Luciole: The Design Story of a Serious Game for Learning English

Created for (and with) Children

Anne-Laure Guinet

1 a

, Audrey Serna

1 b

, L

eo Vanbervliet

, Emilie Magnat

2 c

Marie-Pierre Jouannaud

4 d

, Coralie Payre-Ficout

3 e

and Mathieu Loiseau

1 f

INSA Lyon, CNRS, Universite Claude Bernard Lyon 1, LIRIS, UMR 5205, 69621 Villeurbanne, France

Ecole Normale Sup

erieure de Lyon, CNRS, ICAR, UMR 5191, France

Universit

e Grenoble Alpes, LIDILEM, France

Universit

e Paris 8, TRANSCRIT, France

Keywords:

Serious Games, User-Centered Design, English as a Foreign Language (EFL), Young Learners.

Abstract:

In France, elementary school curriculum mandates 54 hours of foreign language instruction for pupils, typ-

ically English. However, many teachers feel linguistically insecure. Addressing this, we propose Luciole, a

serious game designed to introduce 6 to 8-year-olds to English. Focused on oral comprehension, Luciole al-

lows for autonomous use, facilitating differentiated activities. Luciole was developed through an iterative and

inclusive series of design-development-user test-feedback analysis-redesign. Throughout this process, various

methods were employed, including participatory workshops with children in the context of real classroom

sessions. This article, through the description of the various design stages and successive prototypes, seeks

to clarify a number of more general design issues related to serious learning games intended for use in the

classroom, by pupils new to English, independently, and under the supervision of their teacher, who may be

unfamiliar with the use of digital tools for learning.

1 INTRODUCTION

In 2002, a major change was orchestrated by the

French government in terms of status of Foreign

Language (FL) learning in elementary schools (Dat

and Spanghero-Gaillard, 2005). The Ofﬁcial Bulletin

(BO)

of February 14 2002 made FL an ofﬁcial sub-

ject of elementary school with a weekly hourly rate

(MEN/MR, 2002). This decision was turned into a

law in 2013 (MEN, 2013).

Despite this intention, the enforcement of such

policies is not trivial. Elementary school teachers need

to achieve equilibrium between “fundamental” skills

and other mandatory subjects when some are inte-

grated to the national evaluation campaigns and some

https://orcid.org/0000-0002-2957-1849

https://orcid.org/0000-0003-1468-9761

https://orcid.org/0000-0001-8857-9405

https://orcid.org/0000-0003-1036-381X

https://orcid.org/0000-0002-0156-9122

https://orcid.org/0000-0002-9908-0770

The French education ministry enforces decisions

through publishing Bulletin Ofﬁciel (BO).

are not. Often, FL play the part of adjustment vari-

able, not only because of the difﬁcult balance between

subjects but also due to the teachers’ own difﬁculties

with FL. Few elementary school teachers are FL spe-

cialists and many feel linguistically insecure (MENJ,

2019; Delasalle, 2008). One of their difﬁculties is the

priority set on the oral language, going from compre-

hension to production through repetition (MEN, 2015,

p. 29). Some teachers fear being unable to produce ac-

ceptable utterances and fail to familiarize pupils with

appropriate phonological references (Delasalle, 2008).

We thus decided to create a Serious Game (SG) for

young learners of English (6–8 year-olds). We wanted

to harness the possibilities offered by games to provide

teachers with an application that learners could play

autonomously, thus providing opportunities for differ-

entiated teaching, while also providing learners with

access to some of the building blocks of subsequent

English learning activities (input and lexicon, culture,

comprehension strategies). The product of this work

is called “Luciole”

LUdique au service de la Compr

ehensIon Orale en

Langue

Etrang

ere i.e. playing for improving oral compre-

hension of a foreign language

380

Guinet, A.-L., Serna, A., Vanbervliet, L., Magnat, E., Jouannaud, M.-P., Payre-Ficout, C. and Loiseau, M.

Luciole: The Design Story of a Serious Game for Learning English Created for (and with) Children.

DOI: 10.5220/0013285900003932

Paper published under CC license (CC BY-NC-ND 4.0)

In Proceedings of the 17th International Conference on Computer Supported Education (CSEDU 2025) - Volume 1, pages 380-391

ISBN: 978-989-758-746-7; ISSN: 2184-5026

This paper is a case study of a long-term design

process in a real world setting involving several hun-

dreds of end users (teachers and pupils). It explains

the rationale behind various design iterations carried

out over several years of development. We believe

that our approach highlights key considerations rele-

vant to designing and testing learning games for chil-

dren. The main research questions addressed here are:

1. How to design an experimental protocol to test a

SG ”in the wild”? And what are its consequences

in the design of the application?

2. How to identify areas for improvement in an ap-

plication that already demonstrated positive out-

comes in terms of knowledge acquisition?

3. How to effectively engage young children in the

design process, considering their limited familiar-

ity both with tablet use and with SG?

To tackle these questions, we propose a “design

story”. It begins with a scientiﬁc context closely re-

lated to 1, followed by its inﬂuence on the ﬁrst ver-

sions. Then it tells the part played by user-centered

design methodology for education in the subsequent

iterations and addresses research questions 2 & 3.

Throughout this process, we extract design require-

ments (DR), identiﬁed [DR

∗

], that drove this design

and can be applied to other projects.

2 DESIGNING A LANGUAGE

LEARNING GAME FOR 6–8

YEAR-OLDS

Luciole was developed over the course of 8 years

within different projects and fundings. Its integration

into those projects represented a set of both oppor-

tunities and constraints, which impacted greatly the

design process of the game.

Since 2016, several versions of Luciole have been

developed. Each version has followed an iterative cy-

cle: 1. design (within the team or integrating stake-

holders in collaborative design sessions); 2. develop-

ment; 3. user tests; 4. data collection and analysis.

Each cycle builds on the previous iteration, inte-

grating users’ feedback from the previous cycle. The

data collection methods can vary from one iteration

to the other according to its objectives, and the time

and design constraints. In this article, we present the

various methods for gathering needs, developing and

testing prototypes, and analyzing results used over the

project. Our main research questions for this paper

being methodological, we describe more extensively

the processes linked with the experimental protocol,

the identiﬁcation of areas for improvement and with

the integration of young children in the process.

2.1 On the Design Constraints of

Experimental Design

Luciole ﬁnds its origin in the Fluence project (2017-

2022). It aimed to design and validate digital tools

supporting learning (Mandin et al., 2021). Three

applications were developed, two targeting reading

skills (EVAsion & Elargir) and Luciole supporting

English as a Foreign Language (EFL). To validate the

applications, we needed to prove that each one im-

proved the skill-set it targeted. To do so, the longi-

tudinal protocol (Mandin et al., 2021; Loiseau et al.,

2024) integrated 700+ pupils of 37 schools (ﬁg. 1).

They were separated in groups of consistent distribu-

tion of rural vs. urban schools, socially advantaged

vs. disadvantaged schools. To ensure that control and

experimental groups were comparable, they were cre-

ated with a consistent distribution of pretest scores.

2018

CP (6 y/o) CE1 (7 y/o) CE2 (8 y/o)

2019 2020

EVASION

Luciole

Pretest Posttest Posttest

Luciole

ELARGIR

EVAsion & ELARGIR

point of view

Test groups

Control groups

Posttest

Luciole

point of view

Control groups

Test groups

ELARGIR

Cohort:

52 classes

37 schools

736 pupils

pupils

731

pupils

340

pupils

158

pupils

131

pupils

133

pupils

339

pupils

137

Covid-19 outbreak

Figure 1: Fluence experimental design.

In a “design” paper, it is ﬁtting to detail the im-

plications of the experimental design. The tools

were paired to neutralize the Hawthorne effect, which

states: “those who perceive themselves as members

of an experimental or otherwise favored group tend

to outperform control groups, even in the absence of

applied variables” (Koch et al., 2018, p. 3). In other

words, the test group for each application is the con-

trol group for the other one. This creates constraints:

[DR

] Skill-sets targeted should not overlap across

tested applications. To fulﬁll this Design Require-

ment (DR), there needs to be deep understanding of

the skill-set targeted by the other team(s).

In our case, the core hypothesis behind EVAsion

(a reading app) is that “optimal intervention for Visual

Attention Span (VAS) enhancement should be based

on the properties of action video games while requir-

ing parallel processing of targets that gradually in-

creased in number of visual elements” (Valdois et al.,

2024, p. 3). In our case, [DR

] translated into “the

game should not use action video game structures

Luciole: The Design Story of a Serious Game for Learning English Created for (and with) Children

381

(Green et al., 2010, p. 203) or integrate written text”.

The neutralization of the Hawthorne effect can re-

sult in situations where, in addition to one’s own re-

search questions, the protocol might apply constraints

on the nature of the tools one can design.

2.2 On the Design of Serious Games

The ﬁrst iteration began with a design phase that re-

sulted in a product backlog, which allowed to re-

cruit developers, designers and graphic artists

. The

subsequent development phase was carried out using

Scrum (Kniberg, 2007), allowing design tasks to oc-

cur during development (ﬁg. 2).

2.2.1 On the Notion of Serious Game (SG)

The choice of creating a SG was directed by the prop-

erties attributed to them in literature: improve learner

motivation (Garris et al., 2002; Reinhardt and Thorne,

2019), promote active learning (Vlachopoulos and

Makri, 2017, p. 26), provide feedback on actions and

assessment of player skills (Oblinger, 2004, p. 14), al-

ter the perception of one’s errors (Loiseau and No

us,

2022, p. 69), not to mention take advantage of games

design patterns which are adapted to learning (Gee,

2003). We deﬁne a serious game as “a game in the

sense of (Duﬂo, 1997) designed, prescribed or used

with both the aim of entertaining and an external ﬁ-

nality targeting the player” (Loiseau and No

us, 2022,

p. 76). We will not engage in what Duﬂo means by

game, by quoting this deﬁnition we want to separate

from gamiﬁcation which does not aim at producing

a full-ﬂedged game (Seaborn and Fels, 2015, pp. 14,

16, 27) but also recognize that we might fail in that en-

deavor and create an object which is not perceived as

a game by the user. As a consequence, a DR emerges:

[DR

] the game must be perceived as a game by its

players. We will see later on that it is not a tautology.

2.2.2 Narration in the Game

Among the design elements that are widespread in

video games, the integration of a narrative (Domsch,

2013) was essential in our case but also constrained

by DRs. [DR

] A language learning game for 6-8

y/o should contain (almost) no written text is, in

our case, highly consistent with [DR

], as the other

two Fluence applications target reading. Rather than

being a consequence of [DR

], [DR

] stems from var-

ious factors. 6 year-olds are not yet readers; any writ-

ten text might be understood extremely slowly if un-

derstood, thus impairing the engagement in the game.

Over the project, we collaborated with two companies

and 4 independent graphic artists and developers.

Second Language Acquisition (SLA) theory and of-

ﬁcial instructions also highlight this DR. The cur-

riculum for our age target (6-8 year olds) makes oral

language a priority (MEN, 2015, p. 31). Listening

skills in particular, according to Krashen, constitute

the foundation upon which other language skills are

built. The aim of the second language classroom is

to provide “comprehensible input” (Krashen, 1982,

chap. 3), that is to say, to expose learners to utterances

in their Second language (L2) that they can process.

[DR

] should thus be completed by [DR

] A language

learning game for 6-8 y/o should be the source of as

much L2 input as possible. [DR

] is also consistent

with the situation of certain teachers we evoked in in-

troduction. Since input is one of the building blocks

of later language skills, a game introducing learners

to EFL should provide opportunity for input by native

speakers, thus discharging the teacher from some of

the responsibility.

The narrative of the game addresses (at least in

part) both DRs. The narrative integrates tasks in a co-

herent plot, but is also used to explain game struc-

tures, to introduce cultural elements about English

speaking countries (MEN, 2015, p. 29) and to provide

meta-linguistic information.

These DRs are at the core of the ﬁrst versions of

Luciole (cf. § 3), whose time frame did not allow us to

integrate users in the original design. The subsequent

design phases attempted to close that gap using user-

centered design methodology.

2.3 User-Centered Design Methodology

for Education

Involving end users in the development of learning

environments has raised increasing interest due to

its demonstrated beneﬁts in enhancing usability, user

experience, system adoption, and user engagement.

Various methods and tools derived from Human-

Computer Interaction (HCI) facilitate the integration

of users in the design, implementation, and evaluation

processes of interactive technologies. Because of de-

velopmental differences between children and adults,

researchers have started to explore speciﬁc methods

for the participation of children in design process,

called Child-Computer Interaction (CCI). In their

systematic literature review, Tsvyatkova and Storni

identify methods, techniques and tools from User-

Centered Design (Norman and Draper, 1986), partic-

ipatory design (Simonsen and Robertson, 2013) and

learner-centered design (Good and Robertson, 2006)

developed or adapted to support children’s involve-

ment in design (Tsvyatkova and Storni, 2019). Their

review indicates that most methods primarily support

CSEDU 2025 - 17th International Conference on Computer Supported Education

382

design exploration and prototype evaluation, with a

predominant focus on children aged 7 to 12. Proto-

typing methods are less frequently employed because

they might be too difﬁcult for children. Saiger et al.

reached a similar conclusion in their review on chil-

dren’s involvement in game design, ﬁnding that chil-

dren participated as true design partners in only half

of the studies (Saiger et al., 2023). Five key fac-

tors were identiﬁed as inﬂuencing the effectiveness

of children’s involvement: comprehension, cohesion,

conﬁdence, accessibility, and time constraints. Most

of the existing methods and tools focus on the evalua-

tion of mock-ups or prototypes, especially with young

children, to understand how they interact with the sys-

tem and to identify their emotional responses. These

tools are based on interviewing techniques — This-or-

That (Zaman and Vanden Abeele, 2007), Contextual

Laddering (Zaman and Vanden Abeele, 2010) — or

propose instruments — Fun Toolkit (Read, 2008) —

to identify children’s engagement, likes or dislikes

when interacting with products.

2.4 A Design Story

Luciole was created to respond to a need. Its de-

sign process was iterative and based on speciﬁc de-

sign requirements linked to its object and scientiﬁc

context. We brieﬂy evoked its main design methods,

but an 8-year design process (Fig. 2) cannot be homo-

geneously summarized. In the next sections, we go

through the different versions of the application.

Fig. 2 gives a color to each version of Luciole,

showing its design and development phases as well as

the user tests conducted and the analysis period. This

ﬁgure illustrates the design cycle, both as a sequence

of steps and in terms of how the previous phases pro-

vide input for the next. It also presents the various

milestones to contextualize v5, for which we describe

extensively the process which tackles research ques-

tions 2 and 3.

3 FLUENCE VERSIONS

3.1 v1: The Core

3.1.1 Game Structures

The design period of v1 corresponds to the deﬁnition

of the core DRs and the foundations for all subse-

quent versions (cf. § 2). Once they were identiﬁed,

we started to establish the game’s curriculum. We

separated the use of constructions

in three phases:

1. introduction of new constructions; 2. training on

said constructions, used in simple utterances; 3. con-

textualization of trained constructions, used in more

complex utterances, sometimes mixing constructions

from diverse activities.

Despite its critiques, we adapted the Presentation-

Practice-Production (PPP) model for its reliability for

lower level learners (Anderson, 2016). Integrating

these phases in the narrative aims to make every learn-

ing activity meaningful and targets the realization of

an action oriented task. It thus conﬁrms potentialities

of games in language learning (Cornillie et al., 2012).

The narrative places the learner in the shoes of

Sasha, a French child recruited by an intelligence

agency to save animals. This choice owes to the spe-

cial relationships children can develop with pets (Cas-

sels et al., 2017). Sasha’s missions take him to En-

glish speaking countries. He is helped by his men-

tor, Ash, who speaks French with a heavy English ac-

cent (to help the learner discover the sounds of En-

glish, and also to help focus on communicating and

to play down possible production errors) and Hartguy,

the solely English speaking coach.

To address [DR

] through diversity in the player

activity, we isolated 3 interaction matrices:

• identify & touch: The system provides an au-

dio stimulus and the player needs to identify the

sprite corresponding to the answer and touch it. It

is mostly used in training and contextualization

mini-games where one sprite is the correct answer

for a given cue. In v1, Introduction phases al-

ways used an adaptation of this mini-game: play-

ers need to ﬁnd the relevant sprites to hear the new

vocabulary;

• drag ’n drop: The system provides an audio stim-

ulus prompting the player to identify a draggable

sprite among others and drop it on an associated

drop zone. It is also used both in training and con-

textualization mini-games;

• remember: based on the metaludic rules (Silva

Ochoa, 1999, p. 277) of memory games such as

“Simon” or “Touch Me”

, it is used for vocabu-

lary memorization, and spans across introduction

and practice.

We used these matrices to create 49 activities in v1.

To provide feedback on the learner’s progression in

Strictly speaking constructions are “stored pairings of

form and function, including morphemes, words, idioms,

partially lexically ﬁlled and fully general linguistic pat-

terns” (Goldberg, 2003, p. 219). In this article, we also add

phonemes though they do not have a function and should

not be considered as such.

https://en.wikipedia.org/wiki/Simon (game)

Luciole: The Design Story of a Serious Game for Learning English Created for (and with) Children

383

Fluence

Trans3

Design/Development

User Tests

700 pupils 10 weeks

Experimentation

500 pupils 10 weeks

Experimentation

800 pupils 10 weeks

Experimentation

field survey

 11 interviews 18 pupils

Workshop 1

11 interviews 3 teachers

Workshop 2

interactive tasks 9 pupils

Workshop 3

13 pupils 4 weeks

User test

10 weeks

User test

interrupted COVID

Experimentation

Brainstorming

Scientific research

Benchmark

Pluridisciplinary team

Collaborative design

Collaborative design Collaborative design

Collaborative design

Ecrimo/

LuCoCoph

550 pupils 10 weeks

Experimentation

Analysis

Type of Data Analysis

Qualititative

Mixed

Quantitative

v4b

Alternative authentication

Search ﬁnd & touch + new dashboard

+ trophies + ranking/stars

Missions 0-4

Missions 5-8 + QR code

+ Ranking

2 versions (Normal vs NoPhono)

Traces improvement

“Remember” update

Quadrichromino

Missions 9-10 + Changing room

2017 2018

2019 2020

2021 2023 2024 2025

Figure 2: Luciole design story timeline (detailed version available online).

(a) The Map in v5 adds a geographical map.

(b) The notebook contains all the vocab learned.

Figure 3: Screenshots of the Luciole serious game.

the game we use a map: each activity is represented

by a logo on the map and is associated with a score

(0 to 3 stars) (Fig. 3a). Each activity unlocks the next

one. Learners can replay activities from the map or

the training center, where they are sorted by theme.

To help learners keep track of what they have learned,

we introduced a notebook displaying all the words

they worked on, again grouped by theme (Fig. 3b).

3.1.2 Results

The game was tested for 10 weeks (prescription: 3×20

min/week). We collected pre/post-tests for 679 CP

pupils (cf. ﬁg. 1). We retrieved interaction traces for

310 out of 340 Luciole users and teacher feedback.

Global results were positive: Luciole groups outper-

formed control group in the oral English comprehen-

sion tasks

(Mandin et al., 2021; Loiseau et al., 2024).

Teachers warned us that remember games dis-

couraged pupils. It also appeared that some children

advanced too fast in the game and then were lost for

lack of mastery of previous content.

3.2 v2 & v3: Improving Reﬂexivity and

Expanding the Game

3.2.1 Design Choices

Both qualitative feedback demonstrated lackluster re-

alization of [DR

]: such difﬁculty can disengage

learners. We iterated over remember matrix. The sec-

ond remark pointed to [DR

] Advancing to the next

mini-game should neither be too punitive nor too

permissive. The quest for such balance led us to cre-

ate a global score

. To motivate users, we associated

milestones with a “spy rank” and created checkpoints

that prevented players from accessing an activity un-

less they had reached a given rank.

A new interaction matrix was created to improve

immersion in the game using non-digital resources

(Loiseau et al., 2021). In the QR code mini-game,

players scan one or more QR codes in response to oral

instructions. QR codes are on a map of the British

Isles. It concretizes Sasha’s trip and diversiﬁes game

structures. At the end of each mission (some 10 activ-

ities), a new suspect is interrogated and the next sus-

English score: Luciole = 11.18, SD = 3.5;

EVAsion = 8.41, SD = 3.33;

t = −10.59, d f = 677, p < 10

−23

Score

global

∑

A=1

max(Score

activity

(A)) with A the

activity, 1 the 1st activity in the progression and n the fur-

thest activity reached by the player.

CSEDU 2025 - 17th International Conference on Computer Supported Education

384

pect is identiﬁed by scanning QR codes on the poster.

v2 contained 94 mini-games.

v1 provided only usage information. In v2, we

[DR

] use the game to gather quantitative infor-

mation from the learner’s perception of the game

and their learning. The narrative allows for self-

assessment questionnaires every four missions. Sasha

goes home with his mother who asks questions about

his adventures. These questions constitute the self-

assessment.

3.2.2 Results

The cohort of v1 was split in 4 subgroups. We col-

lected pre/post-tests for 559 CE1 pupils distributed in

4 subgroups (Fig. 1). We successfully retrieved the

interaction traces (including self-assessment) for 405

Luciole players and again collected teacher feedback.

The overall results were also positive

(Mandin et al.,

2021; Loiseau et al., 2024). Yet, the group which used

Luciole two years in a row did not fare better than the

other Luciole groups. This was later explained by a

combination of rolling releases on our part during the

experiment and of a lack of updates on the part of the

teachers (Loiseau et al., 2024).

The trace system was not immune to bugs, some

of the self-assessment was lost. But we could process

the perceived level of mastery across lexical themes

(not detailed here). In terms of user experience, the

responses that stood out were the following: 62.7%

reported understanding the story. 53% found the

game easy, 43.7% found it moderate (sometimes easy,

sometimes difﬁcult), and 3.3% found it very difﬁcult.

Lastly, all pupils gave the maximum score to their

overall appreciation of Luciole.

According to teachers, despite some improve-

ment, remember still needed some work.

3.2.3 v3

The tests of v3 were rapidly interrupted due to the

covid-19 outbreak. It provides a new mini-game

meant to easily extend the game with utterance-image

pairs. It is included as a bonus to players who com-

plete the whole story.

3.2.4 After the Fluence Project

An unexpected result of the Fluence project was that

the v1 Luciole group fared better in phonological

awareness in their First language (L1) (French) than

English score: Luciole − ∗ = 10.45, SD = 4.3;

EVAsion − Elargir = 8.21, SD = 3.77;

t = −6.1, d f = 321, p < 10

−8

the EVAsion group

(Charles et al., 2025). Phono-

logical awareness was included in the tests as a pre-

dictor of reading/writing skills (Valdois et al., 2024).

At the end of the Fluence project, we tried to in-

vestigate the matter with a joint project (called EC-

RIMO2/LuCOCoPh, see Acknowledgements). In or-

der to do so, we replicated the Fluence protocol, using

ECRIMO (a dictation app) for our control group. Al-

though we replicated the results in terms of English

skills

, phonological awareness results were disap-

pointing, which we could attribute to the control task

and our lack of means for individual testing of the

pupils (Charles et al., 2025).

v4 was also an opportunity to improve Luciole’s

trace collection system (cf. § 3.2.2) and redesign the

remember after v2 teacher feedback.This project was

also an opportunity to reﬂect on our design processes

and the need to implicate the users more deeply in the

design phases to accelerate game improvements.

4 INVOLVING CHILDREN

The end of our multiple experimentations also created

time to look more deeply into the game interaction

traces. After four iterations, the most undesirable be-

havior highlighted was the lack of mini-game replay.

Failure or limited star gain did not change that behav-

ior. For example, in the remember game, the average

scores range from 0.07 to 1.19 stars (out of 3). Yet,

they are not replayed to gain more stars. The “train-

ing center” (meant to identify activities where score

improvement was most likely) was underused: it was

displayed an average 15 times per pupil. This lack

of replay can be detrimental to acquisition. Hence,

[DR

] The game should encourage players to replay

mini-games in which they did not fare well.

[DR

] concerns the understanding of speciﬁc fea-

tures (and global user experience). For these reasons,

we decided to adopt a user-centered approach to in-

volve children for the next design iteration. As under-

lined in section 2.3, methods to involve children are

directed toward prototype evaluation. Using proven

tools is especially relevant because children tend to

seek to please adults during design workshops, ex-

press few spontaneous contradictions, and have dif-

ﬁculty making and expressing clear decisions. They

can also be inﬂuenced by group effects, with a leader

Wilcoxon rank-sum tests for Phoneme deletion task:

W = 55280, p < .005 & Phoneme segmentation: W =

67044, p < .0005

Mann-Whitney U test Luciole

f ull

vs. ECRIMO: U =

17015, p < 10

−4

; Luciole

NoPhono

vs. ECRIMO: U =

19123, p < 0.005

Luciole: The Design Story of a Serious Game for Learning English Created for (and with) Children

385

sometimes monopolizing the ﬂoor. This section de-

tails a methodology to work with 6 to 8 year-olds

through 4 interventions with pupils and teachers. Tes-

timonies in this section were translated from French.

4.1 Workshop 1

From March to June 2023, a user test of v4b (v4 with

alternative authentication) was carried out (2 sessions

a week). In June, a ﬁeld survey was carried out with

18 elementary school pupils (CP) from three differ-

ent classes. They were asked about their use of Lu-

ciole, and to demonstrate in-game access to certain

features. These one-on-one interviews helped better

understand some concerns (e.g. lack of replayability).

The main obstacles highlighted were: most chil-

dren are unable to show their progress in the game

(number of stars earned, secret agent rank); they are

unable to explain in their own words the notion of

stars; more critically, they do not understand how to

access certain features (the “training center”, or the

ways to see their progression: the map to replay mini-

games — ﬁg. 3a — or the notebook — ﬁg.3b).

The positive aspects identiﬁed were linked to key

game elements: the storyline (“it’s good to free the

animals”, “I love when we go on adventures”); re-

wards (“I like winning medals”); learning new words

related to the activities (“learning colors”, “dragging

clothes”); or the visual environment (“the images and

sounds are beautiful”).

Less appreciated aspects included activities per-

ceived as too challenging, such as the remember

game, and instances of lengthiness in dialogues (“I

don’t like when they chat for too long”). The vast ma-

jority of children were able to recount the story, or

at least its main elements (characters, setting, goal).

They were also able to spontaneously and instantly

provide between one and a dozen English words (pri-

marily related to animals, numbers, and colors —

consistently with in-game auto-evaluations).

4.2 Workshop 2

In parallel, we carried out semi-structured interviews

with a group of three teachers from the same user test.

They were unanimously positive about their experi-

ence. They posited that children learned new words

and both understood and enjoyed the story. They re-

port general enthusiasm at the start of each play ses-

sion, and a sort of bond with the main characters.

They observed that more advanced pupils some-

times went to help those experiencing more difﬁcul-

ties. The main concern was autonomy: some chil-

dren had difﬁculty understanding and applying in-

structions during certain mini-games, and they were

unable to complete certain tasks without teacher inter-

vention. Then they did not progress in the game sce-

nario, which led to frustration and discouragement.

Teachers conﬁrmed the children’s lack of under-

standing of some features: rank progression, time

management, the means (and interest) of replaying

mini-games. They also reported that the remember

mini-games were not appreciated by all pupils. Some

considered them too hard and insufﬁciently reward-

ing. Although traces analysis revealed a high level of

use of the notebook (ﬁg. 3b), the interviewed teach-

ers expressed a strong desire to encourage students to

listen more times to words in the game, but also to

use these words in group activities outside the game

(they were enthusiastic about our suggestion of a non-

digital sticker album, for example).

4.3 Workshop 3

Based on outcomes of previous iterations, conﬁrmed

by workshops 1 & 2, one of our main goals was [DR

]

(replayability). To achieve this, we undertook a com-

plete overhaul of the dashboard and reward system

(stars and ranks). The objective is to make it eas-

ier for pupils to visualize their progress and increase

their motivation to replay mini-games in which they

underachieved. In a user-centered approach, we de-

cided to involve children in the design process, rather

than depend on their feedback like in workshops 1 and

2. As we have explained in introduction to section 4,

involving such young children is not straightforward,

we thus resorted to the tools introduced in section 2.3.

4.3.1 Participants

In June 2023, 9 French primary school students par-

ticipated in a workshop: 2 CP and 7 CE1 (5 girls and

4 boys). This group was part of a mixed level rural

class. All participants had played Luciole during the

user test. Children were divided into three groups.

Each task was carried out with all three groups. For

each group, two adults were involved.

4.3.2 Protocol

The method employed was inspired by the laddering

technique, based on the Means-End theory (Zaman

and Vanden Abeele, 2010). The name “laddering”

comes from the idea of climbing a ladder, where each

question serves as a rung leading to deeper insights.

This systematic probing aims to achieve more com-

prehensive understanding of the subjects’ attitudes,

motivations, and decision-making processes. This

CSEDU 2025 - 17th International Conference on Computer Supported Education

386

technique is valuable because it helps uncover moti-

vations that individuals may not be fully aware of or

able to articulate autonomously.We used the laddering

method by following this presentation schema: First,

we start with a simple question related to the chil-

dren’s experience with the game “What do you like

about playing with Luciole?”; then we explore their

motivations by asking “Why?” to uncover the moti-

vations behind their preferences. For instance, if a

child mentions enjoying the game because “it’s fun”,

we ask, “Why is it fun for you?”

It is important to frame questions in a way chil-

dren this age can understand and relate to. Instead

of asking questions about values, we focused on con-

crete aspects like favorite characters, game features,

or challenges they enjoy. Moreover, children often ex-

press themselves through play and imagination. We

have to encourage them to expand on their answers

using their creativity. Understanding what motivates

and engages pupils allows for targeted improvements

that enhance their learning experience.

There were two experimenters per group of chil-

dren: the “interactor” who engages with the children

(e.g., asks questions, organizes speaking turns, etc.),

and the “note-taker” who records as much informa-

tion as possible, especially non-verbal behavior (au-

dio was recorded), and assists in workshop set-up.

Experimenters had general instructions: Simple

attire recommended (avoid intimidating clothing such

as suits or ties); Allow children to get accustomed to

the presence of adults (icebreaker), and avoid appear-

ing rushed; Adopt an attitude of ignorance to stim-

ulate the involvement and sharing of information by

the child. Position themselves in a rapport if possible

equivalent by mentioning ignorance on certain points,

which they can respond to. Moreover, it is advisable

to adopt the least adult-like position possible (Zaman

and Vanden Abeele, 2010, p. 160).

4.4 Tasks (Fig. 4)

The workshop was split in 5 stages with a 15-minute

break between the second and third task.

Icebreaker and Recall of Luciole’s Story. The

workshop starts with a presentation involving all par-

ticipants and outlining its objectives and tasks. A Lu-

ciole video is shown to introduce the game and help

children recall it.The goal is to create a common set-

ting and make it clear to the pupils that they partici-

pate to the same tasks no matter which subgroup they

are. Pupils are encouraged to engage in spontaneous

remarks both to engage them, notice we listen to them

Can you help us improve

Luciole?

Non-digital material

Interactive material

Task 1

Star system

Task 2

Ranking systems

Task 3

Sub-ranks

Icebreaker

Task 4

Visualization of

rank progression

Non-digital material

Figure 4: Workshop 3 task plan (Luciole v5).

but also to gauge their enthusiasm and to identify po-

tential leaders who speak up frequently.

Task 1 — Star System. The aim is to improve the

pupils’ understanding of the stars they earn at the end

of each mini-game. We created both paper visuals

and interactive versions to chose between vertical and

horizontal representation of the stars. We start by re-

minding pupils of the “this-or-that” instructions, then

show the paper visuals (to avoid children rushing to

the tablets — untimely clicks, ignoring instructions),

while reminding them of how stars are earned in Luci-

ole. After freely testing the two options on the tablet,

the children indicate on their slate the number of the

proposition they prefer. Their choice is the basis for

a discussion where they explain their choices individ-

ually and collectively. The experimenter makes sure

that all the children participate.

Task 2 — Ranks. In v4, children reported not

knowing their “spy rank”. The rank is supposed to

motivate children and provide an indicator both of the

progression in the game and of the level of mastery.

In this task, the aim is to choose which of the four

proposed animal rankings is more appealing, make

sure children recognize each animal, and adjust the

animal order within the ranking. Children write their

preferences on their slates, then comment on each an-

imal. The experimenter systematically asks questions

to make children justify their choices (Why?).

Task 3 — Sub-Ranks. We created sub-ranks for

each animal: it generates more opportunities to get a

reward, to hear the animal name, and optimizes visu-

alization on the tablet. Animals evolve from baby, to

adult, to superhero. The task targets the children’s un-

derstanding and acceptance of this progression. They

sorted the stages of this evolution and justiﬁed their

choices (Why?). Particular attention was paid to the

“baby” rank: we asked each child whether they would

accept being depicted as a baby lion wearing a diaper.

Luciole: The Design Story of a Serious Game for Learning English Created for (and with) Children

387

Task 4 — Visualization of Rank Progression. Pre-

vious tasks allow to gauge interest for various rank

scales, but to address children not understanding that

indicator, we need the representation of the scale to

be clear. We created two mock-ups on the tablet, one

vertical and one horizontal representation of the rank-

ing system. Children could manipulate both modes

on the tablet. Then each indicated their preferred vi-

sualization on their slate. The experimenter then had

them explain their choices (Why?).

Material. Various material were necessary for this

workshop. They were designed and created speciﬁ-

cally for this study, but could be adapted and used for

another study (Table 1).

Table 1: Materials used for the co-conception workshop.

Physical Material

Animal grade image cards proposed (32 cards) (4×8)

Ranking card track (divided into 2 A4 sheets) ×4

Selection number allowing to associate a number with a ranking or a prototype. A4 sheet divided into 4 or 1 sheet per number.

Adjective image cards associated with intra-grade progression (Lion — 3 cards Baby/Normal/Super)

Vertical example star presentation and horizontal to guide the child during this or that ×1

Vertical example gauge presentation and horizontal to guide the child during this or that ×1

Example sheet of video games using a star system

Numerical Material

Progress bar with grade scale (vertical/horizontal)

Star activity progress bar

Base - without star (i.e., to explain the activity)

Other Supports

Luciole video

Slates × 5

Whiteboard markers × 8

Camera or smartphone

Microphone and recording equipment (e.g., computer)

4.4.1 Results

Data collected came from audio recordings, verbatim,

pictures (slates) and real-time completion of observa-

tion grids by adult observers.

Task 1. The consensus was in favor of the verti-

cal view of the stars. The children’s arguments were:

climbing is vertical (“we climb the stairs to go even

higher”); positive evolution goes up; more aesthetic

and easier to understand. The horizontal view was

considered unattractive and could hinder the comple-

tion of an activity for taking up more space on the

screen. The children also indicated that they wanted

star ﬁlling to be progressive (gauge effect).

Task 2. The order of preference for the animal

rankings was: 1. marine animals (3 votes and 2

groups reaching consensus), 2. birds (2 votes and 1

group reaching consensus), 3. food chain (2 votes),

4. “cool” animals did not receive votes. The two fa-

vorite marine animals were, with two votes each, the

swordﬁsh and the dolphin. Numerous lively discus-

sions led to adjustments, both of the ranking order

(“the octopus is stronger than the crab”) and to the

visual appearance of certain animals (“the dolphin is

too big”). Grades not yet unlocked should be visi-

ble to whet players’ appetite, but grayed out and pad-

locked to show that they are not yet earned.

Task 3. The sub-ranks were understood and unan-

imously approved (no reservations about the “baby”

stage).

Task 4. Consensus was quickly reached on the hor-

izontal view described as clearer and easier to under-

stand, but many questions arose about the space it

would take up on the screen.

4.4.2 Subsequent Development

The workshop was quickly followed by various de-

velopments : New graphics representing the marine

animals: shrimp, seahorse, crab, octopus, seal, dol-

phin, swordﬁsh, orca all declined in three sub-ranks

each, with visual reinforcement of the notion of rank

(medal) with the addition of a ribbon; Padlocking of

not yet reached ranks; possibility to hear both locked

and unlocked ranks; horizontal presentation of ranks,

the sub-ranks are displayed vertically when the rank

is touched.

4.5 Prototype Validation by Pupils

Traces analysis had shown that in introduction

phases, children did not actually listen to the sounds

they were supposed to become familiar with. They

just touched the screen frenetically until all objec-

tives were uncovered. We created a new mini-game

(coined Search, ﬁnd and touch), so that they could

only hear one new item at a time to hear it clearly.

After developing the new ranking system and

mini-game, we tested the new prototype with children

who had never played the previous versions of Luci-

ole. This study consists of three interventions, each

one week apart. During each intervention, observers

watch children participate in a Luciole game session,

and then gather feedback and suggestions.

4.6 Outline of Interventions

Introduction and Game Overview. At the begin-

ning of the ﬁrst intervention, team members introduce

themselves, present the Luciole project, and give a

hint of the story-line, the characters and explain how

to launch the game (account system).

Game Session and Observation. During each inter-

vention, children play Luciole for 20 minutes, each

time resuming their progression. One observer con-

ducts shadowing observations (Czarniawska, 2007)

based on a speciﬁc observational grid.

CSEDU 2025 - 17th International Conference on Computer Supported Education

388

Qualitative Feedback. At the end of each 20-minute

session, researchers conduct a semi-structured discus-

sion with the children, teachers to gather feedback and

suggestions for improvement.

Feature Testing Workshops. In the ﬁnal two inter-

ventions, speciﬁc workshops are organized to test cer-

tain game features with small groups of pupils (N=4),

aiming to collect targeted feedback.

Analysis and Team Member Brainstorming. After

each intervention, the observations and workshop re-

sults are analyzed to identify and reﬁne improvement

areas for the next session.

Autonomous Session of Luciole in Class. Between

each intervention, children continue their progress

with their teacher during two autonomous in-class

sessions.

Final Evaluation. The third intervention ends with

pupils completing the System Usability Scale (SUS)

(Vlachogianni and Tselios, 2022).

4.7 Results

Observation sessions and analysis of the verbatim

showed a high level of understanding of the new play-

ing instructions and rankings. Pupils unanimously ap-

proved the marine animal badges. They were proud

to compare their progress with each other and share

it with the teacher. The summary of comments made

during the test workshops did not reveal any partic-

ular difﬁculties for pupils with the new Search, ﬁnd

and touch mini-game.

The mean score on the SUS questionnaire was 88.7

(standard deviation: 7.2), indicating excellent accept-

ability of the game (even with young subjects). The

items with the highest scores concerned handling of

the game, which was rated as easy to use, light and

requiring no external assistance. Interest was also

shown in using the game more frequently. Analysis of

the 13 pupils’ use of the Luciole game (with access to

information such as progress, number of mini-games

played, stars won, notes consulted, etc.) revealed that

the time spent playing the game was as prescribed,

and that the game’s features were widely explored.

The progress made by the pupils was remarkable.

5 DISCUSSION

This paper tells the story of the design of a serious

game over eight years. Its design has been driven

by DRs that either emanated from the learning ob-

jectives for the players or the experimental protocol

itself. While the ﬁrst source of constraints is com-

mon the second raises more questions. Should the

design of a technological object be driven by how the

object will be tested? In our case, the constraints of

the protocol were in line with our scientiﬁc and learn-

ing objectives and the decision to join the project was

straightforward. But this should not downplay this is-

sue. Serious games have been pointed out for lack

of longitudinal and/or empirical studies (Girard et al.,

2013) and the neutralization of variables such as the

Hawthorne effect come with constraints that should

not be overlooked. In the v4 tests, the choice of the

control group might be the cause for some disappoint-

ing results (Section 3.2.4).

Working within this set of constraints, the regular

tests in ecological conditions allowed us to gather di-

verse data on the use of our game. We had regular

interactions with teachers who signaled bugs and im-

pressions on the use of the system, gathered interac-

tion traces, collected observations, questionnaires and

post-test data. This wide array of data both in nature

and quantity allowed us to continuously improve our

application. Qualitative information (for instance the

class observation that children did not seem to pay at-

tention in introduction phases of versions 1 through

4) could be veriﬁed thanks to quantitative data (traces

analysis). In that, we did not only mobilize usage data

to qualify the behavior of the learners but also to iden-

tify issues and improve the design.

Still, the fact that the game was created all along

the Fluence project put pressure on the development

process and let us settle in a rhythm of gathering infor-

mation at the end of the tests (one iteration per year),

instead of implicating the pupils and teachers directly.

This rhythm showed its efﬁciency limits in the num-

ber of iterations necessary to ﬁx our remember mini-

game (cf. [DR

]).

On the contrary, before the Trans3 project we had

time to analyze our data and organize design work-

shops with actual end-users. Those proved very effec-

tive in the redesign of the visual research game or the

ranking systems. To set-up these workshops we used

design tools that were adapted to children (see § 4).

Yet, we underestimated the length and frequency of

digressions within those workshops. In turn, the time

allotted to digressions made workshop longer than

expected, and exhausted the children. Digressions

should not be avoided, they participate in creating a

trust environment, an atmosphere where children can

really contribute. But they should be accounted for in

the length of the design workshops.

We should also mention that shadowing is some-

times difﬁcult to understand for a 6 year-old child who

wants to advance in the scenario and asks for help

(such interactions or lack thereof were very produc-

tive in the discussion part of the workshop).

Luciole: The Design Story of a Serious Game for Learning English Created for (and with) Children

389

Finally, it should be noted that the design tools we

chose might not be adapted for other life cycles. Had

we implicated the children when we created the game

from scratch, we probably would not have relied on

evaluation tools and prototype version choices. But

the improvement of a tested and validated system was

in line with our objectives.

6 CONCLUSION

Over the course of its design and development, Lu-

ciole has been tested on three cohorts of hundreds

of children (twice on one cohort). In each experi-

ment, the Luciole groups fared better than their coun-

terparts regarding EFL. Each experiment took place

under business as usual conditions, each group be-

ing given an application targeting separate skills to

neutralize the Hawthorne effect. In one occasion, it

also proved to have an effect not only on English but

also on phonological awareness. Over the course of

eight years of work we have made our design prac-

tices evolve to use various source of data.

Now that Luciole has reached a form of stabil-

ity (an almost “ﬁnal” version should be published in

September 2025), many venues are open for experi-

menting. The inﬂuence of the context of use should

be analyzed. Many research questions can be iden-

tiﬁed: how the learning context (school vs. formal

out-of-school learning vs. informal learning) inﬂu-

ences Luciole’s outcomes; what in-class group activi-

ties to carry out to maximize Luciole’s effect; or how

and which non-digital material (and rewards) can in-

ﬂuence children’s engagement in the game, and their

motivation for foreign languages, or for school.

ACKNOWLEDGEMENTS

The authors are grateful to the French government

(France 2030) for its ﬁnancial support through the

Trans3 project (ANR-22-FRAN-0008), managed by

the Agence Nationale de la Recherche (ANR). This

work was also supported by

• the e-FRAN program (“e-FRAN Fluence”

project, 2017–2022), funded by the “Programme

d’Investissement d’Avenir” (PIA2) handled by

the Caisse des D

ots et Consignations (CDC);

• the ASLAN project (ANR-10-LABX-0081) of the

Univ. de Lyon, for its ﬁnancial support within the

French program “Investments for the Future” op-

erated by the ANR: funding of LuCOCoPh.

• the Pole Pilote P

egase, French government funded

“Territoires d’innovation p

edagogique” of the “In-

vestments for the Future” program, operated by

the CDC: funding of the ECRIMO2 project.

REFERENCES

Anderson, J. (2016). Why practice makes perfect

sense: the past, present and future potential of the

PPP paradigm in language teacher education. En-

glish Language Teaching Education and Development

(ELTED), 19(1):14–22.

Cassels, M. T., White, N., Gee, N., and Hughes, C. (2017).

One of the family? Measuring young adolescents’ re-

lationships with pets and siblings. Journal of Applied

Developmental Psychology, 49:12–20.

Charles, E., Magnat, E., Jouannaud, M.-P., Payre-Ficout,

C., and Loiseau, M. (2025). An English listening

comprehension learning game and its effect on phono-

logical awareness. In Frisch, S. and Glaser, K., edi-

tors, Early Language Learning in Instructed Contexts,

Language Learning & Language Teaching. John Ben-

jamins, Amsterdam/Philadelphia.

Cornillie, F., Thorne, S. L., and Desmet, P. (2012). Digital

games for language learning: from hype to insight?

ReCALL, 24:243–256.

Czarniawska, B. (2007). Shadowing: And other techniques

for doing ﬁeldwork in modern societies. Liber.

Dat, M.-A. and Spanghero-Gaillard, N. (2005).

L’enseignement des langues et cultures

etrang

eres

a l’

ecole primaire : un exemple d’utilisation de

document authentique multim

edia. Corela, (HS-1).

Number: HS-1 Publisher: Universit

e de Poitiers.

Delasalle, D. (2008). Enseigner une langue

a l’

ecole : a-t-on

les moyens de relever ce d

eﬁ dans le contexte actuel ?

Etudes de Linguistique Appliqu

ee, 3(151):373–383.

Domsch, S. (2013). Storyplaying: Agency and Narrative in

Video Games. De Gruyter.

Duﬂo, C. (1997). Jouer et philosopher. Pratiques

eoriques. Presses Universitaires de France, Paris.

Garris, R., Ahlers, R., and Driskell, J. E. (2002). Games,

motivation, and learning: A research and practice

model. Simulation & Gaming, 33(4):441–467.

Gee, J. P. (2003). What video games have to teach us

about learning and literacy. Palgrave Macmillan,

Basingstoke. OCLC: 637002203.

Girard, C.,

Ecalle, J., and Magnan, A. (2013). Serious

games as new educational tools: how effective are

they? A meta-analysis of recent studies. Journal of

Computer Assisted Learning, 29(3):207–219.

Goldberg, A. E. (2003). Constructions: a new theoretical

approach to language. Trends in Cognitive Sciences,

7(5):219–224.

Good, J. and Robertson, J. (2006). Carss: A framework

for learner-centred design with children. International

Journal of Artiﬁcial Intelligence in Education, Inter-

national Journal of Artiﬁcial Intelligence in Educa-

tion, 4(16):381–413.

CSEDU 2025 - 17th International Conference on Computer Supported Education

390

Green, C. S., Li, R., and Bavelier, D. (2010). Perceptual

learning during action video game playing. Topics in

Cognitive Science, 2(2):202–216.

Kniberg, H. (2007). Scrum and XP from the Trenches – How

we do Scrum. C4Media.

Koch, M., von Luck, K., Schwarzer, J., and Draheim, S.

(2018). The novelty effect in large display deploy-

ments — experiences and lessons-learned for eval-

uating prototypes. In Gross, T. and Muller, M.,

editors, Proceedings of 16th European Conference

on Computer-Supported Cooperative Work — Ex-

ploratory Papers. European Society for Socially Em-

bedded Technologies (EUSSET).

Krashen, S. D. (1982). Principles and practice in second

language acquisition. Language teaching methodol-

ogy series. Pergamon Press, Oxford.

Loiseau, M., Jouannaud, M.-P., Payre-Ficout, C.,

Emilie

Magnat, Serna, A., and Guinet, A.-L. (2024). Large

scale experimentation of a serious game targeting oral

comprehension in esl for 6–8 year-olds: Some lessons

learnt. In CALICO 2024 — Conﬂuences and Connec-

tions: Bridging Industry and Academia in CALL.

Loiseau, M. and No

us, C. (2022).

A la recherche de

l’attitude ludique. In Silva, H., editor, Regards sur

le jeu en didactique des langues et des cultures, vol-

ume 9 of Champs Didactiques Plurilingues, chapter

1.3, page 61–83. Peter Lang Verlag, Bruxelles, Bel-

gique.

Loiseau, M., Payre-Ficout, C., and

Emilie Magnat (2021).

Le qr code dans le jeu s

erieux “luciole” : lien entre le

monde virtuel et le monde r

eel pour le d

eveloppement

de la compr

ehension orale en anglais

a l’

ecole. In

Actes de “Objets pour apprendre, objets

a apprendre :

Quelles pratiques enseignantes pour quels enjeux?”.

Mandin, S., Zaher, A., Meyer, S., Loiseau, M., Bailly,

G., Payre-Ficout, C., Diard, J., Fluence-Group,

and Valdois, S. (2021). Exp

erimentation

a grande

echelle d’applications pour tablettes pour favoriser

l’apprentissage de la lecture et de l’anglais. In 10e

conf

erence EIAH — Transformations dans le domaine

des EIAH : innovations technologiques et d’usage(s).

MEN (2013). L’enseignement des langues vivantes

etrang

eres et r

egionales (articles 39

a 40).

MEN (2015). Programmes d’enseignement de l’

ecole

ementaire et du coll

ege. Bulletin ofﬁciel sp

ecial du

26 novembre 2015 386, Minist

ere de l’

Education Na-

tionale, Paris.

MENJ (2019). Guide pour l’enseignement des langues vi-

vantes. Technical report, Minist

ere de l’

Education Na-

tionale et de la Jeunesse, Paris.

MEN/MR (2002). Horaires et programmes d’enseignement

de l’

ecole primaire. Bulletin Ofﬁciel hors-s

erie du 14

evrier 2002 1, Minist

ere de l’

Education Nationale et

Minist

ere de la Recherche, Paris.

Norman, D. and Draper, S. (1986). User Centered System

Design: new perspectives on Human-Computer Inter-

action. Lawrence Erlbaum Associate, London, UK.

Oblinger, D. (2004). The next generation of educational en-

gagement. Journal of Interactive Media in Education,

2004(1):1–16.

Read, J. C. (2008). Validating the fun toolkit: an instrument

for measuring children’s opinions of technology. Cog-

nition, Technology & Work, 10(2):119–128.

Reinhardt, J. and Thorne, S. L. (2019). Digital Games as

Language-Learning Environments, chapter 17, page

409–435. MIT Press, Cambridge, MASS.

Saiger, M. J., Deterding, S., and Gega, L. (2023). Chil-

dren and young people’s involvement in designing ap-

plied games: Scoping review. JMIR Serious Games,

11:e42680.

Seaborn, K. and Fels, D. I. (2015). Gamiﬁcation in the-

ory and action: A survey. International Journal of

Human-Computer Studies, 74:14–31.

Silva Ochoa, H. (1999). Po

etiques du jeu. La m

etaphore

ludique dans la th

eorie et la critique litt

eraires

franc¸aises au XX

ecle. Th

ese, Universit

e Paris 3

— Sorbonne Nouvelle, Paris.

Simonsen, J. and Robertson, T. (2013). Routledge Interna-

tional Handbook of Participatory Design. Routledge,

New York, USA.

Tsvyatkova, D. and Storni, C. (2019). A review of selected

methods, techniques and tools in child–computer in-

teraction (CCI) developed/adapted to support chil-

dren’s involvement in technology development. In-

ternational Journal of Child-Computer Interaction,

22:100148.

Valdois, S., Zaher, A., Meyer, S., Diard, J., Mandin, S., and

Bosse, M. L. (2024). Effectiveness of visual attention

span training on learning to read and spell: A digital-

game-based intervention in classrooms. Reading Re-

search Quarterly.

Vlachogianni, P. and Tselios, N. (2022). Perceived us-

ability evaluation of educational technology using the

system usability scale (SUS): A systematic review.

Journal of Research on Technology in Education,

54(3):392–409.

Vlachopoulos, D. and Makri, A. (2017). The effect of

games and simulations on higher education: a system-

atic literature review. International Journal of Educa-

tional Technology in Higher Education, 14(1).

Zaman, B. and Vanden Abeele, V. (2007). How to measure

the likeability of tangible interaction with preschool-

ers. In CHI Nederland, page 57–59, Woerden. Infotec

Nederland.

Zaman, B. and Vanden Abeele, V. (2010). Laddering

with Young Children in User EXperience Evaluations:

Theoretical Groundings and a Practical Case. In Pro-

ceedings of the 9th International Conference on Inter-

action Design and Children, IDC ’10, page 156–165,

New York, NY, USA. Association for Computing Ma-

chinery. event-place: Barcelona, Spain.

Luciole: The Design Story of a Serious Game for Learning English Created for (and with) Children

391