Motivation for Learning
Adaptive Gamification for Web-based Learning Environments
Baptiste Monterrat
1
, Élise Lavoué
2
and Sébastien George
3
1
Université de Lyon, CNRS, INSA-Lyon, LIRIS, UMR5205, F-69621, Lyon, France
2
Magellan, IAE Lyon, Université Jean Moulin Lyon 3, Lyon, France
3
LUNAM Université, Université du Maine, EA 4023, LIUM, 72085 Le Mans, France
Keywords: Gamification, Adaptation, Learning Environment, Motivation, User Model.
Abstract: Many learning environments are deserted by the learners, even if they are effective in supporting learning.
Gamification is becoming a popular way to motivate users and enhance their participation on web-based
activities, by adding game elements to the learning environment. But it still pays little attention to the
individual differences among learners’ preferences as players. This paper presents a generic and adaptive
gamification system that can be plugged on various learning environments. This system can be
automatically personalised, based on an analysis of the interaction traces. We first present the architecture of
the proposed system to support the genericity of the game elements. Then, we describe the user model
supporting the adaptivity of the game elements.
1 INTRODUCTION TO
GAMIFICATION IN LEARNING
Many learning environments have been shown to be
effective when used, but are quickly deserted by
most of the learners because of a lack of motivation.
Gamification is becoming a popular way to motivate
user participation on web based activities. “When
done well, gamification helps align our interests
with the intrinsic motivation of our players”
(Zichermann et al., 2011). Although this concept is
not new (Deterding et al., 2011), yet little research
treats its uses in learning contexts. This paper
proposes a generic and personalised gamification
system to raise motivation in learning environments
that are not intrinsically motivating.
1.1 The Need for Genericity
Turning a learning environment into a game is a
complex process. Currently, if the designers of a
learning environment are interested in gamifying it,
they have to re-design and re-implement the game
elements especially for this environment. This could
be very complex and require a lot of time, whereas
the elements will not be reusable. The existence of
generic game elements would address this problem.
Achievements systems like Mozilla OpenBadges
(Mozilla, 2011) address this need for genericity, but
using badges only is a poor way to gamify.
Maciuszek et al. (2010) used a component-based
framework for implementing educational games. It
allows turning a learning software into a learning
game by changing only the user interface
components. However, the initial software needs to
be already compatible with this framework in order
to be turned into a game, which is not the case with
most existing learning environments.
Thus, we aim to develop gamification as an
independent layer that could be plugged on learning
environments without changing them.
1.2 The Need for Adaptivity
When trying to motivate with games, an important
difficulty comes from the fact that people do not
have the same expectations, and do not have the
same emotional responses to game mechanics (Yee,
2006). A common approach to fulfil these
expectations is to add gamification features for all
the player types within the application, but there is a
high risk of overloading the user interface. That is
why gamification needs to be personalised. Various
researches contributed to the field of adaptive
games, by adapting the user interface, the level of
difficulty (Andrade et al., 2006), the pedagogical
117
Monterrat B., Lavoué É. and George S..
Motivation for Learning - Adaptive Gamification for Web-based Learning Environments.
DOI: 10.5220/0004848101170125
In Proceedings of the 6th International Conference on Computer Supported Education (CSEDU-2014), pages 117-125
ISBN: 978-989-758-020-8
Copyright
c
2014 SCITEPRESS (Science and Technology Publications, Lda.)
scenario (Marne et al., 2013), or the feedback
(Conati et al., 2009). We aim at developing an
adaptive motivational system addressing the three
deficiencies highlighted below.
The first lack identified in existing works is
about games genre and dynamics. Game dynamics
are defined by Zicherman et al. (2011) as “the
player’s interactions with the game mechanics”.
Related works in adaptive games share the goal of
increasing the game’s acceptance and usability, but
the game dynamics remains the same.
The second lack concerns the adaptation of
gamification. While many works focus on the
adaptation of games, few are interested in the
adaptation of gamification. Ferro et al. (2013) are
among the first researchers to conduct works on
personalised gamification. They are trying to relate
directly game mechanics and game elements to both
player types and personality types.
The third lack concerns research on adaptation of
multi-player games. It has been shown in the game
adaptation techniques review of Hocine et al.
(2011). As gamification mainly relies on competitive
and social features, it is important to consider ways
to apply it for groups of users.
1.3 Main Research Questions
Our research works aim at developing a motivating
system, adaptive and adaptable to various web-based
learning activities.
The main research questions related to this goal
are: (1) How to characterise the game elements to
make them generic and pluggable to the learning
environment? (2) Which user model can handle the
adaptivity of the game elements? (3) Which
architecture can support the tracking, the adaption,
and the integration of the game elements?
Section 2 is dedicated to the state of the art
related to gamification and game elements. Then in
section 3 we present the overall architecture of the
system to make gamification generic. In section 4
we provide details about our user model to make
gamification adaptive. We finally conclude about the
contribution of the paper and present future works in
sections 5 and 6.
2 STATE OF THE ART
2.1 Serious Game or Gamification
Games and fun have proven to enhance motivation
in learning activities. But various approaches are
used to add fun in different cases. The most popular
ones are learning games and gamification. Learning
games refer to the use of digital games for learning
purposes (Prensky, 2001). Gamification has been
defined more recently as “the use of game design
elements in non-gaming contexts” (Deterding et al.,
2011). These two approaches are often poorly
distinguished one from the other. However, they
differ by their design process and by the resulting
application (see Table 1).
Table 1: Differences between a learning game and a
gamified application.
Learning game
Gamified
application
Design
process
Designed as a
game from the
beginning
Adding game
mechanics to an
existing application
Resulting
application
A game with
educational
elements
A learning
application enriched
by game mechanics
In this work, we focus on gamification. On the
one hand, it can be based on existing learning
environments. On the other hand, with gamification
the game elements are not central but peripheral,
which fosters their adaptivity. Thus gamification can
become a “fun layer” that could be plugged on
several applications (Monterrat et al., 2013).
2.2 User Model and Adaptation
2.2.1 Distinguishing Learner and Player
In the game-based learning field, user adaptation can
focus on the user as a learner, or as a player, because
each user is both of them. Research on learner model
focus on the relation between the learner and the
knowledge. For example, the theory of adaptive
hypermedia (Brusilovsky, 2001) tends to adapt the
content of the user’s learning activity.
In our work, the role of user modelling is to
adapt the game elements of the gamification layer.
Accordingly, we assume that the learner part of the
user model is handled by the existing learning
application core that manages the learning activity,
while the gamification system focuses on the player
part. That is why we are particularly interested in
player model in next part.
2.2.2 Player Models
Many studies have been conducted about why
people play games. For example, Bartle (1996)
identifies four player types: killer, achiever,
CSEDU2014-6thInternationalConferenceonComputerSupportedEducation
118
socialiser, and explorer. Yee (2006) identifies three
main motivation components: achievement, social
and immersion. Lazzaro (2004) observes four
motivational factors for playing games: hard fun,
easy fun, altered state and people factor. Moreover,
with the growing interest for gamification since a
few years, various companies and game designers
propose their own types of gamers (Kotaku, 2012,
Gamification Co, 2013). In this work, we rely on the
classification of Ferro et al. (2013): dominant,
objectivist, humanist, inquisitive, and creative.
Although their proposal is still a work in progress,
this classification has the advantage to relate the
player types directly to game mechanics and game
elements. This link allows us to personalise and
adapt our system to the players (see section 4.2).
2.2.3 Adaptation Techniques
Many different adaptation techniques can be found
in the state of the art. They are based on various AI
methods, as for instance: reinforcement learning to
build intelligent adaptive agents (Andrade et al.,
2006), Case-Based Reasoning (CBR), Bayesian
network to build a student model (Conati et al.,
2009), and evolutionary algorithm to design the
tracks of a car racing game (Togelius et al, 2007).
This kind of algorithm could be useful to build more
accurate user models. In this work, we chose to use
adaptation rules written by humans in the first place.
2.3 Data for Game Adaptation
According to Kobsa (2001), we distinguish three
forms of adaptation: to user data, to usage data and
to environment data. All these parameter are
important for the game elements personalisation.
2.3.1 User Data
We should pay attention to basic data about users,
like their age and gender, as it has an influence on
their levels of attention and motivation. Charlier et
al. (2012) focused on the influence of the player’s
age. They argue that older adults need games
without pressing time constraints. There are also
gender differences in motivations for playing games.
For example, Eglesz et al. (2005) found that women
prefer Role Playing Games (RPG) while men prefer
action, adventure simulation and sport games.
2.3.2 Usage Data
Most works presented in the review of Hocine et al.
(2011) base their adaptation mainly on the data from
user’s interactions with the system. It is not a
surprise, as this data is generally available without
asking questions to the user. These interactions are
the basic information used to fill the user model,
which may contain the users’ emotional state (Poel
et al., 2004), their way of learning (Bernardini et al.,
2010), their level of success (Andrade et al., 2006),
their level of satisfaction, attention, and engagement.
As increasing engagement is our goal in this work, it
is also a variable we need to track.
The methods for measuring engagement can be
based on humans (De Vicente et al., 2002)
(observation or self-report), hardware (e.g. eye
tracking) or software. The last one is the only one
that we can automate in web based applications.
(Bouvier et al., 2013) defines a typology of engaged
behaviours, to determine if a player is engaged or
not, but the interactions tracked are specific to
games. Mattheiss et al., (2010) present a list of
specific actions that can predict engagement or
disengagement in educational computer games. For
example, if the learner asks immediately for help
without even reading the question, s/he probably
does not want to spend much effort. Cocea (2006)
also proposed useful examples of behaviors
predicting user disengagement, but her approach is
qualitative. In this work, we rely particularly on the
quantitative and computable method proposed by
Beck (2005), called engagement tracing.
2.3.3 Environment Data
Kobsa distinguishes the software environment (e.g.
the browser), the hardware environment (e.g. the
device), and the information about the place (e.g
location and objects in the immediate environment).
It is generally harder to get information from the
third category, but recent technologies like mobile
devices localization can help.
In this work, we are interested in knowing the
human context, because people do not play the same
way if they are alone, with friends, or colleges. For
example, (Cheng et al., 2011) tried to find the good
moments to play at work, while some works focus
on the uses of games in the classroom (Sanchez,
2011). It is also important for us to know the device
used and the learning context, as some ways of
gamifying can be relevant only in some cases.
3 ARCHITECTURE FOR
GENERIC GAMIFICATION
In this part, we explain how we design game
MotivationforLearning-AdaptiveGamificationforWeb-basedLearningEnvironments
119
elements to be generic, adaptable, and pluggable on
already existing learning environments.
3.1 Game Elements as Epiphytes
In order to personalise the fun features, the learning
application needs to be able to work with or without
these features. That is why we propose to use
epiphytic functionalities: applications that are
plugged in another application without being
necessary. Giroux et al. (1995) define epiphytic
systems as follows: (1) the epiphytic system cannot
exist without a host, (2) the host can exist without
the epiphyte, (3) the host and the epiphyte have
independent existences, and (4) the epiphyte does
not affect its host.
By implementing the fun functionalities like
epiphytes, we can enable or disable them
independently for each user, in order to adapt his/her
interface without affecting the learning application.
This is also a way to foster genericity. We provide
below examples of such functionalities that can be
activated:
A leader board of fast learners for competitive
users.
Badges and cups for challenge.
Ability to leave tips to other users.
Ability to share scores and success on social
networks.
A chat feature for users interested in
socializing.
As shown on Figure 1, the epiphytes (E1 and E2)
are distributed in the user interface, but controlled
only by the gamification layer, independently from
the control of the pedagogical activity.
Figure 1: Independence of pedagogical control and game
control.
3.2 Architecture
An overview of the proposed architecture is
presented in Figure 2, which shows the way the
gamification system can be plugged in an existing
learning environment.
The interactions between the user and the
environment are permanently traced (1) and stored
in the database. Secondly, the data collected is used
by the trace analysis system (2), which calculates
frequently the engagement level of the user and
stores it in the same base. When the trace analysis
system detects user disengagement, it sends an alert
to the adaptation engine before the user leaves.
When the adaptation engine (3) receives an alert
about the low engagement level, it updates the
information of the player model in the same base,
according to the history of engagement level and the
use of activated epiphytes (see section 4.3.2), and
selects the epiphytic functionality which best fits the
user’s needs. Finally, the selected functionality (4) is
Figure 2: Architecture of the gamification system.
CSEDU2014-6thInternationalConferenceonComputerSupportedEducation
120
introduced in the learning environment (see section
3.3).
3.3 Integration of the Epiphytic
Functionalities
There are different possible ways of introducing and
integrating the functionalities.
On the one hand, the user needs to be aware of
the introduction of a new functionality, so we have
to inform him/her. On the other hand, the
information must not interrupt the learning activity
(“the epiphyte does not affect its host”), so a popup
window is also not a good solution. As shown on
Figure 3, we propose a small tooltip to inform the
user without requiring any interaction.
Figure 3: Tooltip to inform the user of changes.
The web technologies allow us to integrate the
epiphytes in various ways on the web pages, like
panels for the information displayed permanently,
and tooltips for epiphytes based on punctual
feedback. Examples are shown on Figure 4.
Figure 4: Examples of ways to integrate the epiphytic
functionalities in a web user interface.
Finally, it is important to allow the users to
disable the activated functionalities. The first reason
is that some people do not want to play, and they
should not be forced to, as games are a voluntary
activity. The second reason is that the adaptation
engine may be wrong during the first uses of the
environment, and may propose a functionality that
does not feet the player’s preferences. Thus, the
player can close the functionality. By the way this
provides a useful feedback to the system about what
the user does not like.
4 MODELS FOR ADAPTATION
OF GAMIFICATION
In section 3.2, we presented the architecture of the
system that supports a generic gamification. In this
part, we focus on the adaptation process and the
player model necessary for this adaptation.
4.1 User Model
An overview of the user model is shown on Figure
5, and its parts are details in the following
subsections.
Figure 5: Overview of the collected and calculated data in
the user model.
The data we want to know is registered within the
player model (section 4.2), which tells us which
game elements the user may like. It is calculated
based on the engagement level, and the use of the
epiphytes. The collected data is detailed in section
4.3)
4.2 Player Model for Adaptive
Gamification
In section 2.2 we explained why we chose to base
our model on the classification of Ferro et al. (2013).
The list of its motivational factors is presented in
Table 2.
Table 2: Player classification of Ferro.
Classification Examples of game elements
Dominant Characters, conflicts
Objectivist Objectives, challenge
Humanist Story/Narrative, dramatic art
Inquisitive Aesthetics, boundaries
Creative Resources, world building
MotivationforLearning-AdaptiveGamificationforWeb-basedLearningEnvironments
121
When a new user registers on the learning
environment, the values of each motivational factor
are initialised for him/her according to user data (see
section 4.3.1). During the use of the learning
environment, the values will change according to the
user’s interactions (see section 4.3.2).
In addition, each epiphytic functionality also has
a list of values associated with the motivational
factors. Table 3 provides an example of such
association.
Table 3: Example of values associating the epiphyte
“leader board” to the motivational factors.
Leader board
Dominant 100%
Objectivist 40%
Humanist 20%
Inquisitive 0%
Creative 0%
These values are necessary to choose the
adequate functionality when we know the user’s
player profile.
4.3 Data for Gamification Adaptation
The three types of data we use for adaptation (Kobsa
et al., 2001) are based on the state of the art
presented in part 2.3.
4.3.1 User Data
The user data we use for adaptation are
Birth date
Gender
These data are static, but they have an influence
on the initial values of the player model. Adaptation
rules can be extracted from our knowledge on the
influence of these data, and these adaptation rules
can be used to set better values for the player model
of new users (see Table 4 for examples).
Table 4: Examples of adaptation to user data.
Tom is a man. When he
registers on the learning
environment, his value
for the motivational
factor “competition” is
set at 60%, instead of
40% for a woman
Nadia is 62 years old.
When she registers on the
learning environment, we
set a limit of 2 epiphytes
activated at the same time.
4.3.2 Usage Data
We need to track the user’s interactions with both
the gamification layer and the learning environment,
to evaluate the level of engagement.
Concerning the tracking of the epiphytic
functionalities, we can assume that the more a
functionality is used, the more the player is sensitive
to the motivational factor associated with this
functionality.
Table 5: Examples of adaptation to the use of epiphytes.
Tom is now able to
publish on a social
network when he
finished a learning
session. He uses this
functionality many times.
Accordingly, his value
for the motivational
factor “social” increases.
The button for sharing
activity on social networks
has been introduced in
Nadia’s interface. She
turned it off after one
minute. Accordingly, her
value for the motivational
factor “social” decreases.
As some functionalities do not require direct
interactions, the system has to find a correlation
between the activation of the functionalities and the
engagement of the learner. A functionality is
effective if it is correlated with a high engagement.
Table 6: Examples of adaptation to the engagement level.
The leader board was
added in Tom’s
environment, but no
difference was observed
in his behaviour. His
value for the
motivational factor
“competition” decreases.
Since the leader board was
added in Nadia’s
environment, she is
connected more often and
makes more exercises to
raise her score. Her value
for the motivational factor
“competition” increases.
Our way of calculating engagement and
disengagement is detailed in section 4.4.
4.3.3 Environment Data
In addition, some contextual information are crucial
for the gamification engine. Firstly, it is useful to
know if the learner is at school, at work, or on free
time, as this context has an influence about how
people learn.
We are also interested in the device used by the
player. In the cloud computing domain, various
learning environments are available on mobile
devices as on computers, but all features are not
necessary relevant or available on any device (e.g.
because of the screen size).
Summary of the environment data for
adaptation:
Device used.
Learning context (school, work or personal).
Size of the group (if school or work).
CSEDU2014-6thInternationalConferenceonComputerSupportedEducation
122
Table 7: Examples of adaptation to environment data.
Tom learns at school in
the computer room with
his other classmates.
Accordingly, a chat
feature would be useless
because he speaks
directly with them.
Nadia, whose motivational
factor “competition” is
high, sometimes learns on
her smartphone.
Accordingly, we can
propose her to compete
with players locally near
from her.
4.4 Engagement Tracking
We have access to one information directly
indicating engagement:
The session dates.
A user connected more often and longer can be
considered as more engaged in the activity than
another. However, this allows us to know the
general engagement but not to compare the
engagement level at two distinct times. That’s why
we need another way to track real time engagement.
We use two metrics:
Too short time to read texts and to answer
questions, based on engagement tracing (Beck,
2005).
Too long time to read or answer questions.
4.5 Adaptation Technique
The gamification engine has two roles: updating the
player model and selecting a functionality adapted
for the user.
To update the player model, a simple algorithm
based on adaptation rules increases or decreases the
values of the motivational factors, according to the
observed use of the epiphytes and the engagement
level (see section 4.3.2). Then, the engine has to
select the motivational factor with the highest value,
and to identify an epiphyte corresponding to this
motivational factor, according to the association
table (see Table 3). This behaviour must be balanced
with some random selections. Selecting the
functionality totally randomly would be ignoring the
user model. But if there is no random, the
functionalities implementing new motivational
factors for this user will never be tried.
Finally, as the epiphytes may induce interactions
between users, this engine has to “take in account
the collaborative aspect and heterogeneity between
players, while maintaining the overall coherence of
the game” (Hocine et al., 2011). That is why the
adaptation engine checks if several users of the
group are interested in competing before activating
multi-player functionalities.
5 CONCLUSION AND
DISCUSSION
In this paper, we proposed the architecture of a
system to motivate learners by integrating game
elements in existing web-based learning
environments. This system is both generic and
adaptive.
The genericity is based on the use of game
elements as epiphytic functionalities, which does not
affect the host environment when integrated in the
user interface.
The adaptivity is based on a player model that
defines the player type matching best with the user.
The adaptation process has four steps:
1. Tracing data from the learning environment
and the game elements.
2. Evaluating the engagement level of the user.
3. Updating the player model, based on
adaptation rules, using basic data about the
user, data from the use of the environment, and
data describing the learning context.
4. Integrating within the user interface the
epiphyte matching best with the player model.
This system is not designed with the goal to turn
every learning activity into a game, because games
need to be played voluntary and people in some
contexts are already motivated to learn. Adaptive
gamification should be used with non-intrinsically
motivating activities, like memorizing vocabulary or
mathematical rules.
Despite this system has not been tested yet, it
addresses three lacks in the literature and existing
software:
It proposes the adaptation of game dynamics,
whereas existing systems (e.g Khan Academy,
2006) adapt the learning path and difficulty
level.
It deals with adaptation of gamification,
whereas the literature deals more with
adaptation of games.
It proposes the adaptation of multiplayer
features, whereas existing environments
propose the same game elements for all the
users.
6 FUTURE WORK
We plan various evaluations and improvements for
the system.
Regarding the evaluations, we are currently
implementing the system, which will be plugged on
MotivationforLearning-AdaptiveGamificationforWeb-basedLearningEnvironments
123
“Projet Voltaire”, a web-based environment to learn
French spelling. For the next step, we will plug the
gamification system on other learning environments,
in order to evaluate its genericity.
An experiment will allow us to evaluate the
system described in this paper. For instance, we plan
to compare the automatic adaptation with the
“home made” adaptation: what happens if the user
can choose the new functionalities by himself? In
order to evaluate the gamification adaptation to
users’ profiles, we plan to compare three cases:
Case 1: Selecting game elements according to
user’s profile.
Case 2: Selecting game elements randomly.
Case 3: Selecting game elements at the
opposite of user’s profile.
In this way, we will be able to evaluate the
relevance of the proposed adaptation (case 1).
Furthermore, some improvements will concern
the flexibility of the player model. Sometimes, the
player type is not enough to model the user’s needs,
as they can change during the day. As an example,
two motivational factors of Lazzaro (2004) are
detailed bellow:
Hard fun (Players look for challenge, strategy
and problem solving).
Easy fun (Players enjoy intrigue and
curiosity).
Whether we expect to relax (easy fun) or to be
challenged (hard fun) depends more on our mood
than our personality and player type. Some
contextual information can help to know about this
mood, like the hour and the day. For example, a user
may expect a more relaxing activity after lunch.
Furthermore, expert systems are limited as they are
static. Another improvement we plan to do is the use
of machine learning techniques to automatically
adapt the adaptation rules themselves, based on the
experience with the previous users.
REFERENCES
Andrade G., Ramalho G., Gomes A. S., Corruble V.,
2006. Dynamic Game Balancing: an Evaluation of
User Satisfaction, in: Proceedings of the 2nd Artificial
Intelligence and Interactive Digital Entertainment
Conference (AIIDE06), AAAI Press.
Bartle, R., 1996. Hearts, Clubs, Diamonds, Spades: Players
Who Suit MUDs. Journal of MUD Research 1, 1.
Beck, J., 2005. Engagement tracing: using response times
to model student disengagement. Artificial intelligence
in education: Supporting learning through intelligent
and socially informed technology 125, 88.
Bernardini, A., Conati, C., 2010. Discovering and
recognizing student interaction patterns in exploratory
learning environments, in: Intelligent Tutoring
Systems. pp. 125–134.
Bouvier, P., Lavoué, E., Sehaba, K., George, S., 2013.
Identifying Learner’s Engagement in Learning Games:
a Qualitative Approach based on Learner’s Traces of
Interaction, in: 5th International Conference on
Computer Supported Education (CSEDU 2013),
Aachen, Germany. pp. 6–8.Brusilovsky 2001.
Brusilovsky, P., 2001. Adaptive hypermedia. User
modeling and user-adapted interaction 11 (1-2): 87-
110.
Charlier, N., Michela, O., Remmele, B., Whitton, N.,
2012. Not Just for Children: Game-Based Learning for
Older Adults, in: 6th European Conference on Games
Based Learning, Cork, Ireland. pp. 102–108.
Cheng, T. I. C., n.d. Cheng, L.-T., Shami, S., Casey, D.,
Muller, M., DiMicco, J., Patterson, J., others, 2011.
Finding moments of play at work, in: CHI 2011
Workshop (2–5). Vancouver: Canada.
Cocea, M., 2006. Assessment of motivation in online
learning environments, in: Adaptive Hypermedia and
Adaptive Web-Based Systems. pp. 414–418.
Conati, C., Manske, M., 2009. Evaluating adaptive
feedback in an educational computer game, in:
Intelligent Virtual Agents. pp. 146–158.
Deterding, S., Dixon, D., Khaled, R., Nacke, L., 2011.
From game design elements to gamefulness: defining
gamification, in: Proceedings of the 15th International
Academic MindTrek Conference: Envisioning Future
Media Environments. pp. 9–15.
Eglesz, D., Fekete, I., Kiss, O. E., Izso, L., 2005.
Computer games are fun? On professional games and
players’ motivations. Educational Media International
42, 117–124 (in Hainey, T., Connolly, T., Stansfield,
M., Boyle, E., 2011. The differences in motivations of
online game players and offline game players: A
combined analysis of three studies at higher education
level. Computers & Education 57, 2197–2211.).
Ferro, L. S., Walz, S. P., Greuter, S., 2013. Towards
personalised, gamified systems: an investigation into
game design, personality and player typologies, in:
Proceedings of The 9th Australasian Conference on
Interactive Entertainment: Matters of Life and Death.
p. 7.
Gamification Co, 2013, A New Perspective on the Bartle
Player Types for Gamification - GCo [WWW
Document], n.d. URL http://www.gamification.co/
2013/08/12/a-new-perspective-on-the-bartle-player-
types-for-gamification/ (accessed 10.1.13).
Giroux, S., Pachet, F., Paquette, G., Girard, J., 1995. Des
systèmes conseillers épiphytes. Revue d’intelligence
artificielle 9, 165–190.
Hocine, N., Gouaïche, A., Di Loreto, I., Abrouk, L., 2011.
Techniques d´adaptation dans les jeux ludiques et
sérieux. Revue d’intelligence artificielle 25, 253–280.
KhanAcademy, 2006. www.khanacademy.org.
Kobsa, A., Koenemann, J., Pohl, W., 2001. Personalised
hypermedia presentation techniques for improving
CSEDU2014-6thInternationalConferenceonComputerSupportedEducation
124
online customer relationships. The Knowledge
Engineering Review 16, 111.
Kotaku, 2012, They’ve Narrowed Gamers Down to Eight
Fundamental Types. Which Are You? [WWW
Document], n.d. URL http://kotaku.com/5938464/
theyve-narrowed-gamers-down-to-eight-fundamental-
types-which-are-you (accessed 10.1.13).
Lazzaro, N., 2004. Why we play games : Four Keys to
More Emotion Without Story. Technical Report, XEO
Desig inc.
Maciuszek, D., Ruddeck, G., Martens, A., 2010.
Component-based development of educational games:
The case of the user interface, in: Proceedings of the
4th European Conference on Games-based Learning
(ECGBL 2010), Reading, UK. pp. 208–217.
Marne, B., Carron, T., Labat, J.-M., Marfisi-Schottman, I.,
2013. MoPPLiq: A Model For Pedagogical Adaptation
of Serious Game Scenarios, in: Advanced Learning
Technologies (ICALT), 2013 IEEE 13th International
Conference On. pp. 291–293.
Mattheiss, E., Kickmeier-Rust, M.D., Steiner, C.M.,
Albert, D., 2010. Approaches to detect discouraged
learners: Assessment of motivation in educational
computer games. Proceedings of eLearning Baltics
(eLBa) 10.
Monterrat, B., Lavoué, E., George, S., 2013. Toward
Personalised Gamification for Learning Environments.
In 4th Workshop on Motivational and Affective
Aspects in Technology Enhanced Learning (MATEL
2013) in Conjonction with EC-TEL 2013.
Mozilla, 2011. http://openbadges.org.
Poel, M., Heylen, D., Nijholt, A., Trappl, R., 2004.
Emotion based Agent Architectures for Tutoring
Systems: The INES Architecture, in: Cybernetics and
Systems 2004. Workshop on affective computational
entities.
Prensky, M., 2001. Digital game-based learning. McGraw-
Hill, New York.
Sanchez, E. 2011. A game in the classroom, what did the
students learn? ESERA 2011, Lyon, France.
Togelius, J., De Nardi, R., Lucas, S.M., 2007. Towards
automatic personalised content creation for racing
games, in: Computational Intelligence and Games,
2007. CIG 2007. IEEE Symposium On. pp. 252–259.
De Vicente, A., Pain, H., 2002. Informing the detection of
the students’ motivational state: an empirical study, in:
Intelligent Tutoring Systems. pp. 933–943.
Yee, N., 2006. Motivations for play in online games.
CyberPsychology & Behavior 9, 772–775.
Zichermann, G., Cunningham, C., 2011. Gamification by
Design: Implementing game mechanics in web and
mobile apps. O’Reilly Media, Inc.
MotivationforLearning-AdaptiveGamificationforWeb-basedLearningEnvironments
125
