Method for Joining Information and Adapting Content from

Gamified Systems and Serious Games in Organizations

Mathieu Guinebert

, Joël Fabiani

, Melaine Cherdieu

, Pierre Holat

1,2 d

and

Charlie Grosman

FI Group, 14 Terrasse Bellini, 92800 Puteaux, France

Laboratoire d’Informatique de Paris Nord, Université Sorbonne Paris Nord, 93430 Villetaneuse, France

Keywords: Serious Games, Gamification, Design Method, Scenario Formalization Model, Adaptive Learning.

Abstract: In some work fields, the number of various knowledge and skills one must master can be tremendous.

Therefore, we decided to make work and training more rewarding and motivating. The skills and knowledge

mustered by the employees in both those situations are the same, but the systems responsible for the tracking

and the adaptation of the content are not. Therefore, our contribution is twofold. First, a system that centralizes

the learner’s game, learning and professional profiles and provides the other systems connected to it with the

necessary information to adapt their content thanks to various modules. Secondly, a generic model that should

be respected by any system connecting to our first element. We argue that it is necessary to use both our

method and model to be able to fully exploit the information provided by our system. We tested our model

and method on three different implementations but could not measure the impact of said implementations on

our learners.

1 INTRODUCTION

“Evidence from the field of labour economics

suggests a positive relationship between training and

firm productivity” (Bryan, 2006). Moreover, as it is

shown by the literature (Roussel, 2000), great

motivation implies a greater implication and a greater

efficiency in the given tasks and activities. Research

on gamification and serious games relies directly on

those principles. Indeed, they use game elements as a

motivational motor (Alsawaier, 2018).

With this concept as our basis, we decided to put

in place several systems in our company. Those

systems are dedicated, on the one hand, to making the

employee’s training more playful through learning

games and, on the other hand, to the gamification of

their everyday work. However, as indicated by (Dale,

2014), the efficiency of such methods in the case of

companies is not guaranteed. The use of a design

method allows us to limit the inefficiency risks of

https://orcid.org/0000-0002-1777-6708

https://orcid.org/0000-0002-9880-663X

https://orcid.org/0000-0003-1368-4053

https://orcid.org/0000-0003-4972-8679

https://orcid.org/0000-0001-7100-121X

those systems (Kappen & Nacke, 2013). Given the

fact we wish to put in place various educational

systems targeting the same skills (in a simulated

context, and in a real one), we seek to create and

implement a complex system that would allow those

educational systems to be joined around their player-

learner profiles, knowledge models, and skill models

(both the pedagogical and playful ones). We also aim

for this junction to be made around as their game logs

and the equivalences between “professional” and

“pedagogical” skills. This complex system, which we

named “Joint System” (JS), is destined to be modular.

The JS itself needs, also, a design method. Besides the

tracking of the learners and its skills, our system is

aimed at the increase of the playfulness of everyday

work and training. It focuses itself, on the one hand,

on maintaining the learner’s motivation through the

use of his/her logs to generate adapted playful

content, and, on the other hand, on adapting the

pedagogical content provided to the learner. We

338

Guinebert, M., Fabiani, J., Cherdieu, M., Holat, P. and Grosman, C.

Method for Joining Information and Adapting Content from Gamiﬁed Systems and Serious Games in Organizations.

DOI: 10.5220/0010997500003182

In Proceedings of the 14th International Conference on Computer Supported Education (CSEDU 2022) - Volume 2, pages 338-350

ISBN: 978-989-758-562-3; ISSN: 2184-5026

intend for the JS to allow the future addition of

modules such as ITS, authoring tools, CMS, etc. as

needed.

Moreover, the JS must be able to differentiate

skills issued from training and skills issued from

professional tasks carried out by the employees.

Therefore, the design method must consider the

design and implementation of both our educational

systems and of the adaptive modular system making

the junction between them.

To guarantee the optimal tracking of our player-

learner profiles, we decided to use exclusively

computer-mediated solutions for us to complete the

existing training. Those choices also seem quite

relevant regarding the increase in remote working

caused directly by Covid-19.

Our contribution can be summarized to three main

components. Firstly, a design method for educational

and playful systems taking part in our JS. Secondly, a

generic scenario formalization model that can be used

to describe both playful and pedagogical scenarios,

and, thirdly, how to link said scenarios to the JS.

2 WHY A METHOD AND A

MODEL

Some researchers present gamification and serious

games as synonymous inside of their work

(Caponetto et al., 2014), while others make the

differentiation (Landers, 2014; Deterding et al.,

2013). In our context, we decided to differentiate

them. We place the difference between the nature of

the gamified task. In this paper, a gamified system

refers to systems in which real tasks are being

gamified. Be it with or without game elements, the

tasks wouldn’t be any different and the consequences

on the results of the employee won’t change on a

professional scale. On the other hand, serious games

will refer to any systems constructed with a “serious”

intent and in which the tasks are both playful and

simulated. Therefore, any mistakes made in a

gamified system have “real” consequences as

opposed to mistakes made in a serious game.

2.1 The Need for a Method

The literature gives answer elements to our Research

Question n° 0 (RQ0) “How do we guarantee the

efficiency of gamified systems and learning games in

a company?”. Indeed, as indicated by (Kappen &

Nacke, 2013), the efficiency of a gamified solution is

directly dependent of its design. However, as (Nacke

& Deterding, 2017) suggest, the gamification and

everything linked to it has yet to reach full maturity.

Moreover, the recent literature on serious games

design method also translates a need for it. For

example, we could quote the work of (Avila-Pesantez

et al., 2019), who reminds us of this need in their

literature review before presenting their own method.

Therefore, those observations lead us to our RQ1

(directly obtained from our RQ0): “Which approach

is needed to guarantee both the efficiency and the

relevance of a complex system composed of several

serious games and gamified systems?”

Various leads can be found in the different

approaches available in the literature to help the

design of gamified systems and serious games. We

mostly focused ourselves on four of them that

decomposed their method into phases. Of course,

those four methods are not the only ones to do so but

we had to narrow down our choice to a manageable

subset. Two of those four methods are directly

focused on gamified systems: GOAL (Garcia et al.,

2017) and (Morschheuser et al., 2017). The two

others are centered around learning games: “the 6

facets” (Marne et al., 2012) and (Avila-Pesantez et

al., 2019).

Those four methods do not always agree on the

workflow. Our two learning games methods tend to

give far more freedom on the matter (in particular

(Marne et al., 2012)). There is also a lack of

consensus on the very nature and number of the

phases composing the method. Beyond specific

consideration like the obvious lack of pedagogical

objectives in the gamified systems methods, we can

find several common points such as the development

and evaluation phases.

However, none of those methods could satisfy us

fully. Indeed, we emit the Hypothesis n°1 (H1) that to

guarantee an optimal efficiency for our various

systems destined to be connected to our JS, they need

to be designed with the intent of being connected to

said system and its various modules. Therefore, in

order to verify our hypothesis, we need a design

method taking into account the specificities of our JS

that would allow for the design of both serious games

and gamified systems. Moreover, we also emit the

hypothesis H2 that it is possible to reach equivalences

between pedagogical and “professional” objectives in

such a way that “real” and simulated results could be

used freely by any systems connected to our modular

one. None of the methods and approaches that we

could find in the literature seemed to consider both

those hypotheses. Thus, explaining why, we had to

create our own.

Method for Joining Information and Adapting Content from Gamiﬁed Systems and Serious Games in Organizations

339

2.2 The Need for a Model

(Liu et al., 2017) reminds us that the benefits of an

adaptive system on learning are fully admitted but

needs a particular attention to its design. Moreover,

as indicated by (Peng et al., 2019), the apparition of

new technologies, most notably in the domain of Big

Data and Data Analysis, incites us to construct new

forms of learning using those information and

technologies to better adapt to the learner.

Our JS allows us to link numerous concepts and

identical skills but implemented and evaluated by

different systems. Those concepts and skills can be of

pedagogical nature, of course, but also of gamified

nature. When a learner fails a gamified task, are we

sure that he/she failed because of a lack of

pedagogical skill? Or could he/she have failed

because of gaming aspects? In this context, the

possibilities offered by our joint system on the

modeling of the learner has led us to consider both the

playful and pedagogical adaptation of the content in

our project.

As is indicated by the frequently cited Flow

Theory (Csikszentmihalyi, 2000), it is important to

adapt the difficulty of the task to the learner’s skill

level in order for his/her motivation not to plummet.

Our player-learners are evolving in systems that can

link both playful and pedagogical aspects. Therefore,

the adaptation of the content can’t rely on a unique

Flow curb, but should rely on at least two curbs, one

for the pedagogical aspects and one for the game

aspects.

Thus, our hypothesis H3 in the scope of this

project is that, in the context of serious games and

gamified systems, the scenario to which the learners

are exposed must be both adapted from a pedagogical

point of view and from a playful point of view to

ensure the learner’s optimal motivation. Therefore,

we seek to answer the following RQ2 “Which

formalism or model to adopt in order to ease the

adaptation and differentiation of playful and

pedagogical scenarios?”.

Regarding pedagogical models, IMS-LD

(Hummel et al., 2004) is still today regularly cited

(Ouadoud et al., 2018; El Moudden & Khaldi, 2018).

IMS-LD address every problematic linked to the

modeling, the design, and the organization of a

system’s pedagogical content. The Pleiades method

(Villiot-Leclerq, 2007) is another interesting

approach for the modeling of pedagogical scenarios.

However, none of those two methods has been

truthfully conceived to consider playful elements.

More recent approach such as “MoPPLiq” (Marne

et al., 2013) or “Multiplayer Learning Game

Ontology” (MPLGO) (Guinebert et al., 2017) makes

the link between pedagogical and playful elements.

Yet they are still not perfect for our context. In the

case of MoPPliq for example, every activity sequence

available to the learners need to be defined directly in

the model. An adaptation using this model would thus

be limited to the links defined in the scenario.

In MPLGO, the precedencies between the

activities are flexible and are determined by the game

resources produced and consumed by the players.

However, the knowledge and skills are not considered

in the construction of those precedencies. An

adaptation based on MPLGO would only rely on

playful elements which does not answer our

problematic.

One of the closest answers we could find for our

needs toward a model considering H3 was the model

and methods proposed by (Marfisi-Schottman, 2012)

which infers its pedagogical structure directly from

IMS-LD and differentiate the game scenario. This

model and method have been designed to help the

communication between the various individual

working on the Learning Game and seems to tackle

the game scenario mostly on a narrative scale. Every

detail of the most atomic component, the screen,

which involves the interactions with the Learning

Game itself is left to the screen designer.

The models associated with the gamified such as

the GOAL ontology (Garcia et al., 2017) also fails to

satisfy our needs. The pedagogical aspect is, for

obvious reasons, often nonexistent in said models.

We failed to find a method or model in the

literature that would satisfy our needs to adapt a

scenario on both its playful and pedagogical aspects

depending on the learner profile. Therefore, we had

to create our own model to answer our RQ2.

3 MODEL FOR ANY

EDUCATIONAL SYSTEM

CONNECTED TO OUR JOINT

SYSTEM

To answer our RQ2, we seek to treat 5 specific

aspects:

• Activities granularities

• Playful and pedagogical aspects

differentiation

• Genericity toward any educational system

• Simplicity of use (accessible to a non-

expert)

• Connectivity to the joint system

CSEDU 2022 - 14th International Conference on Computer Supported Education

340

Figure 1: Model for an educational system and its connexion to the joint system.

To construct our model, we decided to take

inspiration from both MPLOG (Guinebert et al., 2017)

and MoPPLiQ (Marne et al., 2013). It is interesting to

note that MoPPLiQ has been conceived along with

“the 6 facets” which we evoked in section 2.1.

To manage the granularities of our activities we

decided on a three degrees scale:

• Levels

• Mission

• Action

Method for Joining Information and Adapting Content from Gamiﬁed Systems and Serious Games in Organizations

341

3.1 Levels

A Level can precede another Level and is composed

of several Missions. There are three key elements to

understand how to manage our Level degree. First,

targeted skills and knowledge are primarily linked to

Missions, not Levels. Therefore, since Missions are

the components of our Levels, their targeted skills are

linked to the Level they are a part of.

However, and it is our second element, the way

Missions are presented to the learner is entirely up to

the educational system we are modeling. Thus, it is

entirely possible to represent Levels as an orderly

sequence of mandatory Missions or as a multiset of

free access Missions with no obligation for the

learners.

For example, you could use our Levels to

represent a quiz for which each question would be

considered a Mission or use them to represent an open

world game where each Mission would be playful

objectives and quest disseminated throughout the

land.

Finally, the third element to note toward our Level

degree is the way we consider precedencies between

Levels. Our model allows you to freely define your

precedencies. There are no restrictions, be it an

absence of precedencies, fixed precedencies such as

in MoPPLiq, dynamic precedencies as in MPLOG or

dynamic precedencies linked to only skills or both

skills and system components.

3.2 Missions

Our Mission degree is here to represent a set of

objectives for the learner. You can use them to

represent any kind of objectives, be it professional,

pedagogical, or playful. For example, achievement,

quests, questions, rapports, meeting objectives,

financial incentives, etc. could all be considered as

Missions depending on the modeled system.

To determine the fulfillment of a Mission, the

system must check if its current state has reached the

objectives part of the Mission. The way precedencies

work for Missions in our model, is the same as in our

Level degree.

Missions can target several knowledge and skills.

It is the fulfilment or not of those Missions by the

learner that should indicate the system whether a

learner mastered, failed, and/or experienced the

targeted skills.

As mentioned earlier, Missions can be linked to

either pedagogical, professional, or playful

objectives. Moreover, we expressed both the

hypothesis that equivalencies could be made between

pedagogical and professional skills, and that several

flow curbs should be considered in our systems (H2

and H3). Therefore, we consider here that skills and

knowledges can be of either of these three natures.

3.3 Action

Our Action degree corresponds to the most atomic

degree of our model. The learner evolves in Levels to

fulfill a Mission, but ultimately, there is only one

thing that he/she ever does: Actions. In the same way

Roles worked in MPLOG, in our model, Actions

consume and produce System Objects and nothing

else.

What is interesting are the consequences of those

production and consumption of Objects. To fully

understand their reach, we must explain both what

can be a System Object and how it affects the system.

A System Object can be anything useful to the

modeling of the scenario or the system’s functioning.

It can be game resources, Boolean flags, files, credits,

points, given answers, etc.

Those Objects can then be found in two categories

of Inventories like what can be found in MPLOG.

They can be found in Environment/World/System

Inventories which represent the objects available to

every learner connected to the system and/or

available to the system itself. They can also be found

in personal inventories. The objects in those

inventories are only related to the learner those

inventories are linked to.

3.4 Connection to the Joint System

Our joint system must fulfill three purposes toward

any system connected to it:

• Trace tracking

• Profile Evaluation

• Adaptive Scenarios

Since Actions are the only things done by

learners, they are what drives any kind of evaluation

and any kind of trace tracking. When an Action is

done, it can, or not, lead to an evaluation by the

system. It is defined by the modeled system.

Therefore, we make a distinction between the

System Evaluation Module and the Joint System

Evaluation Module. The first one is ad hoc to the

modelled system and can be mustered by Actions, end

of Levels, Missions’ fulfilments, etc. The second

evaluates normalized logs (in our case our own xAPI

template, but you could use your own for your own

joint systems) sent to the joint system’s LRS.

Thus, when using our joint system model, the

evaluation sequence is as follows:

CSEDU 2022 - 14th International Conference on Computer Supported Education

342

1. An Action is done by the learner

2. This Action leads ultimately to an ad-hoc

evaluation by the system

3. The system evaluation module generates a

normalized log of the evaluation and Action

and sends it to the LRS

4. The Joint System evaluation module

acknowledges the log

5. It evaluates the impact on the learners

involved and modifies their profile

consequently

The adaptive part is similar but a little bit more

direct. Once again, we have two different modules. A

System adaptative module and a Joint System

adaptative module. Both those modules are activated

only when the modeled system considers it necessary;

that is to say, when the modeled system estimates that

an adaptation of its content could be useful to the

learner. This adaptation could happen at the start of

the session to determine the best Level for the player,

to select the best Mission during a Level or even to

generate new Missions and Level.

The JS adaptive module is modular itself. It can

be perceived as a toolbox usable by the modeled

system’s module. The role of the JS adaptive module

is to provide the various feedback from the tools

queried by the module from the modeled system. In

our case, our current adaptive module has only one

tool in its box. The role of this tool is to establish, for

a given learner, a priority order for any skills or

knowledge that has been passed on to it. Our tool

currently bases itself on four key aspects:

precedencies as in CbKst (Doignon, 1994), scoring,

system type and knowledge maintenance. Therefore,

the sequence of actions between our two modules

using this tool is as follows:

1. The System module signals a need for

adaptation

2. It establishes a list of knowledge and skills

that could be targeted and sends it to the

Joint System Module

3. The Joint System module analyzes the

impact each skill’s mastery would have on

the learner and ranks it by order of priority

4. It sends back the ranking order to the System

Module which uses it in an ad hoc way to

adapt the learner’s scenario.

Thanks to the division of both the evaluation and

adaptation modules as two separate entities, it is

possible to connect any kind of educational system to

our joint system as long as its ad hoc modules make

the interface with our normalized ones. It also makes

the independent evolution of the modeled and joint

systems feasible. This last point is quite important in

a company context where there is a need for constant

production of new training systems and improvement

of old ones. The normalization of the data by and to

the joint system makes it possible.

3.5 Application and Conclusions

We are in a professional context. Therefore, we will

probably never use this model to represent “pure”

video games without any serious components.

However, we need it for 100% pedagogical and/or

professional application. Therefore, we asked

ourselves if our system was generic enough to

represent any kind of system with various degrees of

playful, pedagogical, and professional components.

Moreover, we also wanted to test our hypothesis

which claims that to fully exploit our joint systems, it

is necessary for our educational systems to be

conceived and implemented with our joint system in

mind.

Table 1: Conclusion on Genericity of our model and

possibility of Connection to our JS on various kinds of

systems.

Genericit

Connection to JS

pothetical OK OK

Implemente

OK OK

Publicly

Available

OK Only Evaluation

3.5.1 Genericity

To test our genericity, we established a list of 10

systems that we modeled without any connection to

the Joint System. Of those 10 systems, 4 are

hypothetical systems that could be useful for our

company, 3 are systems we internally developed and

conceived using our method and the last 3 are systems

broadly available to the public.

2 of the hypothetical systems, 2 of the internally

developed ones and 2 of the public ones (namely

Voracy Fish (G Interactive, 2012) and ClassCraft

(Sanchez et al., 2015)) are or would be generally

classified as serious games. The third hypothetical

system is a gamified system, and the third internally

developed system is a quiz system with very limited

playful elements. The fourth hypothetical system tries

to emulate a pedagogical system that would give

information on whether a learner succeeded, partici-

pated and was a present to a non-computer-mediated

training. Finally, the last system we wanted to model

was a pure video game that has often been transformed

or used toward more pedagogical solutions (Bos et al.,

2014; Ekaputra et al. 2013): Minecraft.

Method for Joining Information and Adapting Content from Gamiﬁed Systems and Serious Games in Organizations

343

For the most open world and/or generic systems

like Minecraft or Classcraft we had to establish

generic Missions. Otherwise, we were able to model

every system thus leading us to consider our approach

as sufficiently generic.

3.5.2 Thoughts about the Connection to the

Joint System

Afterwards, we pondered on how we would connect

these models to the joint system. The hypothetical and

implemented systems have been designed with our JS

in mind. Therefore, there is obviously no challenge

whatsoever to model their connection to it. For the

three others, it would be possible to create an

interface that would normalize their produced logs

and link them to the associated skills and knowledge.

Thus, even if a system has not been conceived to be

connected to our Joint System, the evaluation part can

be maintained. This is important because it means we

can use those data to improve the information we

have on a learner profile.

However, the same cannot be said for the

adaptation part. We can distinguish two cases. First-

case scenario, the modeled system has no adaptation

module whatsoever. It is therefore impossible for it to

use any information produced by our joint system.

Second-case scenario, the modeled system has an

adaptation module. If it is possible to interact with the

module, an interface could hypothetically be made to

normalize the data coming from and toward our joint

system adaptation module. But this interface remains

hypothetical because there is no assurance that the

exchange sequence, that we established in section 3.4

for the adaptation modules, will be followed by the

module of the modeled system. In the case where it is

not possible to communicate with the modeled

module, it is, of course, not possible for the modeled

system to use the Joint System adaptation module.

Therefore, only a handful of systems seems to be

able to infirm our hypothesis H1 and their existence

remain hypothetical. Thus, we can consider our

hypothesis H1 to be mostly verified.

4 DESIGN METHOD

4.1 First Abstraction Level

As specified in section 2.1, to answer our RQ1, we

posed two hypotheses, the first one being that it is

possible to have equivalencies between pedagogical

and professional objectives in such a way that “real”

and simulated results can be freely exploited by any

systems using said objectives.

The second one, which we attempted to address

in section 3.5.2 is that it is necessary to take those

equivalences and the connection to our Joint System

into consideration in the design phases of any

educational system destined to be connected to the

Joint System.

None of the approaches we could find in the

literature seemed to take both those hypotheses into

consideration. As indicated in section 3.5, the

gamified systems and learning games are not the only

systems that could benefit from being connected to

the joint system. Yet, most of the systems that are

interesting to us comprise either playful and/or

pedagogical/professional components.

Given those last two facts, there is an obvious

interest for us to inspire ourselves from learning

games and gamified design methods. However, it is

also important to note that our method cannot be a

simple fusion of said methods since we must consider

both the joint system and the genericity of the

method.

To conceive our method, we decided on an

approach dividing it into several phases. Meaningful

examples of such a division for methods can be found

in the literature, notably in (Garcia et al., 2017),

(Morschheuser et al., 2017) and (Avila-Pesantez et

al., 2019). Thus, the first question we had to ask

ourselves to design our method was: “Which division

do we have to adopt in the case of a joint system’s

design method?”

This division and its workflow are not trivial since

our method aims to be able to help design any system

including playful and/or educational elements that

would be connected to our joint system. Indeed, if it

is true that, the approach available in the literature,

share common phases, they also have their

differences toward their composition and/or

articulation. Therefore, it is possible to take

inspiration from them, but, as indicated in section 2.1,

they cannot be used in their current state to answer

our problematic.

To establish our phases, we listed the steps one

needed to/could take in order to construct those

systems. Those steps, that we will further detail in

section 4.2, and the existing phases in the literature

lead us to establish six different phases:

1. Preparation and Analysis

2. Context Determination

3. Junction and Constraints

4. Ideation and Design

5. Development

6. Evaluation

CSEDU 2022 - 14th International Conference on Computer Supported Education

344

Figure 2: Method for the design and implementation of an educational system destined to be connected to a joint system.

Method for Joining Information and Adapting Content from Gamiﬁed Systems and Serious Games in Organizations

345

In our case, the number of employees working on

the design of those systems is little. Thus, we had to

take this into consideration in the workflow of the

phases. We decided to inspire ourselves from the

learning games approach and make it quite free. As

you can see it in Figure 2, it is possible to get back to

previous phases from any of the more advanced

phases. This allows a more agile approach for which

multiple iterations are easier to handle. However,

even with this “free” workflow, we consider it as a

sequence: you must begin with the first phase and

muster your way up to the sixth one.

4.2 Second Abstraction Level

We will detail what is to be done for each phase. We

will also indicate our proposed workflow for its

various steps.

4.2.1 Phase 1: Preparation and Analysis

This phase is common to almost every method or

approach we could find. In this phase, one must

define three key elements:

• The system’s type

• The system’s objectives

• The concerned learners’ profiles

Here, the system’s type refers to the kind of

system you wish to design: a learning game, a

gamified system, a quiz, etc. The very nature of this

system is both what will determine the nature of the

objectives (pedagogical, playful and/or professional)

and the profiles that will be addressed by it.

Yet, it could be argued that it is the objectives that

lead to the creation of the system, and therefore that

the type of system is determined by the objectives one

wishes to address. A similar reasoning can be found

for the learner’s profile. This explains why we

established a free workflow between those steps.

Those steps are the fundamental reason behind the

design and the implementation of your system. This

explains why we grouped them together in this phase.

For example, our juridical team wanted to teach

some juridic aspects to our collaborators, thus,

defining the targeted profiles as well as rhe

pedagogical and professional objectives. We decided

that a learning game would be the best way to provide

the said teaching, therefore defining the system’s

type. Finally, by choosing this type, we had to define

playful objectives for our learning game.

4.2.2 Phase 2: Context Determination

A similar phase can be found in “the 6 facets” (Marne

et al., 2012): in this phase the user must define the

global usage context of his/her system. We

distinguished three steps for which we established a

free workflow:

• Define the presence and nature of other

agents

• Define the context itself

• Define the metrics, KPI and tools available

The first step allows you to establish whether you

intend your learners to interact with other learners or

not and whether you want them to confront

themselves to computer-controlled opponents. We

consider it as part of the context because it directly

impacts the way you can use your system (which is

addressed by the second step).

In the second step, you must establish your

system’s use case, the tasks’ nature, whether your

system is for initial training only or not, whether you

want to involve groups of synchronous plays or

sequences of asynchronous ones, etc. In short, you

must define the situations your system will be used in.

The third and final step of this phase is more

related to the context of the organization itself. You

must ask yourselves which tools, KPI and metrics are

available to you and/or the learners and how you

could use them jointly with your system.

The juridic game we cited as an example in 4.2.1

is a mono-player game with an asynchronous

leaderboard and no computer-controlled opponent. It

is not limited to the initial training of employees and

can be linked to various home tools.

4.2.3 Phase 3: Junction and Constraints

This third phase has once again a free workflow and

is composed of three steps:

• Define the constraints

• Identify the element exploitable by the joint

system

• Identify the element exploitable by the

designed system

Similarly to what can be found in (Morschheuser

et al., 2017), the constraints refer to any kind of

constraints one could apply toward the

implementation of the system. What are your

deadlines? What is your budget? Who is available to

design and implement the system? What about

GDPR/legal questions? etc.

The two other steps are only found in our

approach. As of yet we only tested our method and

joint system with the sharing and adaptation of skills

and knowledge. However, we also think that game

design elements and logs could be used to generate

adaptive opponents and/or interfaces and are

CSEDU 2022 - 14th International Conference on Computer Supported Education

346

currently working on their addition to the JS adaptive

module (see section 3.4).

In those two steps the designer must consider

every information that will be produced by the system

and all information that is available in the joint one to

establish how the designed system could improve the

others or be improved by them.

The juridic learning game that serves as our

example needed to be developed quickly and with no

additional funds besides salaries. It provided the JS

with traces regarding the mastering or not of various

juridic and playful skills and knowledge. By linking

it to the JS it was able to use the learning adaptive

module. This module implies the respect of GRPD

laws, but also allows the learning game to exploit the

other systems’ traces for its adaptation.

4.2.4 Phase 4: Ideation & Design

This phase is composed of six different steps:

• Formalization of the Missions/Activities

• Simulation of the pedagogical/professional

domain

• Identification of the motivating elements

• Identification of the interaction mechanics

with the system

• Determination of the Decorum

• Pedagogical/Professional and Playful

scenario formalization and determination of

its adaptation mechanisms.

Formalizing the Missions and/or Activities (by

using our model) allows you to establish the event and

rules that will drive the system. You can directly link

them to your objectives and use these formalizations

to ensure that every profile and goal has been taken

into consideration.

The simulation of the domain (that you can find

in “the 6 facets”) means that you must establish how

your computer-mediated solution represents and

simulate the activities and/or pedagogical-

professional tasks.

Typically, the motivating elements that you must

identify in the third steps are game design elements.

Is there a reward system? How is it designed? What

about the global economy of your system?

The interaction with the system (also found in

“the 6 facets”) establish and formalize the Actions a

learner can make with the system. Those interactions

codify what can and cannot be done by the user. It can

directly be linked both to the formalization of the

Missions and to the simulation of the domain.

The Decorum is also defined in “the 6 facets”. It

is mostly a playful component and is linked to the

motivating elements. The decorum is defined by the

graphical elements and narration of your system.

The final step helps you consider how the

junction’s steps of the third phase interact with every

other step of the phase 4; notably, the motivating

elements and Missions.

The juridic learning game was designed by

inspiring ourselves from the goose board game. The

Decorum was directly inspired by it and we divided

the obtained board in steps according to the process

we intended to simulate. The learboard served as a

motivating element and the desire to beat it was

modeled by a Mission. A session of the game is

modelled by a Level and each part of the process is

modelled by a Mission.

4.2.5 Phase 5: Development

We determined three steps that one needs to consider

while developing the system:

• Connection to the joint system

• Development of the mechanics

• Development of the system’s evaluation and

adaptation module

The first and second steps can be done in any

order you want. The first step stipulates that you must

develop and consider the way your modules will

connect to the joint system. Depending on the

system’s nature and/or programming language, you

will be able to reuse previously implemented

connection modules.

The second step is the development of the system

itself. We won’t delve into too many details for this

step because we think it should be left to your

decision which development method is the best.

Finally, the third and final step of this phase seek

to implement the final step of phase 4 by relying on

the implementations and development made in the

first two steps of this phase.

4.2.6 Phase 6: Evaluation

Every method needs an evaluation phase. In our case

we divided it in three steps with free workflow:

• Evaluation of the system’s impact on the

learner

• Evaluation of the system’s contribution to

the joint system

• Evaluation of the joint system’s contribution

to the designed system

Those three evaluations are closely linked

together but do not rely on the exact same indicators.

The system’s impact can be measured either by

Method for Joining Information and Adapting Content from Gamiﬁed Systems and Serious Games in Organizations

347

looking into the profile’s evolution or by looking at

the learner’s performance at similar tasks.

The designed system’s contribution is directly

linked to elements and information it shared with the

joint system. Until those elements and information

are used by another system connected to the joint

system, the design system’s contribution will remain

poor. Yet those data can still be used to make reports

on a learner’s performance.

Similarly, if no previously connected systems

shared information and elements useful to the

designed system, the usefulness of the joint system

will be kept at a low point for the designed system.

However, since the joint system produces adaptive

data useful to the designed system it keeps a modicum

of usefulness even in this case.

The juridic learning game is currently the only

system currently providing juridic traces to our JS.

The evaluation of its impact on other systems is

therefore limited at the moment. We were able collect

data and evaluate the juridic skills of our

collaborators. Moreover, the game was able to make

full use of the adaptive module.

5 CONCLUSIONS

We seek to create a complex system that would allow

data, game elements, learners’ profiles, logs, etc. to

be freely shared between systems of various natures

that are used in various contexts. We also want this

system to provide a set of normalized tools usable by

pedagogical, playful and professional systems to

adapt their content to the learner/player/user.

The creation of said system led us to ask ourselves

two different Research Questions:

• RQ1: “Which approach is needed to

guarantee both the efficiency and the

relevance of a complex system composed of

several serious games and gamified

systems?”

• RQ2: “Which formalism or model to adopt

in order to ease the adaptation and

differentiation of playful and pedagogical

scenarios?”

To answer our RQ2, we conceived a generic

model that we used in three case scenarios. Firstly, we

used it to model hypothetical systems that would be

more or less playful. Secondly, we used it to model

existing systems available in the literature or

commerce. Thirdly, we used it to model and

implement systems designed for our company.

Thanks to those three case scenarios, we

concluded that our model seems to fulfill most of the

five aspects we considered in order to answer our

RQ2:

• Activities granularities

• Playful and pedagogical aspects

differentiation

• Genericity toward any educational system

• Simplicity of use (accessible to a non-

expert)

• Connectivity to the joint system

Of those five aspects, only the simplicity of use

could not be tested yet. The results we obtained by

testing the connectivity to the joint system seems to

confirm H1 that stipulates that a system should be

designed with the JS in mind to best exploit it. Indeed,

our results seem to show that it is possible to exploit

data from a system not designed to be used jointly

with our JS, but that it would be difficult for this

system to use the available adaptive tools.

As seen in section 2.1, the best way to answer our

RQ1 seems to be the use of a dedicated method.

Given our hypotheses and that this is still an ongoing

research in the field, we decided to create our own

method.

This method used jointly with our model allowed

us the design and the implementation of three

different systems. Those three still lack skill overlap

with each other thus leading, at the moment, to their

limited evaluation. However, the first results

regarding the usability of the method and model are

quite encouraging. Our current joint system adaptive

module comprises a unique tool similar to a really

lightweight ITS that provides feedback on demand to

the connected system. To do so, this tool takes into

consideration the skill nature, the memorization

process, the precedencies between objectives and the

leaner’s global evaluation.

It is important to note that, as shown previously,

since systems must be designed while knowing the

various existing tools they can use, the addition of

new tools to our adaptive module should mean that

those shouldn’t be usable by previously designed

system.

If such a case were to happen, the only way for

previously designed systems to use those new tools

would be to reiterate our design method on phase 3.

Our future works will be driven by two different

axes. Firstly, verify the equivalence between

pedagogical and professional skills in our systems

(H2). Secondly, improve the JS adaptive module

For our first axis, we aim to design and implement

new gamified systems and training tools that could be

used in parallel to our existing learning games. Those

new systems would provide us with both information

that could infirm or confirm H2. Moreover, with more

CSEDU 2022 - 14th International Conference on Computer Supported Education

348

systems implemented, we would, of course, have

more knowledge and skills overlap and give more

usefulness to our JS.

For our second axis, the improvement of the

module is twofold. First, we need to upgrade our

existing tool. To do so we intend to rely more on

previously acquired data in order to change the way

priorities are decided for our objectives. At the

moment our current adaptive tool tries to improve

knowledge maintenance by making the player/learner

redo the activities at increasing intervals. We intend

to further fine-tune this aspect in the future.

Secondly, we need to make our module more

versatile for it to accept more adaptive tasks than just

the objectives adaptation (note that an objective can

be professional and/or playful and is not limited to

pedagogical objectives). A way to reach this goal is

to create new tools that could be used by our designed

systems. For example, we intend to add adapted

opponent generations by using the shared logs and

profiles to establish adapted behavior and difficulties.

REFERENCES

Alsawaier, R. S. (2018). The effect of gamification on

motivation and engagement. The International Journal

of Information and Learning Technology.

Avila-Pesantez, D., Delgadillo, R., & Rivera, L. A. (2019).

Proposal of a Conceptual Model for Serious Games

Design: A Case Study in Children With Learning

Disabilities. IEEE Access, 7, 161017-161033.

Bos, B., Wilder, L., Cook, M., & O'Donnell, R. (2014).

Learning mathematics through Minecraft. Teaching

Children Mathematics, 21(1), 56-59.

Bryan, J. (2006). Training and performance in small firms.

International small business journal, 24(6), 635-660.

Caponetto, I., Earp, J., & Ott, M. (2014, October).

Gamification and education: A literature review. In

European Conference on Games Based Learning (Vol.

1, p. 50). Academic Conferences International Limited.

Csikszentmihalyi, M. (2000). Beyond boredom and

anxiety. Jossey-Bass.

Dale, S. (2014). Gamification: Making work fun, or making

fun of work?. Business information review, 31(2), 82-

90.

Deterding, S., Björk, S. L., Nacke, L. E., Dixon, D., &

Lawley, E. (2013). Designing gamification: creating

gameful and playful experiences. In CHI'13 Extended

Abstracts on Human Factors in Computing Systems

(pp. 3263-3266).

Doignon, J. P. (1994). Knowledge spaces and skill

assignments. In Contributions to mathematical

psychology, psychometrics, and methodology (pp. 111-

121). Springer, New York, NY.

Ekaputra, G., Lim, C., & Eng, K. I. (2013). Minecraft: A

game as an education and scientific learning tool.

ISICO 2013, 2013.

El Moudden, F., & Khaldi, M. (2018). Developing an ims-

ld collaborative project creation application coproline

(Collaborative Project Online). International Journal of

Engineering Applied Sciences and Technology, 3(4),

38-43.

Garcia, F., Pedreira, O., Piattini, M., Cerdeira-Pena, A., &

Penabad, M. (2017). A framework for gamification in

software engineering. Journal of Systems and Software,

132, 21-40.

Guinebert, M., Yessad, A., Muratet, M., & Luengo, V.

(2017, September). An ontology for describing

scenarios of multi-players learning games: toward an

automatic detection of group interactions. In European

Conference on Technology Enhanced Learning (pp.

410-415). Springer, Cham.

Hummel, H., Manderveld, J., Tattersall, C., & Koper, R.

(2004). Educational modelling language and learning

design: new opportunities for instructional reusability

and personalised learning. International Journal of

Learning Technology, 1(1), 111–126

Interactive, G. E. N. I. O. U. S. (2012). Voracy Fish: New

multiplayer serious game for physical rehabilitation

(2012).

Kappen, D. L., & Nacke, L. E. (2013, October). The

kaleidoscope of effective gamification: deconstructing

gamification in business applications. In Proceedings of

the First International Conference on Gameful Design,

Research, and Applications (pp. 119-122).

Landers, R. N. (2014). Developing a theory of gamified

learning: Linking serious games and gamification of

learning. Simulation & gaming, 45(6), 752-768.

Liu, M., McKelroy, E., Corliss, S. B., & Carrigan, J. (2017).

Investigating the effect of an adaptive learning

intervention on students’ learning. Educational

technology research and development, 65(6), 1605-

1625.

Marfisi-Schottman, I. (2012). Méthodologie, modèles et

outils pour la conception de Learning Games (Doctoral

dissertation, INSA de Lyon).

Marne, B., Carron, T., Labat, J. M., & Marfisi-Schottman,

I. (2013, July). MoPPLiq: a model for pedagogical

adaptation of serious game scenarios. In 2013 IEEE

13th International Conference on Advanced Learning

Technologies (pp. 291-293). IEEE.

Marne, B., Wisdom, J., Huynh-Kim-Bang, B., & Labat, J.

M. (2012, September). The six facets of serious game

design: a methodology enhanced by our design pattern

library. In European conference on technology

enhanced learning (pp. 208-221). Springer, Berlin,

Heidelberg.

Morschheuser, B., Hamari, J., Werder, K., & Abe, J.

(2017). How to gamify? A method for designing

gamification. In Proceedings of the 50th Hawaii

International Conference on System Sciences 2017.

University of Hawai'i at Manoa.

Method for Joining Information and Adapting Content from Gamiﬁed Systems and Serious Games in Organizations

349

Nacke, L. E., & Deterding, C. S. (2017). The maturing of

gamification research. Computers in Human

Behaviour, 450-454.

Ouadoud, M., & Chkouri, M. Y. (2018, October). Generate

a meta-model content for disciplinary information

space of learning management system compatible with

IMS-LD. In Proceedings of the 3rd International

Conference on Smart City Applications (pp. 1-8).

Peng, H., Ma, S., & Spector, J. M. (2019). Personalized

adaptive learning: an emerging pedagogical approach

enabled by a smart learning environment. Smart

Learning Environments, 6(1), 1-14.

Roussel, P., & Laboratoire interdisciplinaire de recherche

sur les ressources humaines et l'emploi (Toulouse).

(2000). La motivation au travail: concept et théories.

LIRHE, Université des sciences sociales de Toulouse.

Sanchez, E., Young, S., & Jouneau-Sion, C. (2015, June).

Classcraft: de la gamification à la ludicisation. In 7ème

Conférence sur les Environnements Informatiques pour

l'Apprentissage Humain (EIAH 2015) (pp. 360-371).

Villiot-Leclercq, E. (2007). La méthode des Pléiades: un

formalisme pour favoriser la transférabilité et

l’instrumentation des scénarios pédagogiques. Sciences

et Technologies de l'Information et de la

Communication pour l'Éducation et la Formation,

14(1), 117-154.

CSEDU 2022 - 14th International Conference on Computer Supported Education

350