Review of Learner Modeling Using Educational Games
Mohamed Ali Khenissi
1
, Fathi Essalmi
1
, Mohamed Jemni
1
and Kinshuk
2
1
Research Laboratory of Technologies of Information and Communication & Electrical Engineering (LaTICE),
Tunis National Higher School of Engineering (ENSIT), University of Tunis,
5, Avenue Taha Hussein, B.P. 56, 1008, Tunis, Tunisia
2
School of Computing & Information Systems, Athabasca University, Athabasca, Canada
Keywords: Educational Games, Learner Model.
Abstract: The learner model is a key component of any adaptive E-Learning system, including educational games, as
it stores the information about the learner, which can then be used to provide personalized experience.
Recently, there has been growing interest in creating learner models using educational games. This is due to
the limitations of traditional E-Learning systems, since the learner interaction with the computer is rather
restricted in such systems. Educational games, on the other hand, not only stimulate learners by increasing
their motivation and engagement, but also facilitate a lot of learner interaction that could be observed to
create learner models. This paper presents a survey of the field of learner modelling using educational
games. In particular, it describes the main methods of learner modelling. This paper also lists several
educational games that are suitable for experimentation. Compiling this information can be important to the
researchers and developers working in this field, especially to new researchers.
1 INTRODUCTION
Recently, there has been growing interest in learner
modelling using all types of games designed for
educational purposes. Learner modelling (the
process of creating learner model) requires several
observations (inputs) collected from learner
interaction with the computer. The learner
interaction with computer is rather restricted in
traditional E-Learning systems (limited to clicks,
time that the learner invested in visiting the page,
etc.) (Pablo et al., 2009; Stathacopoulou et al.,
2007). This information does not really inform us
how the learner interacted with that content. Did
they read the content? Did they examine it? Did they
simply go outside? (Pablo et al., 2009). This lack of
information affects the accuracy of learner
modelling process. On the other hand, educational
games provide ample opportunities of learner
interaction with the computer, which can be used to
create reliable learner models. Therefore, using
educational games (that represent a highly
interactive content) to model the learner is a
promising trend.
This paper presents a survey of the field and
describes the main methods of learner modelling
using educational games. This paper also lists
several educational game platforms that are suitable
for experimentation. This is an attempt to facilitate
experimental research in this field. Compiling this
information can be important to the researchers and
developers working in the field of educational games
and learner modelling, especially to new researchers.
The main content of the paper (sections 2 and 3)
is a review of existing methods that have already
been used to model the learner using educational
games. To make the scope of the review more clear
we define the concepts of modelling methods as set
of ideas and approaches that can be presented at the
conceptual level and used to achieve a specific
objective. This objective is the creation of learner
model.
In this review, we observed those methods that
model learners in explicit way and the ones which
model them in implicit way. The difference between
explicit and implicit methods of learner modelling is
related to the ways of extracting information about
learners. An implicit method aims to extract
information about learners in hidden and
unobtrusive way, without endanger the high level of
engagement provided by educational games. On the
other hand, an explicit method aims to use a direct
and obvious way of extracting information and made
it overt to learners. The next section of the paper
147
Ali Khenissi M., Essalmi F., Jemni M. and Kinshuk ...
Review of Learner Modeling Using Educational Games.
DOI: 10.5220/0004840201470152
In Proceedings of the 6th International Conference on Computer Supported Education (CSEDU-2014), pages 147-152
ISBN: 978-989-758-021-5
Copyright
c
2014 SCITEPRESS (Science and Technology Publications, Lda.)
presents identified methods that have already been
used to model the learner explicitly. Section 3
presents methods that are used to generate learner
models implicitly. Section 4 describes and discusses
several educational games used to model learners.
Lastly, conclusions are drawn and directions for
future work are presented.
2 EXPLICIT METHODS
In this section, we will present identified methods
that were already used to model the learner
explicitly along with examples.
2.1 Evaluation of Questionnaire
Learner modelling can be performed through
evaluating the learner's answers to questionnaire at
the beginning, during or at the end of the educational
game. It can also be done by redirecting learners to
an online survey and asking learner to answer
questions (Fu et al., 2009; Pourabdollahian, 2012).
The uses of questionnaire can provide direct and
precise answers, but this would endanger the high
level motivation provided by educational games in
case of stopping learner from playing and requesting
her/him to answer questions.
As examples of works that used this method, a
scale named EGameFlow was used in (Fu et al.,
2009) to measure learners’ enjoyment of educational
games. The scale contained 42 items, presented in
Likert-type scales, with 1 and 7 respectively
representing the lowest and highest degree to which
respondents agree with the items. Similarly, in
(Pourabdollahian, 2012), learners were requested to
answer a questionnaire in order to evaluate their
level of engagement. The questionnaire contained 21
questions based on five-point Likert scale and
categorized according to five classifications
(challenge, immersion, interest, purpose and
control). The data collected from the survey was
analyzed by both descriptive and inferential statistics
for evaluating the engagement of learners.
2.2 Interpreting Body Posture and
Physiological Signals
Various hardware and software equipments have
also been used for modelling the learners. Behaviour
of the learners can be recognized through identifying
gestures, body posture and physiological signals. It
is done by identifying and labelling different
reactions to game events and then looking for
common features of body expressions (Peters et al.,
2009; Jimenez et al., 2011).
This method can provide additional information
for modelling learner. However, it requires the use
of additional hardware and software equipment.
Furthermore, observations collected using these
equipments can be interpreted in different ways and
this can endanger the reliability of learner models.
(Jimenez et al., 2011) used Electro
Encephalogram readings using a Brain Computer
Interface (BCI) for psychometric input to measure
attention. BCI offered the possibility of reading
electric signals generated by neural activity in the
brain, which could be used to assess the learners’
attention levels. In another example (Peters et al.,
2009), learner's level of attention was evaluated
using two components. The first component detected
user gaze behaviour based on input from a web
camera. This component used either the eye
direction or the head direction to establish the
screen-space coordinates of where the user was
looking. The second component measured attention
levels using a neurophysiological recording device.
3 IMPLICIT AND UNOBTRUSIVE
METHODS
Several methods are used to generate learner models
implicitly. In this section, we are going to identify
and discuss methods that have been used in the
literature.
3.1 Translating Learner's Actions
In this method, the set of actions made by the learner
during the game can be interpreted and translated
into descriptive information useful to model the
learner. In addition, experts or automated systems
define to each set of actions an appropriate
description. These descriptions are then used to
model the learner (Mark, 2012; Conlan et al., 2009).
For example, the learner model in (Mark, 2012)
aimed at creating a representation of learner’s
cognitive traits. To do that, the study established
connections between the behaviour of the learner
and his/her actions in the game. Specifically, the
manifestations of traits that the learner may exhibit
during game-play were identified. This could
provide evidence of the learner’s cognitive abilities.
The learner model’s goal in (Conlan et al., 2009)
was to generate a real-time evaluation of a learner’s
skills. The skill assessment engine component was
responsible for translating each learner’s actions
CSEDU2014-6thInternationalConferenceonComputerSupportedEducation
148
within the game into a list of probabilities that show
the likelihood of each relevant skill having been
acquired by the learner. The learner model
determined which skill states had increased or
decreased in probability.
3.2 Interpreting Interaction Traces
In this method, the game engine monitors and
records interaction between the learner and the
game, generating a trace of actions performed by the
learner during the play sessions and tracking a lot of
different parameters in the game (start game, end
game, quit game, phase changes, significant
variables, time, performance, tasks completed, etc.).
At the end of the game play, these traces can be
delivered to a group of experts or automated systems
in order to identify what information can be
extracted from these traces and infer a lot of useful
information for modelling purposes (Stathacopoulou
et al., 2007; Bouvier et al., 2013a; Bouvier et al.,
2013b). For example, learner’s learning style was
modelled in (Stathacopoulou et al., 2007) by
monitoring learner’s actions over time, where each
response such as a keystroke, mouse move or drag
was timed and recorded. The learning environment
stored all available information about what a learner
was doing in a log file, recording each learner action
with a time stamp. Furthermore, this work used
teachers’ expertise in order to select the appropriate
measures of learners’ observable behaviour to serve
as indicators of learners’ learning style.
Learners’ engagement was identified in (Bouvier
et al., 2013a; Bouvier et al., 2013b) by relying on
their traces of interactions performed in the learning
game. The study proposed an approach that consists
of three stages. The first stage aims to determine the
high-level engaged-behaviours. The second stage
aims to characterize these engaged-behaviours by
identifying the underlying chains of actions. The last
stage aims to detect these chains of actions among
all the actions recorded.
3.3 Interpreting Conversation
In this method, information about the learner is
extracted from his/her communication with a Non
Player Character (NPC). The NPC poses questions
and records learner's choices during these
conversations and questions answered during game
dialogues (Pablo et al., 2007; Genaro et al., 2008).
For example, learner’s learning style and
learner’s preferences were modelled in (Pablo et al.,
2007) by interpreting conversations between the
learner and NPC during the game play. In particular,
the NPC queried the learner on his/her preferences
and other questions. Depending on the learner's
choices during these conversations, the game
detected his/her learning style and also detected
his/her preferences.
A NPC in the (Genaro et al., 2008) monitored
perpetually the learner and requested her/him to
answer questions. The system collected and
processed the learner responses. After that, it
calculated learner's level of motivation. Then it
selected one rule from a set of rules taken from
theory and reacted aiming to sustain or enhance the
current level of motivation.
3.4 Interpreting Learner’s Errors
The goal of this method is to retain the information
of what the learner has learnt, and what the learner
has learnt incorrectly. In addition, this method
records learners’ misconceptions, errors, number and
type of mistake made by the learners, learner failure
and success and the time taken to complete the game
(Champagnat et al., 2010; Khenissi et al., 2013).
For example, learner failure and success in
(Champagnat et al., 2010) are pursued. In the case of
failure, the game has to be permissive with the
learner. This done by punishing the learner (worse
results, handicaps at the time of the following tests,
less points) but not stop him from playing or
reiterate the stage.
Learner model in (Khenissi et al., 2013) detected
learner's deficiency in programming subject. To do
that, the learner model keeps track of the learner’s
wrong answers when he/she arranges a set of
unordered instructions in order to form a program. In
particular, when the learner moves an instruction in
inappropriate place, the learner model records the
incorrect answer and provides him appropriate helps.
3.5 Path Follow
The goal of this method is to follow the path of
learner during the game play, record learner
trajectory towards a goal and then interpret and infer
information about the learner. In fact, each path has
a specific meaning and leads to specific information
about the learner (Moreno-Ger et al., 2008; Genaro
et al., 2009).
For example, (Moreno-Ger et al., 2008)
described the game as state transition systems in
order to assess learner activity inside the game.
During the game play, actions of the learner
triggered state transitions and the sequences of
ReviewofLearnerModelingUsingEducationalGames
149
actions leaded to one or many end states. The game
engine kept track of transitions, checked the states
that the game went through and generated reports
describing them.
A virtual observer within the educational game
in (Genaro et al., 2009) monitored the learner and
identified his/ her learning trajectory. Learning
trajectory identified was compared to different
predefined learning trajectories. This comparison
informed the learner model to take a decision on
how to proceed.
4 EDUCATIONAL GAMES USED
TO MODEL LEARNER
This section describes educational games used to
model the learner. Most of these educational games
include mechanism to extract specific information
about learners. This is an attempt to facilitate
experimental research in this field and help a new
researcher in the field to start his/her own work.
4.1 Prime Climb
Prime Climb is an educational game designed to
help 6th and 7th grade students practice number
factorization. Prime Climb consists of series of
mountains. Each mountain is divided into hexes
labelled with numbers. Two players must collaborate
to climb these mountains. Each player can only
move to a number that does not share any common
factor with the partner’s number. If a wrong number
is chosen, the climber falls and swings from the rope
until the player select a correct number. During the
game, each student has a pedagogical agent which
provides support. This educational game was cited
in many works for different purpose (Conati and
Maclaren, 2009).
4.2 Learning Version of Pacman Game
Learning version of Pacman game (LPG) aims to
motivate learners to correctly answer the questions
of the programming languages. In the traditional
version of Pacman game, players control Pacman
through a maze, eating pac-dots. When all dots are
eaten, Pacman is taken to the next stage. In addition,
there are four enemies who roam the maze, trying to
catch Pacman. If an enemy touches Pacman, a life is
lost. Near the corners of the maze are four dots
known as power pellets that provide Pacman with
the temporary ability to eat the enemies. When all
lives have been lost, the game ends. In the Learning
version of Pacman Game, when Pacman eats a
power star, the learner has to respond to a question
(about the programming language) in order to
continue the game having a 'reverse' role (Pacman
can move freely and eat the enemy for a short
period) (Khenissi et al., 2013c).
4.3 Talking Island
Talking Island is an educational Massively Multiple
Online Role-Playing Game (MMORPG) designed to
teach English vocabulary and conversational skills
to elementary school students. It includes elements
that are generally found in an MMORPG (e.g. team-
work, battles, pets, and role-playing situations). The
learner can practice the pronunciation thanks to
voice-recognition module included in this game. In
fact, this module diagnoses the correctness of the
pronunciation and determines whether the learner
reaches a new level or passes a quest. In addition,
this game allows players to team up, communicate
via voice or texts, and solve quests jointly (Hou,
2012). This educational game is accessible online at
http://www.talking-island.com/
4.4 e-Adventure Educational Game
Engine
E-Adventure is a game engine designed to facilitate
the creation of educational games. Games delivered
by e-Adventure engine are point and click adventure
games. During the game, learner can learn by the
interaction with objects, consultation of in-game
books and conversations with other characters. The
<e-Adventure> engine includes mechanisms that can
monitor the learner’s activities and then provides
adaptation and assessment (Pablo et al., 2007). The
E-Adventure platform is accessible online at http://e-
adventure.e-ucm.es/
4.5 ELEKTRA Game
The ELEKTRA game is a narrative-driven
adventure game. In this game, the learner has to
solve several physics-oriented puzzles. A virtual
character (representing the ghost of Galileo) guides,
advises and encourages the learner during the game
(Conlan et al., 2009). The ELEKTRA project is
accessible online at http://www.elektra-project.org/
4.6 80Days Game
80Days is an adventure game used to teach
CSEDU2014-6thInternationalConferenceonComputerSupportedEducation
150
geography for a target audience of 12 to 14 year
olds and follows European curricula in geography.
The game story is about an alien scout called Feon
which kidnaps a Boy (play character) and travels
with him around the world in a spaceship to collect
relevant geographical information. The player assists
the alien to explore the planet and to create a report
about the Earth and its geographical features. In the
course of the game, the player discloses the aliens’
real intention which is preparing the conquest of the
earth. The player has to save the planet and the only
way to do it is to draw the right conclusion from the
traitorous Earth report (Kopeinik et al., 2012). The
80Days project is accessible online at
http://www.eightydays.eu/
4.7 Instruction Right Place Game
Instruction Right Place Game allows learners to
benefit from the drag and drop technology to
construct a program (from any programming
language) in an amusing way. This educational
game breaks down complex programming tasks and
guides learners through a series of small steps to
form a program interactively (Khenissi et al., 2013a;
Khenissi et al., 2013b; Khenissi et al., 2014).
4.8 Educational Games and Methods of
Learner Modelling
The process of creating learner model requires
several observations (Inputs) collected from the
learner’s interaction with the educational game.
Methods described in this paper will use these
inputs in order to extract specific information about
learner (Outputs). This information will be stored in
the Learner Model. Most of educational games
previously cited include mechanism to extract
specific information about learners and then give
them adaptation. Table 1 summarizes input, methods
and output used by these educational games, and
cited in the literature, in order to model learners.
Table 1 shows that different educational games
have used different methods for creating learner
model. For example, the educational game Prime
Climb has used body expression of the learner, as
input, and interpret body posture and physiological
signals method in order to model learner’s emotions.
Indeed, manifestations of emotion are expressed
immediately through the body. Hence, the best way
to extract these manifestations is to interpret body
posture and physiological signals of the learner
using hardware and software equipment. As another
example, the table shows that learner's deficiency
can be modelled using interpreting learner’s errors
methods. In particular, this method tracks errors of
the learner in order to interpret them.
5 CONCLUSIONS
In this paper, we focused on learner modelling using
educational games. In particular, we gathered
information from various works in the field and
identified the most important methods used.
Furthermore, several educational games used to
model the learner and available for researchers are
presented and discussed
The use of educational games can improve
learner motivation, increase learner’s desire to learn
and develop positive attitudes toward many subjects.
Furthermore, learner modelling through the use of
educational games can create a reliable learner
model that has an extremely important role in
adaptive learning system.
For future work, we will look how to adapt the
game experience to the individual learner using
learner model. In fact, adaptation service in
educational game can improve the learning process.
Table 1: Educational Games and Methods of Learner Modelling.
Educational games Inputs Methods Outputs
1 Prime Climb Body Expression
Interpreting body posture and
physiological signals
Learner’s emotions
2 LPG Questionnaire Evaluation of questionnaire Learner’s level of knowledge
3 Talking Island Learner behaviour Interpreting Interaction traces Learner activity
4 E-Adventure
Conversation of the
learner
Interpreting conversation
Learner’s learning style
5 ELEKTRA game Actions of the learner Translating learner's actions Learner’s skills
6 80Days game Learner behaviour Path follow Learner's competences
7
Instruction Right
Place Game
Errors of the learner Interpreting learner’s errors
Learner's deficiency in
programming subject
ReviewofLearnerModelingUsingEducationalGames
151
Therefore, we will discuss methods and techniques
used in this field.
REFERENCES
Bouvier, P., Lavoué, É., Sehaba, K., George, S., 2013a.
Identifying Learner’s Engagement in Learning Games:
a Qualitative Approach based on Learner’s Traces of
Interaction. Proceedings of 5th International
Conference on Computer Supported Education
(CSEDU 2013), Aachen, Germany, 6-8 May.
Bouvier, P., Sehaba, K., Lavoue, E., George, S., 2013b.
Using Traces to Qualify Learner’s Engagement in
Game-Based Learning. Proceedings of 13th
International Conference onAdvanced Learning
Technologies, Pages 432 - 436.
Champagnat, R., Delmas, G., Augeraud, M., 2010. A
storytelling model for educational games: Hero's
interactive journey. Journal International Journal of
Technology Enhanced Learning archive Volume 2
Issue 1/2, Pages 4-20.
Conati, C., Maclaren, H., 2009. Empirically building and
evaluating a probabilistic model of user affect. User
Modeling and User-Adapted Interaction 19 (3), pp.
267–303.
Conlan, O., Hampson, C., Peirce, N., Kickmeier-Rust, M.,
2009. Realtime Knowledge Space Skill Assessment
for Personalized Digital Educational Games.
Proceedings of the Ninth IEEE International
Conference on Advanced Learning Technologies,
Pages 538-542.
Fu, F. L., Su, R. C., Yu, S. C., 2009. EGameFlow: A scale
to measure learners’ enjoyment of e-learning games.
Computers & Education 52, pp. 101–112.
Genaro, R. M., David, B., Sara, D., 2008. A Model of
Motivation for Virtual-Worlds Avatars. Proceeding
IVA '08 Proceedings of the 8th international
conference on Intelligent Virtual Agents, Pages 535-
536.
Genaro, R. M., Sara, D., Alma, R. G. G., 2009. A Model
of Motivation Based on Empathy for AI- Driven
Avatars in Virtual Worlds. Proceedings of the 2009
Conference in Games and Virtual Worlds for Serious
Applications, Pages 5-11.
Hou, H. T., 2012. Exploring the behavioral patterns of
learners in an educational massively multiple online
role-playing game (MMORPG). Computers &
Education 58, pp. 1225–1233.
Jimenez, C. O. S., Mesa, H. G. A., Rebolledo-Mendez, G.,
de Freitas, S., 2011. Classification of cognitive states
of attention and relaxation using supervised learning
algorithms. Proceedings of the IEEE International
Games Innovation Conference IGIC, Pages: 31-34.
Khenissi, M. A., Essalmi, F., Jemni, M., 2013a.
Presentation of a Learning Game for Programming
Languages Education. Proceedings of 13th IEEE
International Conference on Advanced Learning
Technologies, July 15-18, 2013, Beijing, China, Pages
324-326.
Khenissi, M. A., Essalmi, F., Jemni, M., 2013b. Toward
the personalization of learning games according to
learning styles. Proceedings of the international
Conference on Electrical Engineering and Software
Applications (ICEESA), March, 21-23, 2013
Hammamet, Tunisia, Pages.1-6.
Khenissi, M. A., Essalmi, F., Jemni, M., 2013c. A
Learning version of Pacman Game. Proceedings of the
Fourth IEEE International Conference On
Information and Communication Technology and
Accessibility (ICTA), October, 24-26, 2013,
Hammamet, Tunisia.
Khenissi, M. A., Essalmi, F., Jemni, M., 2014.
Comparison between serious games and learning
version of existing games. Proceedings of the 6th
World Conference on Educational Sciences. 06-09
February 2014, Malta, in press.
Kopeinik, S., Nussbaumer, A., Bedek, M., Albert, D.,
2012, Using CbKST for Learning Path
Recommendation in Game-based Learning,
Proceedings of the 20th International Conference on
Computers in Education, November, 26-30, 2012,
Singapore.
Mark, L., 2012. Identification OF Cognitive Traits
Through Game Method. Master of Science In
Information Systems. Athabasca University.
Moreno-Ger, P., Burgos, D., Sierra, J. L., Fernández-
Manjón, B, 2008. Educational game design for online
education. Computers in Human Behavior, 24, pp.
2530-2540.
Pablo, M. G., Pilar, S. T., Iván, M. O., José, L. S.,
Baltasar, F. M., 2007. Adaptive units of learning and
educational videogames. Journal of Interactive Media
in Education, Pages 1-15.
Pablo, M. Ger., Daniel, B., Javier, T., 2009. Digital games
in e-Learning environments: current uses and
emerging trends. Simulation & Gaming 40 (5), Pages
669-687.
Peters, C., Asteriadis, S., Rebolledo-Mendez, G., 2009.
Modelling user attention for human-agent interaction.
Proceedings of the 10th Workshop on Image Analysis
for Multimedia Interactive Services, Pages 266-269.
Pourabdollahian, B., Taisch, M., Kerga, E., 2012. Serious
Games in Manufacturing Education: Evaluation of
Learners’ Engagement, Proceedings of the 4th
International Conference on Games and Virtual
Worlds for Serious Applications. Procedia Computer
Science. Volume 15, Pages 256–265.
Stathacopoulou, R., Grigoriadou, M., Samarakou, M.,
Mitropoulos, D.,2007. Monitoring students' actions
and using teachers' expertise in implementing and
evaluating the neural network-based fuzzy diagnostic
model. Expert Systems with Applications. Volume 32,
Issue 4, May 2007, Pages 955–975.
CSEDU2014-6thInternationalConferenceonComputerSupportedEducation
152
