science, Innovation and Universities of Spain (Didascalias, RTI2018-096401-A-
I00). 

REFERENCES 

[1] European Comission, “Skills and qualifications in Europe.” [Online]. Available: 
https://ec.europa.eu/social/main.jsp?catId=1146&langId=en. 

[2] M. Hoffman, J. Ferri, C. Sison, D. Roter, L. Schapira, and W. Baile, “Teaching communication 
skills: an AACE survey of oncology training programs,” J. Cancer Educ., vol. 19, no. 4, pp. 220–
224, 2004. 

[3] M. Fukuda, H.-H. Huang, N. Ohta, and K. Kuwabara, “Proposal of a parameterized atmosphere 
generation model in a virtual classroom,” in Proceedings of the 5th International Conference on 
Human Agent Interaction, 2017, pp. 11–16. 

[4] I. García Aróstegui, “Competencias comunicativas del maestro en formación,” 2008. 

[5] Universidad Complutense de Madrid, “Competencias del Máster de Formación de Profesorado 
de la UCM.” 



[6] U. de Barcelona, “Objectives and competences in master’s degree in Teacher Training in 
Obligatory Secondary and Upper Secondary School Education, Vocational Training and 
Languages.” 

[7] J. D. Segovia, J. L. G. Ortega, I. G. Aróstegui, and A. R. Fuentes, “Competencias comunicativas 
de maestros en formación,” Profesorado. Rev. Currículum y Form. Profr., vol. 14, no. 2, pp. 303–
323, 2010. 

[8] B. O. Rothbaum et al., “Virtual reality exposure therapy for PTSD Vietnam veterans: A case 
study,” J. Trauma. Stress Off. Publ. Int. Soc. Trauma. Stress Stud., vol. 12, no. 2, pp. 263–271, 
1999. 

[9] P. Anderson, B. O. Rothbaum, and L. F. Hodges, “Virtual reality exposure in the treatment of 
social anxiety,” Cogn. Behav. Pract., vol. 10, no. 3, pp. 240–247, 2003. 

[10] B. O. Rothbaum, L. Hodges, B. A. Watson, G. D. Kessler, and D. Opdyke, “Virtual reality 
exposure therapy in the treatment of fear of flying: A case report,” Behav. Res. Ther., vol. 34, no. 
5–6, pp. 477–481, 1996. 

[11] G. Frank, C. Guinn, R. Hubal, P. Pope, M. Stanford, and D. Lamm-Weisel, “Just-talk: An 
application of responsive virtual human technology,” in Proceedings of the Interservice/Industry 
Training, Simulation and Education Conference, 2002, pp. 773–779. 

[12] R. Prada, I. Machado, and A. Paiva, “Teatrix: Virtual environment for story creation,” in 
International Conference on Intelligent Tutoring Systems, 2000, pp. 464–473. 

[13] J. C. P. Chan, H. Leung, J. K. T. Tang, and T. Komura, “A virtual reality dance training system 
using motion capture technology,” IEEE Trans. Learn. Technol., vol. 4, no. 2, pp. 187–195, 2010. 

[14] H. Kaufmann, D. Schmalstieg, and M. Wagner, “Construct3D: a virtual reality application for 
mathematics and geometry education,” Educ. Inf. Technol., vol. 5, no. 4, pp. 263–276, 2000. 

[15] J. L. Rubio-Tamayo, M. Gertrudix Barrio, and F. García García, “Immersive environments and 
virtual reality: Systematic review and advances in communication, interaction and simulation,” 
Multimodal Technol. Interact., vol. 1, no. 4, p. 21, 2017. 

[16] O. R. Sanmartín, “Profesores denuncian que los conflictos en las aulas con niños de Primaria 
superan ya a los de Secundaria,” EL MUNDO, 2018. 

[17] M. Flores and J. Fernández-Castro, “Creencias de los profesores y estrés docente en función 
de la experiencia profesional Teacher’s beliefs and stress as a function of professional 
experience,” Estud. Psicol., vol. 25, pp. 343–357, Nov. 2004. 

[18] T. MAHVAR, M. A. FARAHANI, and A. ARYANKHESAL, “Conflict management strategies in 
coping with students’ disruptive behaviors in the classroom: Systematized review,” J. Adv. Med. 
Educ. Prof., vol. 6, no. 3, p. 102, 2018. 

[19] S. P. Thomas, “Competitive versus collaborative: exploring the negotiation strategy impact on 
relational outcomes in ongoing buyer-supplier relationships,” 2013. 

[20] C. López Raventós, “El videojuego como herramienta educativa. Posibilidades y problemáticas 
acerca de los serious games,” Apert. (Guadalajara, Jal.), vol. 8, no. 1, p. 0, 2016. 

[21] T. Susi, M. Johannesson, and P. Backlund, “Serious games: An overview.” Institutionen för 
kommunikation och information, 2007. 

[22] F. Biocca and M. R. Levy, “Virtual reality as a communication system,” Commun. age virtual 
Real., pp. 15–31, 1995. 

[23] M. M. North, S. M. North, and J. R. Coble, “Virtual reality therapy: an effective treatment for the 
fear of public speaking,” Int. J. Virtual Real., vol. 3, no. 3, pp. 1–6, 1998. 

[24] G. Riva, E. Molinari, and F. Vincelli, “Interaction and presence in the clinical relationship: virtual 
reality (VR) as communicative medium between patient and therapist,” IEEE Trans. Inf. Technol. 
Biomed., vol. 6, no. 3, pp. 198–205, 2002. 

[25] Ministerio de Educación, “Real Decreto 132/2010, de 12 de febrero, por el que se establecen los 
requisitos mínimos de los centros que impartan las enseñanzas del segundo ciclo de la 
educación infantil, la educación primaria y la educación secundaria.” [Online]. Available: 



https://www.boe.es/buscar/pdf/2010/BOE-A-2010-4132-consolidado.pdf. 

[26] Adobe, “Adobe Fuse Web Page.” [Online]. Available: 
https://www.adobe.com/uk/products/fuse.html. 

[27] K. M. Cooper, B. Haney, A. Krieg, and S. E. Brownell, “What’s in a name? The importance of 
students perceiving that an instructor knows their names in a high-enrollment biology classroom,” 
CBE—Life Sci. Educ., vol. 16, no. 1, p. ar8, 2017. 

[28] M. de E. y F. P. Profesional, “Datos y cifras - Curso escolar 19/20,” 2019. 

[29] Adobe, “Adobe Mixamo Web Page.” [Online]. Available: https://www.mixamo.com/#/. 

[30] Microsoft, “Voice input in Unity.” [Online]. Available: https://docs.microsoft.com/en-
us/windows/mixed-reality/voice-input-in-unity. 

[31] J. Algaba Aranda and M. Hernández Nájera-Alesón, “Desarrollo de una aplicación para analizar 
y evaluar el comportamiento de oradores en realidad virtual,” 2019. 

[32] Vokaturi, “Vokaturi emotion recognition software.” 

[33] R. J. Campbell, L. Kyriakides, D. Muijs, and W. Robinson, Assessing teacher effectiveness: 
Developing a differentiated model. Psychology Press, 2004. 

[34] M. El-Yamri, A. Romero-Hernandez, M. Gonzalez-Riojo, and B. Manero, “Emotions-Responsive 
Audiences for VR Public Speaking Simulators Based on the Speakers’ Voice,” in 2019 IEEE 19th 
International Conference on Advanced Learning Technologies (ICALT), 2019, vol. 2161, pp. 
349–353. 

 





References

ABC. Más de la mitad de los profesores españoles no llega formado en el
contenido, la pedagogía y la práctica de la materia que imparte. 2019.
https://bit.ly/2A2Qc4s.

Adobe. Adobe Fuse Web Page. 2020. https://www.adobe.com/uk/
products/fuse.html.

Adobe Mixamo. Adobe Mixamo Web Page. 2020. https://www.mixamo.
com/#/.

Algaba Aranda, J. y Hernández Nájera-Alesón, M. Desarrollo de
una aplicación para analizar y evaluar el comportamiento de oradores en
realidad virtual. 2019.

Anderson, P., Rothbaum, B. O. y Hodges, L. F. Virtual reality expo-
sure in the treatment of social anxiety. Cognitive and Behavioral Practice,
vol. 10(3), páginas 240–247, 2003. ISSN 1077-7229.

Asset Store Originals. Snaps Art HD | School from Unity Asset Sto-
re. 2020. https://assetstore.unity.com/packages/templates/packs/
snaps-art-hd-school-158106.

Atlas.ti. Atlas.ti Main Page. 2020. https://atlasti.com/.

Azadegan, A. y ckh Riedel, J. Serious games integration in companies:
a research and application framework. En 2012 IEEE 12th International
Conference on Advanced Learning Technologies, páginas 485–487. IEEE,
2012. ISBN 1467316423.

BBVA. Concedidas diez Ayudas a Equipos de Investigación Científica en las
áreas de Economía y Sociedad Digital y de Humanidades Digitales. 2017.
https://bit.ly/31cOJna.

BNE. BALLET NACIONAL DE ESPAÑA. 2020. http://
balletnacional.mcu.es/.

Bonfiglio, N. DeepMind partners with gaming company for AI research.
2018. https://www.dailydot.com/debug/unity-deempind-ai/.

73

https://bit.ly/2A2Qc4s
https://www.adobe.com/uk/products/fuse.html
https://www.adobe.com/uk/products/fuse.html
https://www.mixamo.com/#/
https://www.mixamo.com/#/
https://assetstore.unity.com/packages/templates/packs/snaps-art-hd-school-158106
https://assetstore.unity.com/packages/templates/packs/snaps-art-hd-school-158106
https://atlasti.com/
https://bit.ly/31cOJna
http://balletnacional.mcu.es/
http://balletnacional.mcu.es/
https://www.dailydot.com/debug/unity-deempind-ai/


References 74

Campbell, R. J., Kyriakides, L., Muijs, D. y Robinson, W. Assessing
teacher effectiveness: Developing a differentiated model . Psychology Press,
2004. ISBN 0415304784.

Chan, J. C. P., Leung, H., Tang, J. K. T. y Komura, T. A virtual
reality dance training system using motion capture technology. IEEE
Transactions on Learning Technologies, vol. 4(2), páginas 187–195, 2010.
ISSN 1939-1382.

ComunicArte. ComunicArte: Comunicación efectiva a través de la realidad
virtual y las tecnologías educativas. 2020. http://nil.fdi.ucm.es/?q=
projects/comunicarte.

Cooper, K. M., Haney, B., Krieg, A. y Brownell, S. E. What’s in
a name? The importance of students perceiving that an instructor knows
their names in a high-enrollment biology classroom. CBE-Life Sciences
Education, vol. 16(1), página ar8, 2017. ISSN 1931-7913.

Dinsmore, M., Langrana, N., Burdea, G. y Ladeji, J. Virtual reality
training simulation for palpation of subsurface tumors. En Proceedings of
IEEE 1997 Annual International Symposium on Virtual Reality , páginas
54–60. IEEE, 1997. ISBN 0818678437.

Edulearn. EDULEARN20 - International Education Conference. 2020.
https://iated.org/edulearn/.

El-Yamri, M., Romero-Hernandez, A., Gonzalez-Riojo, M. y Ma-
nero, B. Emotions-Responsive Audiences for VR Public Speaking Simu-
lators Based on the Speakers´ Voice. En 2019 IEEE 19th International
Conference on Advanced Learning Technologies (ICALT), vol. 2161, pági-
nas 349–353. IEEE, 2019. ISBN 1728134854.

EmoFilm. EmoFilm - A multilingual emotional speech corpus. 2018. https:
//zenodo.org/record/1326428#.XvIruedS9hE.

Engadget. ‘Assassin’s Creed Origins´ virtual tours can actually
teach history. Engadget , 2018. https://www.engadget.com/
2018-02-13-assassins-creed-origins-discovery-tour.html.

Epic Games. A first look at Unreal Engine 5. 2020a. https://www.
unrealengine.com/en-US/blog/a-first-look-at-unreal-engine-5.

Epic Games. Blueprints Visual Scripting. 2020b. https://docs.
unrealengine.com/en-US/Engine/Blueprints/index.html{#}:{~}:
text=TheBlueprintsVisualScriptingsystem,orobjectsintheengine.

Epic Games. Unreal Engine documentation on Multi-User Edi-
ting. 2020c. https://docs.unrealengine.com/en-US/Engine/Editor/
MultiUser/index.html.

http://nil.fdi.ucm.es/?q=projects/comunicarte
http://nil.fdi.ucm.es/?q=projects/comunicarte
https://iated.org/edulearn/
https://zenodo.org/record/1326428#.XvIruedS9hE
https://zenodo.org/record/1326428#.XvIruedS9hE
https://www.engadget.com/2018-02-13-assassins-creed-origins-discovery-tour.html
https://www.engadget.com/2018-02-13-assassins-creed-origins-discovery-tour.html
https://www.unrealengine.com/en-US/blog/a-first-look-at-unreal-engine-5
https://www.unrealengine.com/en-US/blog/a-first-look-at-unreal-engine-5
https://docs.unrealengine.com/en-US/Engine/Blueprints/index.html{#}:{~}:text=The Blueprints Visual Scripting system,or objects in the engine.
https://docs.unrealengine.com/en-US/Engine/Blueprints/index.html{#}:{~}:text=The Blueprints Visual Scripting system,or objects in the engine.
https://docs.unrealengine.com/en-US/Engine/Blueprints/index.html{#}:{~}:text=The Blueprints Visual Scripting system,or objects in the engine.
https://docs.unrealengine.com/en-US/Engine/Editor/MultiUser/index.html
https://docs.unrealengine.com/en-US/Engine/Editor/MultiUser/index.html


References 75

Epic Games. Unreal Engine Web Page. 2020d. https://www.
unrealengine.com/en-US/.

Esposito, N. A short and simple definition of what a videogame is. 2005.

Europa Press. La conflictividad en el aula va “cada vez a edades más
tempranas” y ANPE insta al profesorado a “denunciar”. Europa Press
Galicia, 2019. https://bit.ly/31agqNp.

European Comission. Skills and qualifications in Europe. 2020. https:
//ec.europa.eu/social/main.jsp?catId=1146&langId=en.

Flores, M. y Fernández-Castro, J. Creencias de los profesores y estrés
docente en función de la experiencia profesional Teacher’s beliefs and stress
as a function of professional experience. Estudios de Psicología, vol. 25,
páginas 343–357, 2004.

Frank, G., Guinn, C., Hubal, R., Pope, P., Stanford, M. y Lamm-
Weisel, D. Just-talk: An application of responsive virtual human tech-
nology. En Proceedings of the Interservice/Industry Training, Simulation
and Education Conference, páginas 773–779. Citeseer, 2002.

Freesound. Freesound Web Page. 2020. https://freesound.org/.

Fukuda, M., Huang, H.-H., Ohta, N. y Kuwabara, K. Proposal of a
parameterized atmosphere generation model in a virtual classroom. En
Proceedings of the 5th International Conference on Human Agent Interac-
tion, páginas 11–16. 2017.

García Aróstegui, I. Competencias comunicativas del maestro en forma-
ción. 2008.

Github For Unity. Github for Unity. 2020. https://unity.github.
com/.

Hoffman, M., Ferri, J., Sison, C., Roter, D., Schapira, L. y Baile,
W. Teaching communication skills: an AACE survey of oncology training
programs. Journal of Cancer Education, vol. 19(4), páginas 220–224, 2004.
ISSN 0885-8195.

HTC. VIVE. 2020. https://www.vive.com/eu/.

Kaufmann, H., Schmalstieg, D. y Wagner, M. Construct3D: a virtual
reality application for mathematics and geometry education. Education
and information technologies, vol. 5(4), páginas 263–276, 2000. ISSN 1360-
2357.

https://www.unrealengine.com/en-US/
https://www.unrealengine.com/en-US/
https://bit.ly/31agqNp
https://ec.europa.eu/social/main.jsp?catId=1146&langId=en
https://ec.europa.eu/social/main.jsp?catId=1146&langId=en
https://freesound.org/
https://unity.github.com/
https://unity.github.com/
https://www.vive.com/eu/


References 76

López Raventós, C. El videojuego como herramienta educativa. Posibili-
dades y problemáticas acerca de los serious games. Apertura (Guadalajara,
Jal.), vol. 8(1), página 0, 2016. ISSN 1665-6180.

MAHVAR, T., FARAHANI, M. A. y ARYANKHESAL, A. Conflict
management strategies in coping with students´ disruptive behaviors in
the classroom: Systematized review. Journal of Advances in Medical Edu-
cation & Professionalism, vol. 6(3), página 102, 2018.

Manero, B., Torrente, J., Serrano, Á., Martínez-Ortiz, I. y
Fernández-Manjón, B. Can educational video games increase high
school students´ interest in theatre? Computers & education, vol. 87, pá-
ginas 182–191, 2015. ISSN 0360-1315.

MasterD. ¿Qué es Unreal Engine y para qué sirve? 2019. https://www.
masterd.es/blog/que-es-unreal-engine-y-para-que-sirve/.

Ministerio de Educación. Real Decreto 132/2010, de 12 de febrero, por
el que se establecen los requisitos mínimos de los centros que impartan
las enseñanzas del segundo ciclo de la educación infantil, la educación
primaria y la educación secundaria. 2010. https://www.boe.es/buscar/
pdf/2010/BOE-A-2010-4132-consolidado.pdf.

Ministerio de Educación y Formación Profesional. Datos y ci-
fras - Curso escolar 19/20. Informe técnico, Ministerio de Educación y
Formación Profesional, 2019.

Minotti, M. NPD: U.S. game sales hit a record $43.4 billion in
2018. Venture Beat , 2019. https://venturebeat.com/2019/01/22/
npd-u-s-game-sales-hit-a-record-43-4-billion-in-2018/.

MvPlanet. Rdar Chart in Unity Asset Store. 2015. https://assetstore.
unity.com/packages/tools/gui/radar-chart-50074.

Oculus. Oculus Integration - Unity Asset Store. 2020a.
https://assetstore.unity.com/packages/tools/integration/
oculus-integration-82022.

Oculus. Oculus integration guide in Unity. 2020b. https://developer.
oculus.com/documentation/unity/unity-gs-overview/.

Oculus. Oculus Main Page. 2020c. https://www.oculus.com/?locale=
es_ES.

Okan, Z. Edutainment: Is learning at risk? British Journal of Educational
Technology , vol. 34(3), páginas 255–264, 2003. ISSN 0007-1013.

Pearce, C. Towards a game theory of game. First person: New media as
story, performance, and game, vol. 1, páginas 143–153, 2004.

https://www.masterd.es/blog/que-es-unreal-engine-y-para-que-sirve/
https://www.masterd.es/blog/que-es-unreal-engine-y-para-que-sirve/
https://www.boe.es/buscar/pdf/2010/BOE-A-2010-4132-consolidado.pdf
https://www.boe.es/buscar/pdf/2010/BOE-A-2010-4132-consolidado.pdf
https://venturebeat.com/2019/01/22/npd-u-s-game-sales-hit-a-record-43-4-billion-in-2018/
https://venturebeat.com/2019/01/22/npd-u-s-game-sales-hit-a-record-43-4-billion-in-2018/
https://assetstore.unity.com/packages/tools/gui/radar-chart-50074
https://assetstore.unity.com/packages/tools/gui/radar-chart-50074
https://assetstore.unity.com/packages/tools/integration/oculus-integration-82022
https://assetstore.unity.com/packages/tools/integration/oculus-integration-82022
https://developer.oculus.com/documentation/unity/unity-gs-overview/
https://developer.oculus.com/documentation/unity/unity-gs-overview/
https://www.oculus.com/?locale=es_ES
https://www.oculus.com/?locale=es_ES


References 77

Prada, R., Machado, I. y Paiva, A. Teatrix: Virtual environment for
story creation. En International Conference on Intelligent Tutoring Sys-
tems, páginas 464–473. Springer, 2000.

Ridenour, C. S. Finding the horizon: Education administration students
paint a landscape of cultural diversity in schools. Journal of School lea-
dership, vol. 14(1), páginas 4–31, 2004. ISSN 1052-6846.

Robb, D. U.S. Film Industry Topped $43 Billion In Revenue Last Year,
Study Finds, But It’s Not All Good News. DEADLINE , 2018. https:
//bit.ly/2YsR3od.

Romero-Hernandez, A., Gonzalez-Riojo, M., El-Yamri, M. y Ma-
nero, B. Dancing a treasure: A videogame to motivate young au-
diences toward Spanish dance. En Proceedings of the 27 th Internatio-
nal Conference on Computers in Education. Taiwan: Asia-Pacific So-
ciety for Computers in Education (editado por A.-P. S. f. C. in Edu-
cation), páginas 567–572. Asia-Pacific Society for Computers in Educa-
tion, Kenting, Taiwan, 2019. http://ilt.nutn.edu.tw/icce2019/dw/
ICCE2019ProceedingsVolumeI.pdf.

Rothbaum, B. O., Hodges, L., Alarcon, R., Ready, D., Shahar, F.,
Graap, K., Pair, J., Hebert, P., Gotz, D. y Wills, B. Virtual reality
exposure therapy for PTSD Vietnam veterans: A case study. Journal
of Traumatic Stress: Official Publication of The International Society for
Traumatic Stress Studies, vol. 12(2), páginas 263–271, 1999. ISSN 0894-
9867.

Rothbaum, B. O., Hodges, L., Watson, B. A., Kessler, G. D. y
Opdyke, D. Virtual reality exposure therapy in the treatment of fear
of flying: A case report. Behaviour Research and Therapy , vol. 34(5-6),
páginas 477–481, 1996. ISSN 0005-7967.

Rubio-Tamayo, J. L., Gertrudix Barrio, M. y García García, F.
Immersive environments and virtual reality: Systematic review and advan-
ces in communication, interaction and simulation. Multimodal Technolo-
gies and Interaction, vol. 1(4), página 21, 2017.

S Yoon, J. Teacher characteristics as predictors of teacher-student rela-
tionships: Stress, negative affect, and self-efficacy. Social Behavior and
Personality: an international journal , vol. 30(5), páginas 485–493, 2002.
ISSN 0301-2212.

Jurado-de-los Santos, P. y Tejada-Fernández, J. Disrupción y fra-
caso escolar. Un estudio en el contexto de la Educación Secundaria Obli-
gatoria en Cataluña. 2019. ISSN 1578-7001.

https://bit.ly/2YsR3od
https://bit.ly/2YsR3od
http://ilt.nutn.edu.tw/icce2019/dw/ICCE2019 Proceedings Volume I.pdf
http://ilt.nutn.edu.tw/icce2019/dw/ICCE2019 Proceedings Volume I.pdf


References 78

Segovia, J. D., Ortega, J. L. G., Aróstegui, I. G. y Fuentes, A. R.
Competencias comunicativas de maestros en formación. Profesorado. Re-
vista de Currículum y Formación de Profesorado, vol. 14(2), páginas 303–
323, 2010. ISSN 1138-414X.

SoundBible. SoundBible Web Page. 2020. http://soundbible.com/.

SoundSnap. SoundSnap Web Page. 2020. https://www.soundsnap.com/.

Susi, T., Johannesson, M. y Backlund, P. Serious games: An overview.
2007.

Thomas, S. P. Competitive versus collaborative: exploring the negotiation
strategy impact on relational outcomes in ongoing buyer-supplier rela-
tionships. 2013.

UAB. Universitat Autònoma de Barcelona. 2020. https://www.uab.cat/
es/.

Ubisoft. DISCOVERY TOUR MODE OF ASSASSIN’S CREED: ORI-
GINS. 2018. https://support.ubisoft.com/en-gb/Faqs/000031846/
Discovery-Tour-Mode-of-Assassin-s-Creed-Origins-ACO.

Unity. Unity Web Page. 2020. https://unity.com/.

Unity Asset Store. Unity Asset Store. 2020. https://assetstore.
unity.com/.

Unity Blogs. HDRP Guide. 2020. https://bit.ly/31bvjit.

Unity Documentation. Animator Controller in Unity. 2020a. https:
//docs.unity3d.com/Manual/class-AnimatorController.html.

Unity Documentation. Animator Override Controllers in
Unity. 2020b. https://docs.unity3d.com/es/530/Manual/
AnimatorOverrideController.html.

Unity Documentation. Building a NavMesh in Unity. 2020c. https:
//docs.unity3d.com/Manual/nav-BuildingNavMesh.html.

Unity Documentation. Colliders in Unity. 2020d. https://docs.
unity3d.com/es/2019.4/Manual/CollidersOverview.html.

Unity Documentation. Keyword Recognizer in Unity. 2020e.
https://docs.unity3d.com/ScriptReference/Windows.Speech.
KeywordRecognizer.html.

Universidad Complutense de Madrid. Competencias del Máster de
Formación de Profesorado de la UCM. 2020. https://www.ucm.es/
masterformacionprofesorado/competencias.

http://soundbible.com/
https://www.soundsnap.com/
https://www.uab.cat/es/
https://www.uab.cat/es/
https://support.ubisoft.com/en-gb/Faqs/000031846/Discovery-Tour-Mode-of-Assassin-s-Creed-Origins-ACO
https://support.ubisoft.com/en-gb/Faqs/000031846/Discovery-Tour-Mode-of-Assassin-s-Creed-Origins-ACO
https://unity.com/
https://assetstore.unity.com/
https://assetstore.unity.com/
https://bit.ly/31bvjit
https://docs.unity3d.com/Manual/class-AnimatorController.html
https://docs.unity3d.com/Manual/class-AnimatorController.html
https://docs.unity3d.com/es/530/Manual/AnimatorOverrideController.html
https://docs.unity3d.com/es/530/Manual/AnimatorOverrideController.html
https://docs.unity3d.com/Manual/nav-BuildingNavMesh.html
https://docs.unity3d.com/Manual/nav-BuildingNavMesh.html
https://docs.unity3d.com/es/2019.4/Manual/CollidersOverview.html
https://docs.unity3d.com/es/2019.4/Manual/CollidersOverview.html
https://docs.unity3d.com/ScriptReference/Windows.Speech.KeywordRecognizer.html
https://docs.unity3d.com/ScriptReference/Windows.Speech.KeywordRecognizer.html
https://www.ucm.es/masterformacionprofesorado/competencias
https://www.ucm.es/masterformacionprofesorado/competencias


References 79

Universitat de Barcelona. Objectives and competences in master’s
degree in Teacher Training in Obligatory Secondary and Upper Secondary
School Education, Vocational Training and Languages. 2020. https://
www.ub.edu/portal/web/education/university-master-s-degrees/
-/ensenyament/detallEnsenyament/1068147/1.

University of Toronto. Toronto emotional speech set (TESS). 2010.
https://tspace.library.utoronto.ca/handle/1807/24487.

Vallejo, A. P. La gestión de conflictos en el aula. Factores determinantes y
propuestas de intervención. Pérez, M., & Rivera, A.(2005). La orientación
escolar en centros educativos, vol. 319, 2005.

VALVE. Valve Base Stations. 2020. https://www.valvesoftware.com/
es/index/base-stations.

Video Presentation. Presentation Video for EDULEARN20. 2020.
https://www.youtube.com/watch?v=xd_T4_yvWfY.

VIVE. HTC VIVE Controller. 2020. https://www.vive.com/us/
accessory/controller/.

Vokaturi. Vokaturi emotion recognition software. 2020. https://
vokaturi.com/.

Webb, K. The $120 billion gaming industry is going through mo-
re change than it ever has before, and everyone is trying to cash
in. Business Insider , 2019. https://www.businessinsider.com/
video-game-industry-120-billion-future-innovation-2019-9?IR=
T.

Wikiwand. Adobe Fuse Image. 2020. https://www.wikiwand.com/en/
Adobe_Fuse_CC.

YouTube Video. Classroom VR - First scenario. 2020a. https://www.
youtube.com/watch?v=-tz3eJo5_A8.

YouTube Video. Classroom VR - Second scenario. 2020b. https://www.
youtube.com/watch?v=Qvl3f4wH_2g.

YouTube Video. Classroom VR - Third scenario. 2020c. https://www.
youtube.com/watch?v=QEqhbF7CQ-g.

https://www.ub.edu/portal/web/education/university-master-s-degrees/-/ensenyament/detallEnsenyament/1068147/1
https://www.ub.edu/portal/web/education/university-master-s-degrees/-/ensenyament/detallEnsenyament/1068147/1
https://www.ub.edu/portal/web/education/university-master-s-degrees/-/ensenyament/detallEnsenyament/1068147/1
https://tspace.library.utoronto.ca/handle/1807/24487
https://www.valvesoftware.com/es/index/base-stations
https://www.valvesoftware.com/es/index/base-stations
https://www.youtube.com/watch?v=xd_T4_yvWfY
https://www.vive.com/us/accessory/controller/
https://www.vive.com/us/accessory/controller/
https://vokaturi.com/
https://vokaturi.com/
https://www.businessinsider.com/video-game-industry-120-billion-future-innovation-2019-9?IR=T
https://www.businessinsider.com/video-game-industry-120-billion-future-innovation-2019-9?IR=T
https://www.businessinsider.com/video-game-industry-120-billion-future-innovation-2019-9?IR=T
https://www.wikiwand.com/en/Adobe_Fuse_CC
https://www.wikiwand.com/en/Adobe_Fuse_CC
https://www.youtube.com/watch?v=-tz3eJo5_A8
https://www.youtube.com/watch?v=-tz3eJo5_A8
https://www.youtube.com/watch?v=Qvl3f4wH_2g
https://www.youtube.com/watch?v=Qvl3f4wH_2g
https://www.youtube.com/watch?v=QEqhbF7CQ-g
https://www.youtube.com/watch?v=QEqhbF7CQ-g



	Página de TÃtulo
	Ãndices
	Tabla de contenidos
	List of figures
	List of tables

	Introduction
	Objectives

	State of the art
	The problem of current education
	Teacher training and conflict management techniques
	The Four Pillars of Learning

	Video games
	What is a video game?

	Virtual reality
	Virtual environments applied to education
	Serious games or educational video games
	Serious games applications
	a. Edutainment
	b. Educational games
	c. Games for health
	d. Companies, institutions and business corporations

	Serious games at Universidad Complutense de Madrid
	a. The ComunicArte (PR2005-174/01) project
	b. The Didascalia (RTI2018-096401-A-I00) project



	Video game creation tools
	Unity
	Unreal Engine


	Development and implementation
	Design of scenarios
	Selection criteria
	Obtaining the reports
	Proposed scenarios
	First scenario: Inappropriate comment from a student
	Second scenario: Two students refuse to sit apart
	Third scenario: A student is not welcomed by his or her classmates

	Adaptation of the scenarios to Unity
	First scenario
	Second scenario
	Third scenario


	Implementation
	Student models and animations
	First draft of the class
	Asset pack selection process
	Initial asset pack
	Final asset pack

	Implementation of the menu
	Tools
	Controllers
	Colliders
	NavMesh
	SoundLoudness
	KeywordRecognizer
	Vokaturi
	a. Efficiency analisis
	b. Unity adaptation

	Virtual reality adaptation

	Audios

	Explanation of the scenarios
	Common structure
	Context screen
	TipsPanel
	Feedback screen

	First scenario: Inappropriate comment from a student
	Context and problem
	Paths
	Scenario script

	Second scenario: Two students refuse to sit apart
	Context and problem
	Paths
	Scenario script

	Third scenario: A student is not welcomed by his classmates
	Context and problem
	Paths
	Scenario script



	Individual contribution
	Alejandro Díaz
	Álvaro López
	Mario Bocos
	Collaborative work

	Conclusions and future work
	Conclusions
	Accomplished objectives
	Main conclusions

	Future work

	Appendices
	Acceptance Letter
	EDULEARN20 Paper

	References
	Fin