Content uploaded by Maria Rauschenberger
Author content
All content in this area was uploaded by Maria Rauschenberger on Aug 15, 2018
Content may be subject to copyright.
A Tablet Game to Target
Dyslexia Screening in Pre-Readers
Maria Rauschenberger
WSSC Group, DTIC
Universitat Pompeu Fabra
maria.rauschenberger@upf.edu
Luz Rello
HCI Institute
Carnegie Mellon University
luzrello@cs.cmu.edu
Ricardo Baeza-Yates
WSSC Group, DTIC
Universitat Pompeu Fabra
rbaeza@acm.org
Permission to make digital or hard copies of part or all of this work for personal or
classroom use is granted without fee provided that copies are not made or distributed
for profit or commercial advantage and that copies bear this notice and the full citation
on the first page. Copyrights for third-party components of this work must be honored.
For all other uses, contact the owner/author(s).
Copyright held by the owner/author(s).
MobileHCI ’18 Adjunct, September 3–6, 2018, Barcelona, Spain
ACM 978-1-4503-5941-2/18/09.
https://doi.org/10.1145/3236112.3236156
Abstract
Using serious games to screen dyslexia has been a suc-
cessful approach for English, German and Spanish. In a
pilot study with a desktop game, we addressed pre-readers
screening, that is, younger children who have not acquired
reading or writing skills. Based on our results, we have
redesigned the game content and new interactions with
visual and musical cues. Hence, here we present a tablet
game, DGames, which has the potential to predict dyslexia
in pre-readers. This could contribute to around 10% of the
population that is affected by dyslexia, as children will gain
more time to learn to cope with the challenges of learning
how to read and write.
Author Keywords
Dyslexia; Screening; Detection; Pre-Readers; Serious
Games; Computer-based Assessment; Universal Screen-
ing; Language-Independent; Visual cues; Musical cues;
Gamification
ACM Classification Keywords
K.4.2 [Computers and Society]: Social Issues—Assistive
technologies for persons with disabilities; K.3 [Computers
in Education]: Computer Uses in Education—Computer
assisted instruction
Introduction
Around 5% to 15% of the world population are affected
by dyslexia: a specific learning disorder [1]. While visual
and auditory difficulties might cause troubles in writing and
reading, the general intelligence of a person with dyslexia
is not affected. Nevertheless, school failure and frustration
is part of the daily routing for children and parents until the
child is finally diagnosed.
Children with dyslexia (CWD) are, until now, mainly distin-
guished by their reading and writing mistakes compared
to their peer group. Hence, screening pre-readers needs
new indicators. Current approaches to screen pre-readers
require expensive personnel (i.e., a professional therapist)
or special hardware (i.e., MRI machines). Our work tries to
simplify the screening of dyslexia for pre-readers.
First, we created a web-based prototype MusVis (mainly
for desktops) and conducted a study which served as a
proof-of-concept with students from 7 to 12 years old (n
=178) [12]. We addressed the participant’s feedback as
well as the game usage data and created a new application
for pre-readers that: (a) simplifies the musical game-play;
(b) provides musical content that is perceivable by pre-
readers; and (c) uses input methods that are adequate for
pre-readers. The major changes were done mostly on the
musical part of the game as well as for the adaptation of the
input method, i.e., for a tablet instead of a desktop.
In this demo, we present the new tablet application DGames,
with new game interactions and new content. We will use
the game to measure the children’s performance to conduct
an online user study to distinguish pre-readers with and
without dyslexia.
Related Work
The phonological skills deficiencies associated with phono-
logical coding deficits are probably the reason for dyslexia
[20]. Therefore, investigations on the visual and auditory
perception of dyslexia in relation to language acquisition of
pre-readers [8], rapid auditory cues with infants, and visual-
spatial attention [3] on kindergarten children are conducted.
Different games [4, 16, 10] or approaches [19] aim to
screen children with dyslexia mainly related to linguistic
knowledge. The AGTB 5 –12 aims to screen pre-readers
but is only available for the age of 5 till 12. The Cronbachs
Alpha is between .58 and .98 for children at the age of five
till eighth [7]. We could not find any other published ac-
curacy of the prediction process for pre-readers. The ap-
proach for pre-readers needs to be simple (e.g., tablet vs.
desktop) and should not assume existing knowledge of
literacy or phonological awareness.
We are combining findings from previous literature, which
are known to cause troubles for CWD, to create a game
environment to find solid differences for predicting dyslexia
in the future. At the same time, the game should be fun
and not too difficult. We expect people with dyslexia to take
more time and interact differently with the game than the
control group.
Game Content
The game DGames is a major revision of the game MusVis
[12]. Both games aim to detect differences in the perception
of children with and without dyslexia while playing with
musical and visual cues. Only the interaction for the visual
part and eighth visual cues are duplicated from MusVis. We
derived the new design of the game DGames from the pilot
study and implementation of MusVis [13, 12]. All changes
are reported below.
Although children (age 8 till 12) and their parents gave very
positive feedback on the game-play and content, parents
of pre-readers reported difficulties in the interaction of the
game. These parents reported, that their children had dif-
ficulties in understanding the game-play and distinguished
the very short and similar sounds of the musical part. Most
of them quit the game because of that. An example quote
from a dad of a boy (4 years): he was overwhelmed by the
game. He could not distinguish the sounds and just touched
randomly on any card. Also, the input method, i.e., com-
puter mouse was not adequate for younger children.
To be able to target pre-readers, we changed from a desk-
top to a tablet device, as well as, recreated completely the
musical content and interaction. Additionally, results from
MusVis and Dytective [16, 17] showed that CWD did not
make more mistakes while playing games, in spite that
CWD are diagnosed by the amount of written errors they
make. Therefore, also non-related linguistic visual and musi-
cal content is added to evaluate the game interaction itself.
Each game part (musical and visual) has 16 rounds which
are counter-balanced with Latin Squares [2].
Content Design with Visual Cues
The main changes from MusVis [12] are (a) adding non-
linguistic content, (b) tablet adaptation (e.g., double click),
and (c) video introduction to the game.
The visual part has 8 stages and 16 rounds. Each stage
is assigned to one visual type:
symbol, z, rectangle, face,
fruit, kitchen, plant & animal
and four visual cues for each
stage are presented (see Figure 1, where the first four vi-
sual types are duplicated from MusVis [12]). One visual cue
is the target which the participants need to find and click
(see Figure 2, a). The other three visual cues are distrac-
tors for the participants. Each stage has two rounds with
first a 4-squared (see Figure 2, b) and then a 9-squared
Figure 1: Overview of the designed visual cues. The figure shows
the target cue (top) and distractor cues (below) for the eight
different stages (symbol, z, rectangle, face, fruit, kitchen, plant,
animal ) of the visual part of the game DGames.
design (see Figure 2, c). The target and all three distractors
are displayed in the 4-squared design. In the 9-squared
design, the target is displayed twice as well as the distractor
two and three. Only distractor one is displayed three times.
Content Design with Musical Cues
The musical part has for each round a new musical type:
substitution, omission, phoneme, structure
(once Span-
ish and German vowel; Spanish consonant),
rhyme
(twice
Spanish and German; four times English),
combinations,
& rhythm
. Each musical type has one musical cue target
and three musical cue distractors. The new musical cues
are designed with the knowledge of previous literature (see
Table 1 and the new analysis of the published German
errors resource [11, 14]). The matrix shows the relations be-
tween our designed musical types and the literature which
provide evidence to distinguish a person with dyslexia.
The children click on the play-button and can listen to the
musical cue target as often as they like (see Figure 3, a).
After that a row of four buttons is displayed (see Figure 3, b)
Features Explanation of the Features Subs. Omiss. Struct. Phon. Rhyme Comb. Rhythm
Beginning
70% of the spelling errors are at the
third position of a word for German and
Spanish [15, 11].
x x x x x
Length
The average word length for German and
Spanish is just above 7 letters [15, 11]. x x x x x x
Simple
For 73.3% of the analyzed words for Span-
ish the Damerau-Levenshtein distance
was 1, which means that only one letter
mistakes were made [15]. For German it
is even 81.3% [11].
x x x x
Substi-
tution
The error category Submission (exchang-
ing a letter for another one) is frequent for
German, English and Spanish [15, 11].
x x x
Omission
The error category Omission (leaving a
letter out) is frequent in German [11]. x x x
Structure
CWD find it more difficult to recall a target
item with a similar prosodic structure [5]. x x x x x x x
Phono-
logical
STM
CWD showed impairments in the phono-
logical short-term memory (STM) [5]. x x x x x x x
Short-
Interval
Perception
Copying and discrimination tasks are used
to predict phonological awareness [9]. x x x
Pitch Mod-
ulation
CWD have difficulties in processing pitch
patterns [18]. x x x x x
Combi-
nations
Discrimination of rise time is related to
language processing [6]. x x
Complexity
CWD have difficulties with the phonolog-
ical similarity effect and the phonological
neighbourhood when long memory spans
are used [5].
x x x
Table 1: Mapping of the evidence from literature to distinguish a person of dyslexia to design the musical type for each stage in the musical
part of the game DGames.
Figure 2: Example of the visual part of the game DGames with
the priming of the target cue animal (a) and then the
fourth-squared (b) and nine-squared design including the
distractors for each animal (c).
and automatically the assigned musical cues for each but-
ton are played one after another. The buttons are disabled
until the auto-play is done to ensure children listen to all mu-
sical cues. The first button/musical cue is never the musical
cue target to force a distraction for the player. The order
of musical cues is randomly assigned and starts always
from left to right and with the Play all sounds again-button
children can listen to all cues as often as they like.
Implementation Details
Both game parts are developed as a web-application using
JavaScript, jQuery, CSS, HTML5 and a backend with a
PHP server and a MySQL database to make the game
easily adaptable for different devices. The visual part is also
implemented with Angular.
Because of the web implementation technique, a double
click on a web-application generally zooms the application
on a tablet. As young children were observed touching the
Figure 3: Example of the musical part of the game DGames with
the priming of the target cue (a) and then the distractors for each
musical cue (b).
application very quickly and triggering the zoom-effect, this
caused interruptions while playing and was not coherent
with the experience of using a native tablet application.
Therefore, we used a viewport meta tag to control the layout
settings for mobile devices.
All instructions within the game are presented with video or
audio media to address pre-readers. Android prevents by
default automatic play of sound or video and asks for a user
interaction. Therefore, we designed the whole game with
a sequence that starts with a user interaction followed by
audio sounds.
Conclusions and Future Work
The main advantages of the new tablet Game DGames
makes it playable for pre-readers and should improve the
current results for screening dyslexia in young children. Ad-
ditionally, the game is playable on different devices and the
device information are taken into account for the analysis
of the differences between the two user groups. To prove
our improvements, we plan to conduct a large-scale study
in different languages e.g. German, Spanish and English.
Acknowledgements
This demo is supported by the fem:talent Scholarship from
the Applied University of Emden/Leer as well as by the
Deutschen Lesepreis 2017 from the Stiftung Lesen and the
Commerzbank-Stiftung. Also, thanks to H. Witzel for his
advice during the development of the visual part and to M.
Blanca, and J. Carrion on the translation for the Spanish
version.
REFERENCES
1. American Psychiatric Association. 2013. Diagnostic and
Statistical Manual of Mental Disorders. American Psychiatric
Association, London,England. 991 pages. DOI:
http://dx.doi.org/10.1176/appi.books.
9780890425596.744053
2. Field and others. 2003. How to design and report experiments.
SAGE Publications, London. 384 pages.
3. Sandro Franceschini, Simone Gori, Milena Ruffino,
Katia Pedrolli, and Andrea Facoetti. 2012. A Causal
Link between Visual Spatial Attention and Reading
Acquisition. Current Biology 22, 9 (may 2012), 814–819.
DOI:http://dx.doi.org/10.1016/j.cub.2012.03.013
4. Luc Geurts, Vero Vanden Abeele, Véronique Celis,
Jelle Husson, Lieven Van den Audenaeren, Leen
Loyez, Ann Goeleven, Jan Wouters, and Pol
Ghesquière. 2015. DIESEL-X: A Game-Based Tool for
Early Risk Detection of Dyslexia in Preschoolers. In
Describing and Studying Domain-Specific Serious Games.
Springer International Publishing, Switzerland, 93–114.
DOI:
http://dx.doi.org/10.1007/978-3-319-20276- 1_7
5. Usha Goswami, Lisa Barnes, Natasha Mead,
Alan James Power, and Victoria Leong. 2016a.
Prosodic Similarity Effects in Short-Term Memory in
Developmental Dyslexia. Dyslexia 22, 4 (2016), 287–304.
DOI:http://dx.doi.org/10.1002/dys.1535
6. Usha Goswami, Ruth Cumming, and Angela Wilson.
2016b. Rhythmic Perception, Music and Language: A New
Theoretical Framework for Understanding and Remediating Specific
Language Impairment Background to the Project. Technical
Report. University of Cambridge. 1–7 pages.
http://www.cne.psychol.cam.ac.uk/pdfs/
NuffieldBriefinghttps://www.cne.psychol.cam.ac.
uk/people/themes/nuffield
7. Marcus Hasselhorn and Christof Zoelch. 2012.
Funktionsdiagnostik des Arbeitsgedächtnisses. Hogrefe Verlag,
Göttingen. 1–193 pages.
8. Heikki Lyytinen, Jane Erskine, Jarmo Hämäläinen,
Minna Torppa, and Miia Ronimus. 2015. Dyslexia -
Early Identification and Prevention: Highlights from the
Jyväskylä Longitudinal Study of Dyslexia. Current
Developmental Disorders Reports 2, 4 (dec 2015), 330–338.
DOI:http://dx.doi.org/10.1007/s40474-015-0067-1
9. Catherine Moritz, Sasha Yampolsky, Georgios
Papadelis, Jennifer Thomson, and Maryanne Wolf.
2013. Links between early rhythm skills, musical
training, and phonological awareness. Reading and Writing
26, 5 (2013), 739–769. DOI:
http://dx.doi.org/10.1007/s11145-012-9389-0
10. Nessy. 2011. Dyslexia Screening - Nessy UK. https:
//www.nessy.com/uk/product/dyslexia-screening/.
(2011). [Online; accessed 18-September-2017].
11. Maria Rauschenberger, Silke Füchsel, Luz Rello, and
Jörg Thomaschewski. 2017. Supplement:
DysListGerman Resource: A Language Resource of
German Errors Written by Children with Dyslexia.
(2017). DOI:
http://dx.doi.org/10.5281/zenodo.809801
12. Maria Rauschenberger, Luz Rello, Ricardo
Baeza-Yates, and Jeffrey P. Bigham. 2018. Towards
language independent detection of dyslexia with a
web-based game. In W4A ’18: The Internet of Accessible
Things. ACM, Lyon, France, 4–6. DOI:
http://dx.doi.org/10.1145/3192714.3192816
13. Maria Rauschenberger, Luz Rello, Ricardo
Baeza-Yates, Emilia Gomez, and Jeffrey P. Bigham.
2017. Towards the Prediction of Dyslexia by a
Web-based Game with Musical Elements. In The Web for
All conference Addressing information barriers – W4A’17. ACM
Press, Perth, Western Australia, 4–7. DOI:
http://dx.doi.org/10.1145/3058555.3058565
14. Maria Rauschenberger, Luz Rello, Silke Füchsel, and
Jörg Thomaschewski. 2016. A Language Resource of
German Errors Written by Children with Dyslexia. In
Proceedings of the Tenth International Conference on Language
Resources and Evaluation (LREC 2016) (23-28). European
Language Resources Association (ELRA), Paris,
France.
15. Luz Rello, Ricardo Baeza-Yates, and Joaquim Llisterri.
2016a. A resource of errors written in Spanish by
people with dyslexia and its linguistic, phonetic and
visual analysis. Language Resources and Evaluation 51, 2 (feb
2016), 1–30. DOI:
http://dx.doi.org/10.1007/s10579-015-9329-0
16. Luz Rello, Miguel Ballesteros, Abdullah Ali, Miquel
Serra, Daniela Alarcón, and Jeffrey P. Bigham. 2016b.
Dytective: Diagnosing Risk of Dyslexia with a Game.
https://dytectivetest.org/. (2016). [Online;
accessed 06-January-2017].
17. Luz Rello, Enrique Romero, Maria Rauschenberger,
Abdullah Ali, Kristin Williams, Jeffrey P Bigham, and
Nancy Cushen White. 2018. Screening Dyslexia for
English Using HCI Measures and Machine Learning. In
Proceedings of the 2018 International Conference on Digital Health -
DH ’18. ACM Press, New York, New York, USA, 80–84.
DOI:http://dx.doi.org/10.1145/3194658.3194675
18. Emily J. Rolka and Michael J. Silverman. 2015. A
systematic review of music and dyslexia. Arts in
Psychotherapy 46 (2015), 24–32. DOI:
http://dx.doi.org/10.1016/j.aip.2015.09.002
19. Claudia Steinbrink and Thomas Lachmann. 2014.
Lese-Rechtschreibstörung. Springer Berlin Heidelberg,
Berlin, Heidelberg. DOI:
http://dx.doi.org/10.1007/978-3-642-41842- 6
20. Frank R. Vellutino, Jack M. Fletcher, Margaret J.
Snowling, and Donna M. Scanlon. 2004. Specific
reading disability (dyslexia): what have we learned in
the past four decades? Journal of Child Psychology and
Psychiatry 45, 1 (jan 2004), 2–40. DOI:http:
//dx.doi.org/10.1046/j.0021-9630.2003.00305.x