Universal Access in the Information Society (2020) 19:739–781

https://doi.org/10.1007/s10209-019-00689-4

LONG PAPER

A framework to design vocabulary‑based serious games for children

with autism spectrum disorder (ASD)
Kamran Khowaja1   · Siti Salwah Salim2

Published online: 2 August 2019

© Springer-Verlag GmbH Germany, part of Springer Nature 2019

Abstract
The use of serious games to help children with autism spectrum disorder (ASD) learn different skills has increased in the
recent past. Games, however, have not been exploited to enhance the vocabulary of these children. The serious game design
framework (SGDF) and its components provide support throughout the design process, but there are no existing SGDFs
that can be used to design serious games for these children to learn vocabulary skills. Therefore, in this research, a SGDF is
proposed to overcome the problem of designing serious games for children with ASD. The components of the framework
were identified after an extensive review of the literature on ASD and existing SGDFs. The input, process and output model
was used as an underlying structure of the framework, and the identified components were logically placed in each phase of
the model. The proposed framework was iteratively evaluated through expert reviews with academic experts and research-
ers working in the area of serious games, as well as game designers. The modified version of the framework was produced
based on the comments provided by experts. This modified version was then used in the applicability evaluation of the
framework with researchers and game designers. A proposed SGDF was produced based on the comments of experts. The
main findings from expert reviews show that the structure, components and details of the proposed framework have been
effectively refined. The use of the framework to produce serious game design and the prototype demonstrates its use in real
time. The evaluation of the prototype with autistic children shows the improvement in their vocabulary learning before and
after using the prototype.

Keywords  Autism spectrum disorder (ASD) · Language comprehension skills · Vocabulary · Computer-based intervention
(CBI) · Serious game design · Prototype development · Virtual reality · Augmented reality

1 Introduction difficulty interpreting the meaning of the text, and vocabu-

The US Center for Disease Control and Prevention (CDC)
reported in 2014 that out of every 68 children, one is diag-
nosed with autism. These children face difficulties in social
communication or interaction and have restricted, repetitive
patterns of behaviour or interests. The symptoms and sever-
ity vary from one child to another. This means that while all
individuals with autism may share certain similar difficul-
ties, the symptoms and severity may affect their lives dif-
ferently. One common problem found among these children
is that while they are able to read the text, they may face
* Kamran Khowaja
kamran.khowaja@gmail.com
1
Isra University, Hyderabad, Pakistan
2 reason is that it requires the learners to identify the ways ‘to
University of Malaya, Kuala Lumpur, Malaysia
lary plays a vital role in overcoming this problem [1, 2].
According to the research conducted by Fukkink et al. [3],
for readers to understand and interpret what they are read-
ing, they need to know 95% of the words from that text. This
indicates that processing at the word level (vocabulary, in
other words) is the most essential part of comprehension; the
number of words the children know affects the interpretation
of the text they read. Words play an extremely important role
in day-to-day communication, and it is virtually impossible
to convey a message to someone without the proper use of
words [4]. Thornbury [5] explains that it is because of the
characteristics to make sense of the meaning of words that
vocabulary is indispensable to expanding the communica-
tion skills of the learner [6]; in this research, the learner is
a child with autism. Learning of vocabulary is a challeng-
ing task for the learner of any language, and one important

13
Vol.:(0123456789)

740
Fig. 1  Serious game development process [20]
learn, acquire, comprehend, retain, recall, use and expand
their vocabulary’ [7]. Without good comprehension skills,
the participation of these children in various classroom and
other curricular activities, as well as communication with
peers in a classroom environment and chatting platforms,
is hindered [8].
The importance of reading comprehension in children
with autism is highlighted in four review papers [1, 2, 9, 10]
that have been published in the recent past. It was found that
computer-based interventions (CBI) have been underutilised
in developing solutions for these children to learn vocabu-
lary. The research by Khowaja and Salim has highlighted
various technologies that have been used for the intervention
of other skills related to these children. They pointed out
two significant areas: (1) a number of researches focus on
the applications of serious games and (2) the use of serious
games for future research to facilitate the learning of vocabu-
lary and improvement in reading comprehension.
The concept of serious games was first used by the US
military in 2002 for military training of their staff through
game-based technology. It is apparent from the literature
review that there is no single accepted definition of serious
games. From one perspective, the use of the word ‘serious’
means embedding some pedagogical element in the games
[11]. From another perspective, ‘serious’ refers to the pur-
pose of these games [12]. An important characteristic of
serious games is to provide education of some skills and
facilitate users to improve these skills. Sawyer [13] stressed
that learning through serious games is not as fun and enter-
taining as typical games. Kiili [14] emphasised that fun
should not be a primary factor in the design of a serious
game. This indicates that a balance between fun and educa-
tion of some skill is necessary for a game. Therefore, edu-
cational theories need to be integrated into the design of
serious games to provide an interactive, engaging, motiva-
tional, fun-based and meaningful learning experience, while
13
Universal Access in the Information Society (2020) 19:739–781
children are learning by playing games. Therefore, a serious
game will not be the same as a typical game, the sole pur-
pose of which is to provide entertainment and fun.
The use of serious games for children with autism in
the form of therapy or education has increased since the
last decade. This is based on two recent reviews on serious
games for children with autism spectrum disorder (ASD;
[15, 16]). The skills that have been targeted in the serious
games include social and communication skills, the concept
of money, visual motor coordination and first-aid learning,
among others. From the analysis of the targeted skills, it was
found that the potential of serious games has not been tapped
to help children with ASD learn vocabulary; this finding is
similar to that of [2]. Noor and colleagues shortlisted and
reviewed thirteen studies between 2002 and 2011 (inclu-
sive), while forty studies were shortlisted and reviewed by
Zakari and colleagues between 2004 and 2014 (inclusive).
The results of most of the studies among these review papers
show that serious games have improved learning among
these children. The literature review of serious games high-
lighted that frameworks could be used to design or develop
a new serious game or evaluate any existing serious games.
Regardless of its purpose (design, develop or evaluate), a
framework consists of a structure, the components of which
are logically placed across the structure based on the rela-
tionships between components [17–19]. The framework and
its components play a vital role as they highlight what needs
to be incorporated in the game.
E-Games [20] has described five processes, as shown
in Fig. 1, to develop a serious game: (1) concept develop-
ment, (2) design, (3) implementation, (4) testing and (5)
deployment. It can be seen that the implementation of a seri-
ous game and its testing highly depend on the design (as
highlighted in grey colour) of the serious game; this shows
that design of the serious game is crucial. In the context of
designing for neurotypical children and children with special
Universal Access in the Information Society (2020) 19:739–781 741
needs, Hayes et al. [21] state that ‘Children with special
needs are increasingly using computers for a variety of tasks
and activities. However, designing for these children, even
those who are neurotypical can be extremely challenging.
Children develop and change mentally, emotionally, and
physically at a rapid pace’. This is also logically true for
children with ASD, as each child must be considered differ-
ent even if they are diagnosed with the same impairments.
The success of any serious game developed for children with
ASD lies in the design of the game.
Considering the importance of design in serious games, a
dedicated search for serious games design frameworks was
carried out, which included the generalised and specialised
frameworks. The components of the generalised frameworks
cater to the design of the serious game, irrespective of users
and their needs. Some examples of generalised SGDFs found
include those of [22–24]. The components of the special-
ised design frameworks take into consideration the users
and their needs in addition to the basic design of the serious
game. The number of constructions of specialised SGDFs
has increased in the recent past. Examples of specialised
frameworks include: avoiding phishing attacks [25]; educa-
tion of music [26]; elderly users [27]; social networks [28];
and online education [29]. These specialised frameworks
extracted from the Google Scholar database and after apply-
ing the forward searching technique on each of these articles
show that they have been used to design games. All of these
researches reflect the importance and need for construct-
ing specialised frameworks to serve specific purposes. The
search did not reveal any framework that could be specifi-
cally used to design serious games to provide learning of
vocabulary to children with ASD; however, the search did
reveal some related materials described below.
Park et al. [30] have developed a framework to design
assistive technologies that can be used to teach emotions to
these children. This framework only considers components
from the pedagogical perspective, which provide guidance
on how emotions can be taught step by step to these children.
The other main components are not considered, especially
those from the perspective of autistic children and those that
facilitate the designing of a serious game in general. There-
fore, this framework only meets the minimal requirements
for the design of games to learn vocabulary.
Whyte et al. [31] identified seven components for the
design of serious games to keep students motivated in play-
ing the game, as skills to be learned may be difficult and
may require a huge amount of time (weeks or months). The
components, gathered from two sources [32, 33], include
game story, goals, rewards and feedback, increasing levels of
difficulty, individualised training and choice, as cited in [31].
Children with ASD require an intensive intervention pro-
gramme to support them in the learning of skills associated
with the spectrum [34]. The intensive programme requires the
availability of the manpower necessary to support these chil-
dren in the learning. Bono and colleagues have developed an
automated serious gaming platform known as Gaming Open
Library for Intervention in Autism at Home (GOLIAH) to
support these children in the learning of two important skills:
imitation and joint attention (JA) at home as well as clinical
settings [35]. The platform includes eleven games of which
seven are related to imitation, while the remaining four are
related to JA. The main aim of the platform is to identify the
cognitive level of the children by allowing them to play games
with varying difficulty levels. Then, a series of games and their
difficulty levels are planned by the therapist for each child indi-
vidually to practise both skills at home. The results at the end
of the three-month trial show the improvement in the children.
Malinverni et  al. [36] proposed an inclusive design
approach to develop video games for autistic children. The
approach overcomes two major weaknesses found in the
game design: the therapeutic objectives were not met; the
games were unable to properly engage the children in play-
ing. This approach uses four phases to design video games;
clinicians’ expertise is used to elicit requirements in the first
phase. The second phase uses children to identify their inter-
est, motivation and preferences among other information. In
the third phase, the game designer uses information from the
previous two phases to define game mechanics and compo-
nents required throughout the game. The last phase is the
evaluation of the game with these children to determine its
suitability. The problem with this approach is that instead
of providing components that could be used to guide the
designing of games, this approach solely relies on the game
designers to identify the necessary components required for
designing games themselves, based on the information gath-
ered from the clinicians and children.
Since no known existing framework can be used to design
games for autistic children, this research attempts to construct
a specialised framework to design serious games for them.
The framework is constructed using an existing framework
as a base, and an alternate version of the existing structure is
created. The components are enriched through an extensive
review of the literature from the following perspectives: (1)
children with ASD, (2) children without an impairment, dis-
ability or a neurotypical person and (3) the game design itself.
The components associated with ASD have been enriched with
the details related to learning of vocabulary among the autistic
children; thus, the framework can then be utilised to design,
develop and evaluate serious games for this specific purpose.
2 Method
A conceptual or theoretical framework may be described
as a visual or written narration of the main aspects to be
studied in the form of key factors, variables and presumed
13

742
Fig. 2  Literature sources used for the identification of components
relations among them [37]. There are four different ways in
which a framework can be constructed: (1) experimental
knowledge; (2) existing theory and prior research; (3) pilot
and exploratory research; and (4) thought experiments [38].
In this research, existing theory and prior research are used
as a basis to construct a framework.
There are three essential processes that need to be per-
formed to construct a framework (SGDF): (1) identify com-
ponents to be incorporated in the framework; (2) identify an
underlying structure that can be used as a basis to construct a
new framework; and (3) the logical placement of these com-
ponents in the identified structure. The first two processes
are the outcomes of the review of the literature, while the
third process uses each component and its purpose from the
first phase and maps it on the identified structure from the
second process. Each of these processes is described in the
following subsections.
2.1 Identification of components
The literature sources that contributed to the framework
are shown in Fig. 2: an arrow from one source (current) to
another source (next) shows how a review of the literature
has been expanded based on the gaps found in the current
source. The review of the literature started with a focus on
strategies used to teach vocabulary to children with ASD.
Irrespective of the contents to be taught to these children, it
is important to identify the techniques that have been used
to motivate and provide an effective learning environment
for these children. A systematic literature review (SLR) was
used to locate all the relevant studies to evaluate, interpret
and synthesise the information from the studies to answer
research questions [39]. This approach requires a searching
strategy and certain inclusion and exclusion criteria, among
others, to locate and shortlist studies, which result in the in-
depth review of the area being investigated.
Research on ASD has shown that children diagnosed
with ASD often exhibit co-occurring sensory processing
13
Universal Access in the Information Society (2020) 19:739–781
problems, for which they are provided intervention to self-
regulate themselves in day-to-day life [40]. Although the
problems associated with the sensory processing are neither
universal nor specific to ASD, the prevalence of such abnor-
malities in these children is relatively high [41]. Therefore,
perception of the modalities supported by the CBI and its use
among these children may vary from one child to another.
Thus, an approach of SLR was also followed to investigate
the use of modalities in CBIs for these children and the sup-
port in the generalisation and maintenance of skills learned.
An approach of narrative literature review (NLR) was fol-
lowed for the review of remaining sources which are related
to ASD, common components used in the existing SGDFs
for autism, children and game design, and game-based learn-
ing attributes.
Table 1 shows the components that have been identified
from both types of reviews i.e. SLRs and NLRs. Column 1
shows the components. The purpose of each component is
briefly described in column 2, while all the sources in which
each component is used are presented in the last column.
Winn [23] described the heart of a serious game as shown in
Fig. 3, requiring consideration of components from three dif-
ferent perspectives, namely theory, content and game design.
In the context of this research, there is a need to con-
sider components from the perspectives of ASD as well.
This leads to the heart of serious game design for children
with ASD, as shown in Fig. 4, where these perspectives,
namely theory, contents and game design, are the same as
the perspectives shown in Fig. 3, while the addition of the
ASD perspective is new to this figure. The figure shows the
mapping of all the components from Table 1 to the four
perspectives. The following are brief descriptions of each
perspective in the context of ASD:
• Theory represents all the relevant theories that can con-
tribute to the design of the game for these children, be
they related to ASD, learning, or psychology of child or
communication with a child, among others;
Universal Access in the Information Society (2020) 19:739–781 743

Table 1  Components, their purpose in the framework and the sources used to gather their details
Component Purpose Sources

Autism behaviours and symptoms Intervention of behaviours American Psychiatric Association [43], Bos-
seler and Massaro [44], Center for Disease
Control and Prevention [CDC] [46], Charac-
teristics of autism [42], Johnson [45]
Instruction method Effective intervention National Autism Center [57]
Instructional strategies Motivation and focused learning Khowaja and Salim [2]
Modalities Alternate ways of interaction with system Basil and Reyes [79], Bosseler and Massaro
[44], Hetzroni and Shalem [80], Hetzroni and
Tannous [81], Massaro and Bosseler [82],
Moore and Calvert [83), Whalen et al. [84]
Desired capabilities Develop skills Yusoff [24]
Instructional contents Provide subject matter Garris and Ahlers [52], Yusoff [24]
Intended learning outcomes Define goals Garris and Ahlers [52], Yusoff [24]
Learning activity Engaging activities Yusoff [24]
Game genre Categorise activities Yusoff [24]
Game mechanics Rules and regulations Winn [23], Yusoff [24]
Game dynamics Emergent behaviour Winn [23] gaming websites
Game story Develop story Winn [23]
Storytelling Designers’ & player’s story Winn [23]
Narratives Written or spoken words of designers’ story Winn [23]
Non-player characters Artificial intelligence-based non-player characters Winn [23]
Game-based learning attributes Validate learning as done in classroom environment Garris and Ahlers [52], Yusoff [24]
Reflection Improve thinking Yusoff [24]
User profile User details Yusoff [24]
User achievements User performance Yusoff [24]

• ASD represents the key components to be considered

from the ASD perspective in the design;
• Game design comprises of all the components that con-
tribute to the design of a serious game.

2.2 Identification of structure

The review of the literature on existing serious games design

Fig. 3  Heart of serious game design [23]
• Contents represent the subject matter to be taught to an
individual child with ASD; in this research, the subject
matter related to the vocabulary is represented as con-
tents;
frameworks has revealed that an input, process and output
(IPO) model proposed by Garris et al. [22] is widely used
as a basis to construct a framework for games. This model
is considered as a generic framework, as it allows the design
or development of different games based on two aspects:
(1) types of outcomes to be achieved and (2) contents to
be learned through game. This model consists of three
phases: (1) input; (2) process; and (3) output. The input
phase requires consideration of outcomes, contents and
game-based learning attributes. The key of the IPO model
resides in the process; a central hallmark of the game is
not that a user plays the game once and leaves it aside, but
that the game should generate motivation to keep on playing
over and over again. The output validates the performance
of the user against the objectives of the game; this indi-
cates the overall success of playing the game. The criteria

13

744 Universal Access in the Information Society (2020) 19:739–781

Fig. 4  Heart of serious game for

children with ASD

Table 2  Logical placement of components in the phases of IPO model

Phase Role Components

Input Contents for the intervention of behaviours Autism behaviours and symptoms, desired capabilities, instructional contents, intended
learning outcomes
Process Learning activities Instruction method, instructional strategies, modalities, reflection, learning activity, game
genre, game mechanics, game dynamics, game story, storytelling, narratives, non-player
characters, game-based learning attributes
Output Debrief and save user performance User profile and achievements

to measure the performance of the user may vary from one
game to another. It may be in the form of a number of cor-
rect answers given by the user, time spent to provide answers
or another type of measurement. The structure of the IPO
model depicts the learning process used in a game in which
a user selects the content to learn as an input and then under-
goes the rigorous sub-process of learning by taking part in
various activities in which assessment is also conducted to
evaluate the learning in the user; the result of the assessment
determines the rate of achievement of the expected learning
outcomes.
Considering the wide applicability of IPO in the construc-
tion of the framework, and the design or development of
games, this structure has been used in the present research
to construct an alternative version of the existing SGDF.
the components are synchronised with that of the IPO by
Garris et al. [22]. Table 2 shows the logical placement of all
the components in the phases of the IPO model. It can be
seen that input comprises four components, namely autism
behaviours and symptoms, desirable capabilities, instruc-
tional contents and intended learning outcomes. The process
comprises thirteen components, namely instruction method,
instructional strategies, modalities, reflection, learning
activities, game genre, game mechanics, game dynamics,
game story, storytelling, narratives, non-player characters
and game-based learning attributes. The output contains two
components: user profile and achievements.
3 Serious game design framework (SGDF)

2.3 Logical placement of components
in the structure
Each component identified in Sect. 2.1 needs to be logically
placed in the phases of the IPO model identified in Sect. 2.2.
This ensures that both the structure of the framework and
Figure 5 shows the serious game design framework (SGDF)
based on the identified IPO structure and the twenty compo-
nents logically placed into input, process and output of the
structure using the information presented in Table 2. The
following subsections describe all the phases of IPO and the
components associated with it.

13
Universal Access in the Information Society (2020) 19:739–781 745
Fig. 5  A serious game design framework for children with ASD
3.1 Input
Input refers to the data process that identifies the behaviour
related to the vocabulary in which children with ASD face
difficulties, objectives that need to be achieved and accord-
ingly the appropriate contents for these children with ASD
to learn and improve the behaviour related. The following
subsections describe the components of the input phase.
3.1.1 Autism behaviours and symptoms
Each child with ASD exhibits a unique set of behaviours.
Before the game design process begins, it is important to
identify the behaviours that need to be targeted in the serious
game. The identification of these behaviours will help in the
selection and creation of appropriate instructional content
and learning activities. Table 3 shows a list of behaviours
specific to communication skills of ASD which have been
classified into language comprehension, verbal communi-
cation, non-verbal communication and general behaviours.
These behaviours have been compiled and consolidated from
various sources [42–46] among others, rather than taking it
from one school of thought.
3.1.2 Desirable capabilities
Desirable capabilities refers to the development of one or
more of three types of skills, namely cognitive, psychomotor
and affective skills that players of the game can develop as a
part of playing the game [24]. These skills are based on the
cognitive, psychomotor and affective domains.
Cognitive skills refer to the development of thinking abil-
ity in the players which they use to solve problems of vary-
ing degrees, from easy to difficult, either as a part of learn-
ing activities or evaluations following learning of content
through the game [47]. The categories of cognitive skills
have been drawn from Bloom’s taxonomy which includes
recall, understanding, application, analysis, synthesis and
evaluation.
Psychomotor skills are related to the development of
motor skills, physical movements and coordination of body
parts so that an affected child can manually perform all the
tasks [48]. The psychomotor skills have not been catego-
rised by Bloom and his colleagues; however, there are other
researchers who have categorised the skills and created
their own psychomotor taxonomies. These include Simp-
son’s psychomotor domains [48], Harrow’s psychomotor
domains [49] and Dave’s psychomotor domain [50]. The
categorisations by Simpson and Harrow are useful for the
development of this skill from children to adults. Table 4
shows how psychomotor skills have been categorised by dif-
ferent researchers.
Affective skills are all about learning how to deal with
things emotionally, including feelings, values, appreciation,
enthusiasm, motivation and attitude [51]. The taxonomy cre-
ated by these authors for the affective domain includes five
categories, namely receiving, responding, valuing, organisa-
tion and value complex.
13

746 Universal Access in the Information Society (2020) 19:739–781

Table 3  Categorisation of behaviours associated with communication skills

Language comprehension Verbal communication Non-verbal communication General behaviours

1. Content and grammar may be
delayed
2. Rigid understanding of words
3. Difficulty with the concept that
objects can have more than one
name
4. Difficulty in identifying nouns
5. Reverse pronouns (e.g. says
‘you’ instead of ‘I’)
6. Inappropriate use of verbs
7. Inappropriate use of adjectives
8. Struggle to receptively or
expressively label places, people,
objects
9. Language may be slow to
develop
1. Speech may be delayed, or there
may be no speech at all
2. If verbal, might not know how
to start, sustain or end conversa-
tions
3. Frequently use echolalia
(repeating words of others)
4. Show unusual tone of voice
(e.g. monotone, robotic or high
pitch)
5. No consistent response to name
6. Unable to express wants and
needs
7. Talks in a flat, robot-like or
sing-song voice
8. Less likely to make comments
or ask questions
9. Less likely to make bids for
social attention
1. No effort to use non-verbal 1. No response in normal teaching
communication methods
2. Use fewer gestures 2. Less likely to share experiences
3. Difficulty in following direc- 3. Gives unrelated answers to
tions questions
4. Appear not to hear at times
5. Unable to point or wave
6. No accurate interpretation of
puns, sarcasm, idioms, etc.
7. Does not pretend in play
8. Does not understand jokes,
sarcasm or teasing

Table 4  Categorisation of psychomotor skills by various researchers
Sources Categorisation
Simpson [48] 1. Perception (awareness)
2. Set
3. Guided response
4. Mechanism (basic proficiency)
5. Complex overt response (expert)
6. Adaptation
7. Origination
Harrow [49] 1. Reflex movement
2. Fundamental movements
3. Perceptual abilities
4. Physical abilities (fitness)
5. Skilled movements
6. No discursive communication
Dave [50] 1. Imitation
2. Manipulation
3. Precision
4. Articulation
5. Naturalisation
3.1.3 Intended learning outcomes
The intended learning outcomes are the overall goals that a
player has to achieve by playing a serious game [24, 52]. It
also refers to the abilities acquired by the players once they
have completed playing a serious game [53]. Each outcome
of a serious game is created based on the combination of
the specific capability to be achieved in the user, as well as
specific instructional contents. Each learning activity to be
incorporated in the serious games should have at least one
learning outcome that can be achieved by the child (player
of the serious game). These outcomes are decomposed into
all the activities of the serious game based on the overall
instructional contents to be learned by the players. This way,
players know in advance what needs to be accomplished
and what to expect from the material to be learned within
a specific activity. According to the pedagogical guidelines
based on Gagne’s nine events of instruction [54], these
learning outcomes should be communicated to the user.
The outcomes are typically shown at the introduction of the
serious game and are also available through the help menu
of the game.
3.1.4 Instructional contents
Instructional content refers to the specific content or sub-
ject matter that a player can learn through the serious game
[24, 52]. The categories of instructional content and specific
contents supported by the game could be an exhaustive list.
According to Gilbert and Gale [55], there are four differ-
ent types in which content can be classified, namely facts,
concepts, procedures and principles. Each of these types is
briefly described below.
Fact A statement that is true or real; it could also refer to
something that can be shown to be real or true. It consists of
two specifics ‘A’ and ‘B’ so-called fact pair about something;
these specifics could refer to a definition, a name or a num-
ber. For instance, one litre of milk weighs one kilogram or
Malaysia is a federation consisting of thirteen states (Negeri)
and three federal territories (Wilayah Persekutuan).
Concept A basic understanding of something that is nec-
essary to make sense out of it. This type of learning is typi-
cally based on values, ideas, symbols, events or things that
are associated with it. A concept has name ‘X’, a superordi-
nate class ‘Y’ of which X is a member and a list of attributes
‘An’ and values ‘Vn’ where ‘Vi’ corresponds to ‘Ai’. These

13
Universal Access in the Information Society (2020) 19:739–781 747
attributes serve a purpose to classify or sort objects of differ-
ent types. The use of the phrase ‘is a’ distinguishes a concept
from a fact. For instance, ‘an apple’ is a ‘fruit’ and list of
attributes could be ‘colour’, ‘weight’ and ‘country of origin’,
‘taste’ among others.
Procedure Learning how to perform a sequence of step-
by-step actions to accomplish a task. A procedure typically
has a name ‘X’ and it is used in some situation ‘S’ to achieve
goal ‘G’ using set of steps ‘E’ by using some tool ‘T’. For
instance, to eat a pomegranate, a person first cuts its crown
with a sharp knife, then scores the pomegranate with cuts as
if to break it into quarters and then soaks it in cold water as
this allow the seeds to be easily collected. While the pome-
granate is under water, it is gently pulled apart into quarters.
Then, one runs fingers through each quarter slice to start
separating the seeds.
Principles Propositions which contain a name ‘X’ are
applied in a situation ‘S’ and involve the specification of
cause–effect relationships (CERs) between objects or events
‘E’.
3.2 Process
This refers to the process of using data from the input phase
and developing all the necessary learning activities that
meet one or more of the identified outcomes; all the compo-
nents in the doughnut and the sub-components of modali-
ties, game-based learning attributes and story contribute to
the learning activities. The key of the IPO model used in
this research resides in the process; a central hallmark of
the game is based on the design of activities that keep chil-
dren motivated through the interactive medium of learning
instructional content that can increase their interest level to
play the game over and over again. Each activity needs to
address one or more of the learning outcomes to ensure that
the child would be learning the same content and also facili-
tate the same capabilities as identified in the input phase.
The components that facilitate the design of learning activi-
ties are described in the following subsections.
3.2.1 Reflection
Serious games typically consist of a set of interrelated learn-
ing activities where each activity builds on top of the previ-
ous activity in terms of the difficulty level of activity. Reflec-
tion is about giving an opportunity to the player to think
about the reasons behind undertaking the current learning
activity [24]. This thinking process is based on all the learn-
ing activities the player has performed since he/she started
playing the game. This allows the player to understand and
identify the skills and approaches used to solve similar prob-
lems in the previous activities and identify approaches and
skills needed to solve problems in the current activity.
This process can be facilitated by providing information
to the players towards the end of each activity about: (1) the
current activity was undertaken; (2) their achievement in
the activity; and (3) the objectives accomplished through
the activity. This information could be useful for the player
as they may otherwise forget the actual objective behind the
current activity. Garris et al. [22] have also recommended
that a similar set of information be provided to the player.
The set of information includes: (1) a description of the cur-
rent activity undertaken; (2) the purpose behind undertaking
the current activity; (3) all the errors made throughout the
activity; and (4) suggestions on how to correct those errors.
According to Seamon [56], Kolb’s problem-solving cycle
can also be used to support a player’s learning based on
the reflection. This cycle includes four steps that need to
be incorporated into the learning activity: doing, reflecting,
understanding and applying.
3.2.2 Instruction method
Instruction methods for ASD are the evidence-based meth-
ods that have proved to be effective for some individuals
on the spectrum [57]. The NAC report states that there is
no universal method as such which has proved to be the
most effective method for all the children on the spectrum.
The use of instruction methods in the intervention for these
children highly depends on the skills targeted in the inter-
vention. The skill targeted throughout this research is the
vocabulary of these children. Table 5 provides the advan-
tages and disadvantages of five different instruction methods
that can be used to provide intervention of vocabulary in
particular or language in general.
3.2.3 Instructional strategies
Strategies are the techniques which facilitate instructors to
get the attention of children, stimulate their interest towards
the learning of contents, keep them engaged in the learning
process, develop their thinking skills and increase interac-
tion so that they remain focused on the learning [2]. The
instructor typically uses strategies to connect ideas associ-
ated with the instructional content to the real world; this
provides an opportunity for these children to generalise the
contents from one situation to another. Figure 6 provides a
list of strategies that can be used for vocabulary instruction
and text comprehension instruction for these children.
3.2.4 Learning activities
Learning activities are a set of activities designed to keep
players actively involved when they are playing serious
games. The effective design of these activities ensures that
players stay engaged and immersed without getting bored.
13

748 Universal Access in the Information Society (2020) 19:739–781

Table 5  Summary of instruction methods

Instruction method Age range Advantages Disadvantages

Discrete trial training (DTT) 0–21 Useful for basic language skills, numbers, Prompt dependency
letters, colours, etc Need to reprogramme for spontaneous skill
Teacher controlled use
Little thought involved Does not build fluency
Easy to assess the data
Treatment and Education of Autistic and 0–18 Independence Social interaction and verbal communica-
related Communication Handicapped Predictability tion are not emphasised
Children (TEACCH) Routine Does not promote interaction with typical
Structure peers
Consistency
Incidental teaching 0–9 Can be done anytime, anywhere, by Depends on engagement
anyone Cannot control number of trials/data analy-
Does not require massive training sis issues
Workable in a developmental classroom Teacher must remain hypervigilant or cre-
setting ated opportunities
Short episodes Must create opportunities
Natural, activity-based
Spillover to play skills
Less protest/escape/aversive control
Encourages spontaneous skill use
Pivotal response training 3–9 Child initiates Labour intensive
Natural reinforcers Staff must be adequately trained in the
Increased motivation method
Deters inappropriate behaviour
Verbal response training Assessment of basic language and learn- Cost may be high to the school districts
ing skills (ABLLS) allows for tracking Labour intensive
of a child’s progress Requires small staff-to-pupil ratios
Useful verbal operators
May be used at school and in the home

Fig. 6  Strategies with the frequency of use for vocabulary instruction and text comprehension instruction of children with autism [2]

Another important aspect is to ensure that the learning mate-
rial used is appropriate and challenging for players who are
seeking improvement in their competency slightly above
their current level of competency [58]. These activities make
use of the material from the instructional contents through-
out the game.
3.2.5 Game genre
The game genre is a term used to categorise games based
on the interaction between the players’ so-called game play
rather than differences in visual or narratives [24]. It is
defined by a set of game play challenges. These genres range

13
Universal Access in the Information Society (2020) 19:739–781 749
from action–adventure to a combination of action–adven-
ture, role-playing, simulations, strategy and sports, among
others.
3.2.6 Game mechanics
Game mechanics are a set of actions, behaviours and control
mechanisms given to a player within the context of a game
[23, 59, 60]. These mechanics are used to define a rule-based
system for the game environment by specifically stating all
the objects that would be available within the game environ-
ment, how each object would behave and how the user can
interact with them in the game world.
3.2.7 Game dynamics
Game dynamics refer to the change that occurs in the game
play when the game mechanics are activated within the envi-
ronment of the serious game [23, 60]. This provides more
fun, enjoyment and engagement for the player in the games.
Assume the user is playing a time-based game in which the
user needs to collect specific fruits falling from trees (placed
at a different horizontal location with varying height) in the
basket before they touch the ground. These fruits are mixed
with other fruits that act as distractors. The user is penal-
ised for catching one of the distractors. Both actual and dis-
tractor fruits fall at the same constant speed until 30 s are
remaining, after which the game play changes; the speed is
increased by twice and the penalty factor is also doubled.
3.2.8 Game story
The overall story of the game consists of one or more of
the three components, i.e. storytelling, narratives and the
characters involved in the game. These three components of
the game story are described in the following subsections.
3.2.8.1  Storytelling  The use of games as a medium for sto-
rytelling has largely increased in the past decades [61]. This
has become possible because of improvements in the tech-
nology; the game designers are able to introduce characters
into the game environment. Sinclair states that ‘Games have
done more work to pioneer best practices in interactive and
participatory storytelling than any medium since the ancient
theatre traditions’.
Storytelling in a serious game can take two different
forms: (1) designer’s story and (2) player’s story [62] as
cited in [23]. A designer’s story highly depends on the inte-
gration of various aspects related to the serious game, i.e.
the context in which a serious game is to be played, instruc-
tional content to be learned by the user, learning activities
to be created and how characters and narratives are syn-
chronised with each other and used in different activities.
The resultant story is incorporated into a serious game. The
designers tend to describe the situation to others exactly the
way they perceived it (Rouse). The storytelling that users
experience while playing the serious game is based on the
designer’s version of the story, as well as all the interactions
and the selections made by a user throughout the serious
game. The overall resultant experience becomes the player’s
own version of the story, and it is unlikely that the stories
of two players would be the same. The use of designer’s
story highly depends upon the game genre of the serious
game, i.e. their story is strongly used in the adventure or
role-playing games, whereas in game genres like logic or
board games, it is possible that the designer’s story is little
used or not used at all. However, each serious game does
include a player’s version of the story, which at the mini-
mum includes all the activities played and the challenges
performed throughout the game. It is at the discretion of
the game designer to decide which type of story should be
experienced by the player.
3.2.8.2  Narrative A narrative refers to one of the story
events presented to the player within the game world [63].
It consists of a non-interactive presentational material either
in the form of text only or a combination of text, audio and
animation shown to the user as a part of the narration. As
Adams describes, the purpose of a narrative is to present
an event that the user cannot control or interact with. The
narrative is typically used to establish the setting of an envi-
ronment in which a player would provide initial motivation
towards the game, but often, it is not the main focus of game
play. It is also shown when the user moves from one activity
or type of content to another, towards the end of the level
when the user has accomplished the goals or towards the end
of the game as well. The narrative allows designers to keep
players engaged in the game through the way a story is told
[61]. Although the players know the predefined outcome of
the game, they are kept engaged because of the unfolding of
uncertain events and the way the story progresses.
3.2.8.3  Non‑player characters (NPC) Aarseth [64] catego-
rises characters into three kinds based on their depth/shal-
lowness and malleability/potential in order for players to
control the characters. These categorised characters include
the following: (1) bots [65], also referred to as robots, and
sometimes interchangeably used with the term non-player
character (NPC); (2) shallow characters which are given
names and have personal appearances in the game but with
limited personality; and (3) deep characters which are fully
controlled by the player.
The player of real-time games, excluding strategy games,
typically controls one or more characters within the game
environment and plays against a set of characters either con-
trolled by other players as typically happens in a multiplayer
13

750
Table 6  Mapping of game-based learning attributes to the theories of learning and psychology
Behaviourism Cognitivism
Attributes
1. Interaction (equipment) 1. Incremental learning
2. Interaction (interpersonal) 2. Linearity
3. Interaction (social) 3. Attention span
4. Language or communication 4. Transfer of learnt skill
5. Reward 5. Motivation
6. Practice and drill
7. Feedback
8. Rules/goals
9. Positive feelings
10. Conflict
environment or controlled by the computer [66]. These
authors refer to the computer-controlled characters as NPCs;
the behaviours of these characters are controlled by a piece
of code specially written in the artificial intelligence (AI)
language such that their actions, reactions and appearance
are extremely natural. As a part of their research, they have
presented the model-based approach to control the behav-
iours of these NPCs.
Thus, the role of each character used within a serious
game may vary from one another. They may be used to pro-
vide support to the user whenever they are stuck within the
game environment or not sure what to do next. They can be
part of the user’s team in a strategy-based game to support
the player in achieving her goals or they can even be in the
opposite team as your enemy. These characters can also play
the role of some random character who is freely moving
around within the game environment and may or may not
have an ability to interact with the character when both are
next to each other.
3.2.9 Game‑based learning attributes (GBLAs)
The learning through serious games needs to be validated
against current teaching practices to ensure it is at the same
level as that of traditional classroom learning [24, 52].
Thus, there is a need to incorporate certain aspects of cur-
rent teaching practices in serious games that have proved to
be successful in the classroom. These successful aspects are
referred to as game-based learning attributes. These attrib-
utes support learning and engagement throughout the game;
this allows users to remain active and think critically while
learning and playing games. These attributes are based on
different theories, namely behaviourism, cognitivism, con-
structivism and theory of psychology. Table 6 shows a list of
all the attributes and their classification into one of the four
mentioned theories. These attributes are based on two frame-
works, i.e. [22, 24], and three research articles, i.e. [67–69].
13
Universal Access in the Information Society (2020) 19:739–781
Constructivist Psychology
1. Scaffolding 1. Situated and authentic learning
2. Learner control 2. Accommodating the learner’s style
3. Progress 3. Intermittent feedback
4. Adaptation 4. Assessment
5. Location 5. Fantasy
6. Surprise 6. Pieces or players
7. Representation
8. Sensory stimuli
9. Safety
10. Challenge
11. Mystery
12. Intensity
3.2.10 Modalities
Modality acts as a channel of communication between the
user and the computer (more precisely, the serious game
in the context of this research) itself [70]. It can be used
to convey some information to the serious game or acquire
some information from the serious game in return. Games
that make use of just one modality are known as unimodal,
whereas games in which two modalities are used are known
as bimodal and games that integrate more than two modali-
ties are known as multimodal games [71]. The combination
of two or more modalities for input is known as multimodal
fusion, and the partitioning of information into two or more
communication modalities is also known as multimodal fis-
sion (ibid.). Examples of input modalities include speech,
gestures, gait and facial expression, among others, whereas
text, graphics, animation, video model, virtual character and
force feedback are a few examples of output modalities. Fig-
ure 7 provides a list of potential modalities (non-underlined
modalities) that could be used in the studies related to lan-
guage comprehension and decoding skills for children with
ASD [72]. The remaining modalities (underlined) represent
those modalities which have been used for both above-men-
tioned skills.
3.3 Output
This part of game design focuses on the user profile, and
the debriefing of all events would happen as a child plays
the game. Each learning activity from the earlier phase is
taken into consideration for the development of user profile
and to ensure that all essential information related to user
performance in the activity is retained in the profile. The
overall outcome of following through all three phases of the
IPO model reveals an integrated design and all the neces-
sary details of a serious game for children with ASD to learn
vocabulary. The components of output are described below.
Universal Access in the Information Society (2020) 19:739–781 751
Fig. 7  Potential modalities for the CBIs of language comprehension and decoding skills for children with ASD [72]
3.3.1 User profile and achievements
Serious games typically allow users to set up their own pro-
files that contain their basic information. Once created, this
profile is updated regularly with all the accomplishments as
they progress through various activities of the games. This
way, users can also look back at all of their achievements.
It can also allow them to practise a previously learned skill
to improve their learning. Achievement in the games can
be shown in the form of scores, a total number of resources
gathered in the games or the time taken to accomplish that
task.
3.3.2 Debriefing
The purpose of debriefing is to provide a review of all the
activities in which the user took part while playing the game
[22]. For each activity, this review may provide a description
of the activity performed, their performance in the activity,
highlighting all the incorrect responses as well as the correct
responses that should have been selected. The use of debrief-
ing or post-experimental analysis in the educational setting
was first reported by Lederman [73]; it was first used in the
simulation or game. Debriefing plays an important role in
serious games to improve the overall learning and experi-
ence of the user (Garris). Consideration needs to be given
on how to display the information so that the user does not
feel overloaded and can easily concentrate on the corrective
actions that need to be taken.
4 IPO‑based frameworks to design serious
games
Section 2.2 has highlighted the wide applicability of IPO
in the construction of serious games design frameworks.
In this section, two frameworks constructed based on IPO
structure are shown in Table 7: (1) the serious games design
framework proposed in this research which is specifically for
children with ASD to learn vocabulary and (2) the serious
games design framework in general [52].
The IPO details (i.e. information gathered and outcomes)
of the framework constructed in this research (i.e. second
column) are based on the components identified from the
extensive review of the literature on ASD and existing
SGDFs for ASD, typical children and game design in gen-
eral. The IPO details of the framework by Garris and Ahlers
[52] are generic and taken from the perspective of game
design in general, i.e. irrespective of any specific user.
5 Evaluation of the proposed framework
The initial version of the proposed framework was thor-
oughly evaluated through multiple evaluation methods,
including expert review studies, an applicability test and
applicability survey. The methods used and the outcome pro-
duced throughout the evaluation are shown in Fig. 8. First,
an expert review study was conducted to evaluate the initial
version of the framework from the perspective of academic
experts. The changes recommended by the experts were
13

752 Universal Access in the Information Society (2020) 19:739–781

Table 7  IPO-based serious game design frameworks

Phases of IPO Serious game design framework proposed in this research Serious game design framework by Garris and Ahlers [52]

Input Information gathered Information gathered

Intervention of behaviours related to learning of vocabulary Contents to be learned
Objectives to be achieved Game-based learning attributes to be incorporated
Contents to be learned Outcome
Capabilities to be developed Contents and game-based learning attributes to create game cycle
Outcome
Objectives to be achieved in the learning activities
Process Information gathered Information gathered
Learning activities to be developed Game cycle based on user judgment, user behaviour and system
Components that constitute each learning activity feedback
Game-based learning attributes to be incorporated into the Outcome
game; motivation and other elements that drive the process Game cycle
of Garris are part of it
Outcome
Learning activities
Output Information gathered Information gathered
User profile to save performance of learning activities How to provide debriefing
How to provide debriefing How to validate performance against learning outcomes of game
How to validate performance against objectives of game Final outcome
Final outcome Design of a game for typical users
Design of a serious game to provide intervention of identi-
fied behaviours related to learning of vocabulary among
children with ASD

Fig. 8  Methods and flow of evaluations of the proposed framework

incorporated into the framework, leading to an improved
version of the proposed framework. Two studies were con-
ducted to evaluate the framework from the collective view-
point of two different types of experts, including research-
ers working on serious games and game designers. First, an
expert review study was conducted; second, an applicability
test was conducted to assess the application of the frame-
work followed by a survey to gather feedback regarding the
applicability of the framework. The changes recommended
by the experts from the first study were incorporated to pro-
duce a modified version of the framework. This version of
the framework was used in the second study as a part of its
applicability. The proposed version of the framework was
produced based on the comments provided by experts in the
second study.
Tables 8, 9 and 10 present the goal, question, metric
(GQM) statements to perform the expert review studies,
and applicability test.
6 Expert evaluation 1 with academic
experts
This section describes the recruitment process of the partici-
pants, the instruments used, the reviews procedures carried
out and lastly the analysis of results, leading to an improved
version of the proposed framework.

13
Universal Access in the Information Society (2020) 19:739–781 753

Table 8  GQM statement to perform expert evaluation with academic experts

To analyse the components used, their usefulness in the framework, relations between components, structure of the proposed
framework
In order to evaluate the framework to design serious game for the vocabulary acquisition of children with ASD
From the perspective of the academic experts working in the field of serious games
In the context of face-to-face discussion and survey-based questionnaire
Because these experts either design, use or evaluate such frameworks on a regular basis. Therefore, their expert opinion is used
to improve the proposed framework

Table 9  GQM statement to perform an expert evaluation with the game designers or researchers

To analyse the components used

In order to evaluate the framework to design serious game for the vocabulary acquisition of children with ASD
From the perspective of the two types of experts including game designers or researchers working in the field of serious games
In the context of face-to-face discussion and survey-based questionnaire
Because these experts are the ones who work with these components during the design of serious games in real
life. Therefore, their input is taken to improve the proposed framework

Table 10  GQM statement to perform the applicability test with the game designers or researchers

To analyse the application of the serious game design framework to design serious game for the vocabulary acquisition of children
with ASD
In order to assess With respect to
Understandability
Learnability
Adaptability
Usability
Memory load
Expectations
Usefulness
Complexity
Development
Willingness
From the perspective of the game designers or researchers working in the field of serious games
In the context of survey-based questionnaire
Because the practical use of the framework to produce serious game design by the potential users can help to improve the
framework in terms of aspects assessed in the study

6.1 Participants and recruitment in the study. The demographic information of participating

Academic experts working in the area of serious games
were invited to take part in this study. The selection of
these experts was carefully done using Google Scholar1 and
ResearchGate2, and the Google3 search engine. The experts
were also asked to provide recommendations of other
experts to invite them to take part in this evaluation. The
invitation to take part in the expert review study was sent to
a number of participants from which 9 experts showed their
willingness through positive acknowledgement and took part
1
  https​://schol​ar.googl​e.com.
2
  https​://www.resea​rchga​te.net.
3
  https​://www.googl​e.com.
experts is shown in Table 11.
6.2 Instruments used
A questionnaire consisting of four parts and comprising a
combination of closed-end and open-ended questions was
prepared to gather feedback from the experts. The purpose
of each part and type of questions asked in it are described
in Table  12, while the structure of the questionnaire is
shown in Fig. 9, and the complete questionnaire is avail-
able in "Appendix". All closed-end questions in the survey
are based on 3-point Likert scale (‘disagree’, ‘not sure’ and
‘agree’).

13

754 Universal Access in the Information Society (2020) 19:739–781

Table 11  Academic experts’ demographic information

Id Gender University Country Research interest Experience

1 Female Murdoch University Australia Educational technology 20 years

2 Male Curtin University Australia Game-based learning, complex systems, digital 30+ years
media learning and psychometrics
3 Male VU University Amsterdam The Netherlands multimedia and game technology, interactive 20+ years
video, serious games
4 Male The Serious Games Institute, Coven- The UK Serious games and real and virtual educational 9 years
try University, Coventry robotics for secondary and tertiary mathemat-
ics education
5 Male NHL University of Applied Sciences The Netherlands Serious gaming a systemic approach; health and 4 years in serious
wellbeing; vocational safety; social innovation gaming; 20+ years
in research
6 Male Charles University in Prague Czech Republic Game-based learning; artificial intelligence for 10 years
computer games; computational ethology;
computational neurobiology
7 Male RMIT Europe Spain Game design, user experience, play 15 years
8 Male Coventry University UK Games Science (serious games, gamification), 10 years
pervasive learning, simulation
9 Male University of Ottawa Canada Gaming systems and networks 30 years

Table 12  Details of parts and questions in the questionnaire

Part Purpose Questions

Part A Heading: Importance, description and details of components, and their adequacy Closed-end questions and comments/suggestions
(missing, removing) of components used in the proposed framework
Details To get feedback on the importance of each component in the game, determine
whether the description is readable and the details in description are sufficient or
require more information to be added and whether there are components that need
to be removed from or added to the framework
Part B Heading: Structure of the framework Closed-end questions and comments/suggestions
Details To validate the logical division of components to the three phases corre-
sponding to input, process and output of IPO and the structure of the framework
Part C Heading: Relations between components Closed-end and open-ended questions
Details To ascertain whether the relations between components are logical or need
to be fixed and to determine the missing relations (if any) so that framework can be
improved accordingly
Part D Heading: Comprehensiveness of the framework Open-ended question
Details To get suggestions or recommendations for the improvement in the frame-
work

The first level in Fig. 9 shows the shorter name from the
headings of four parts, i.e. part A to part D, mentioned in
Table 12, while from the second level onwards the questions
and sub-questions in each part are shown. Each part contains
at least one qualitative question for an expert to provide their
descriptive responses regarding questions asked in that part.
Part A contains the same set of questions (Q1, Q1a to
Q1c and Q2) for each component of the framework; the
twenty-one components of the framework are represented
through numbers from Co1, to Co21. The question number
in the plain text indicates that it is a quantitative question,
while the bold and underlined question number represents a
qualitative question. Altogether, this part contains 84 quan-
titative and 23 qualitative questions.
Part B comprises four sub-parts; input, process and output
are related to the phases of the framework, while IPO cor-
responds to the structure used in the framework. Altogether,
this part contains 23 quantitative and 3 qualitative questions.
Part C consists of 6 quantitative and 3 qualitative ques-
tions related to the relationships between components.
Part D is related to the comprehensiveness of the frame-
work and consists of just one qualitative question.

13
Universal Access in the Information Society (2020) 19:739–781 755
Fig. 9  Structure of the survey questionnaire
6.3 Study protocols
Each expert was independently invited via an email to take
part in this review study. In the first invitational email, each
expert was briefed about the objective of the study and also
to get their consent to be part of the study. The experts who
positively acknowledged the invitations were sent two files
through an email to perform the evaluation: the first file
included descriptions of the framework all the components,
while the survey questionnaire was included in the second
file. An online Skype session was conducted with the par-
ticipants to brief them about the background of the study, its
objectives, an introduction to the SGDF and its components,
and answer any queries related to the framework or the sur-
vey questionnaire sent earlier. There were certain cases in
which it was not possible to hold a Skype session for reasons
such as tight schedule of an expert, restriction to use Skype
in the premises of their university or time differences. In
such cases, experts were asked to email details of any query
related to the framework or survey questionnaire.
6.4 Data analysis
6.4.1 Quantitative analysis of the frequencies of opinions
The data provided by the experts in response to closed-end
questions were analysed using the frequency of responses
in relation to the 8 questions and their sub-questions (Q1,
Q1A–Q1C, Q4A–Q4D, Q5A–Q5N, Q6A–Q6B, Q7, Q8, Q9,
Q10A–Q10F) of Parts A, B and C as shown in Fig. 9. The
frequency for these questions and sub-questions was calcu-
lated as a cumulative sum of responses for ‘disagree’, ‘not
sure’ and ‘agree’.
6.4.2 Qualitative analysis of the descriptive response
Seaman [74] presented several qualitative methods for data
collection and analysis. One of the methods explained by
Seaman for the analysis of qualitative data is to identify
categories that can help in the classification of descriptive
responses by the experts for further analysis. This technique
is also used in this review study for the analysis of responses
provided by the experts. Each response provided by the
13

756 Universal Access in the Information Society (2020) 19:739–781

Fig. 10  Part A—cumulative
frequency of responses for the
components in the input phase

Fig. 11  Part A—cumulative frequency of responses for the components in the process phase

experts in 7 questions and sub-questions Q2, Q3a–Q3b, Q4e,

Q5o, Q6c, Q10g, Q11a–Q11b, Q12 of Parts A, B, C and D
as shown in Fig. 9 is classified into four categories: sugges-
tion, concern, agreement or disagreement. The descriptions
of these sub-categories are as follows: (1) suggestion is some
sort of action required to make necessary changes in the
proposed framework; (2) concern by an expert that require
attention; (3) agreement is an acceptance of the game com-
ponents or connections between them which does not require
any change to be made in the SGDF; and lastly (4) disagree-
ment is the concern raised by an expert on the SGDF and
require attention.
Fig. 12  Part A—cumulative frequency of responses for the compo-
6.5 Results nents in the output phase

6.5.1 Frequency of the opinions provided by the experts

This section presents the results of the quantitative ques-
tions (Sect. 6.4.1) in the form of cumulative frequencies
of responses. The responses for components in the input,
process and output phases of the framework from the ques-
tions of Part A are shown in Figs. 10, 11 and 12, respec-
tively. For each component, four results are shown: (1) ‘I’
for the importance of component from Q1, (2) ‘U’ for use
of component in design from Q1a, (3) ‘R’ for readability
of description from Q1b and (4) ‘D’ for sufficient details in
the description from Q1c. Each figure shows 3 lines. A red
line or line with triangle markers indicates the number of
experts who selected ‘disagree’ as their answer to the ques-
tion; the experts who mentioned ‘not sure’ as an answer are
represented by a blue line or line with square markers, while
a green line or line with circle markers represents a number
of experts who responded ‘agree’ to question. The cumula-
tive sum could indicate that most of the experts (1) agreed
with the information asked in the question, (2) disagreed
with the information asked in the question, (3) were not

13
Universal Access in the Information Society (2020) 19:739–781 757

Fig. 13  Cumulative frequency of responses for the placement of the components in different phases

sure about the information asked in the question or (4) had

a mixed opinion about the information asked in the ques-
tion which means the number of the experts who selected
‘agree’ is almost the same as of the number of the experts
who selected either ‘disagree’ or ‘not sure’.

6.5.1.1  Part A—components of  the  framework Figure  10

shows that all four components in the input phase are impor-
tant and useful and that their description is readable, while
for the details in the description of the component, mixed
opinions were found among experts. This indicates that for
some experts, the description of components has sufficient
detail, while for others more details need to be added in the
description.
Figure 11 shows that all components in the process phase
are important and useful and that their description is read-
able, while for the details in the description of the compo-
nent, a mixed opinion of experts was found.
Figure 12 shows that all components in the output phase
are important, useful and readable, and have sufficient detail.
6.5.1.2  Part B—placement of  components in  phases
and IPO structure for the framework Figure  13 shows the
cumulative frequency of responses for placement of compo-
nents in different phases from Q4 to Q6 of Part B. The opin-
ion of experts shows that distributions of all components in
input, process and output phases are logical.
Figure 14 shows the cumulative frequency of responses
for the IPO structure of framework from Q7 and Q8 of Part
B4. The expert opinion for Q7 shows that IPO-based struc-
ture is easy to understand and there is a similarity between
structures of IPO and typical games.
Fig. 14  Cumulative frequency of responses for the IPO structure of
the framework
6.5.1.3  Part C—relationships between  components Fig-
ure  15 shows the cumulative frequency of responses for
relationships between components from Q10 of part C.
Most of the relationships are correct based on the expert
opinion; the relationships: (1) learning outcomes leads to
instruction contents (LO → IC) and (2) game-based learning
attributes are based on the theories (GBLAs → Theories),
have received mixed opinion from the experts.
6.5.1.4  Descriptive responses provided by  the  experts A
total of 94 descriptive responses were collected from the
feedback of experts. Table  13 shows the distribution of
responses into an agreement, disagreement, concern and
suggestion.
N = 10 (11%) of the overall descriptive responses cor-
respond to an agreement; these responses are related to the

13

758 Universal Access in the Information Society (2020) 19:739–781

Fig. 15  Cumulative frequency of responses for relationships between the components

Table 13  Frequency of responses classified into five categories asso- 6.6 Changes in the initial version of the proposed
ciated with four parts framework
Agreement Disa- Concern Suggestion Total
gree- The analysis from the quantitative and qualitative responses
ment revealed three types of changes that need to be incorporated
Part A 7 2 19 31 59 into the framework to produce an improved version of the
Part B 1 2 8 1 12 proposed framework. These changes are about (1) details of
Part C 1 1 7 2 11 the components, (2) new components to be added and (3)
Part D 1 0 9 2 12 relationships among components.
Total 10 5 43 36 94
6.6.1 Details of the components

importance and usefulness of components, relationships
between components and the comprehensiveness of the
framework.
N = 5 (5%) of responses indicate the disagreement by the
experts regarding information asked in the question. Disa-
greement is mainly about details in the description com-
ponent ‘characters’, placement of components ‘user profile
and achievements’ in the output phase than the input phase
and the lack of explanation about the relationships between
components.
N = 43 (46%) of the responses involve concerns raised
by the experts, and half of these concerns highlight the fact
that some of the components need additional information in
the description to make it easy to read, understand and fol-
low. For instance, a component named ‘autism behaviours’
provides details about the components but does not indicate
which behaviours are supported by the framework and how
each behaviour can be used in the design.
Similarly, N = 36 (38%) of the responses are in the form
of suggestions provided by the experts. These suggestions
are related to adding more detail, providing some examples
and also adding references in the components, relationships
between components and a sample prototype developed
based on the proposed framework.
Table 14 shows the samples of the experts’ responses for
agreement, disagreement, concern or suggestion.
The following changes have been incorporated into the
components:
(1) An example of each type of learning outcome has been
added in the description of component ‘learning out-
come’;
(2) A description of debriefing has been added;
(3) Examples of instructional contents related to vocabu-
lary have been added in the description of the compo-
nent;
(4) Autism behaviours: References have been added in the
description for further information on the behaviours
that can be used for the intervention. The name of the
component is renamed to ‘autism behaviours and symp-
toms’;
(5) Instruction method: References have been added in
the description for further information on the standard
methods for ASD that can be used for the intervention;
(6) Strategies: References have been added in the descrip-
tion for further information on the strategies that can
be used in the serious game for the intervention;
(7) The name of component ‘game attribute’ has been
renamed to ‘game-based learning attributes (GBLAs)’.
6.6.2 New component
A new component ‘context’ is added in the learning activity.
A description of this component is given below based on the
research by de Freitas and Jarvis [75].

13
Universal Access in the Information Society (2020) 19:739–781 759

Context The context of any serious game includes infor-

component would need a full listing of the key

and a reference to a more complete listing and

examples of what it means for design practice

references for a game developer to follow and

clear, a game prototype is needed to establish

behaviours or a listing of behaviour domains
Autism behaviours To be useful in design, the
mation about the three aspects related to the game, namely

Reflection Give an example to make it clearer

the effectivity and coherence of the compo-

Instruction methods Again, give some main

… the theoretical framework is sufficiently

access, technical support and place. Access is about the

learn what are these methods typically?

identification of the location where the game will be used,
determining who can access the serious game. If it is to be
used in the classroom environment, then all the students of
the class can use the game; however, if it is to be used some-
where outside the classroom, then it is available to the wider
range of audience than just students of that class. The tech-
nical support is about determining whether assistance can
be provided to the users of the game if they are stuck in the
Suggestion

nents …
middle of the game. Lastly, the place is to determine whether
the location where the game would be used is suitable for
learning? Is it a quiet place for children to concentrate, play
and its instructors. I guess this is particularly
description seems rather limited at this stage
spectrum? Could this be made more specific

and learn, or an outside environment where it may be dif-

teristics? And are you referring to a specific

Learning activity Contextualisation is impor-

what extent behaviours have to be captured

one devices using the framework should fit
‘Autism behaviours’ is a very generic term. I

true for children with ASD. I am unsure to

the urgency and practice of its participants
The methodology, gaming or otherwise that
assume it refers to autistic behaviours and
symptoms or behavioural traits or charac-

ficult for the user to concentrate?

Instructional contents clear, but from the

6.6.3 Relationships between components
tant for all game play activities

The relationship of components ‘Game-based learning

attributes → Theories’ was changed to ‘Game-based learning
attributes ← Theories’ to indicate that game-based learning
attributes are based on the different theories of learning and
psychology theory.
perhaps?
Concern

7 Expert evaluation 2 with researchers

and serious games designers
Reflection A very worthwhile component and Character Not necessarily controlled by AI—

and is prior knowledge to the game play…

the behaviour can be fixed for, e.g. NPCs,

the learning activity to user achievements

Characters Part of game design, typically hid- For the relations where I disagree it is very

with the comment that no attention is paid to insufficient. For example, the relations of
makes sense because the descriptions are
User profile is usual input at the beginning

should be explained in much more detail

involving with a fixed dialogue structure

difficult to assess whether the relation

Expert evaluation 2 was carried out by two different types

of experts, i.e. researchers working on serious games and
game designers. The instruments used, reviews procedures
and data analysis methods for quantitative and qualitative are
the same as described in Sect. 3; hence, they are not being
described again in this section. The review of the frame-
during the game

work for this study mainly focuses on the components in the

not always…
Disagreement

input, process and output phases of the framework, i.e. the

twenty-one components as shown in Part A of Fig. 2. Thus,
it consists of 84 quantitative and 23 qualitative questions.

7.1 Participants and recruitment
Table 14  Sample responses from the experts

(again) reflection can also be promoted by

it is good that multimedia instructions are

playing, e.g. a summary animation of the

… the framework is fairly comprehensive,

Instruction method No doubt relevant and

Researchers working in the field of serious games or game

design were invited to part in this study. They were invited
session that was just finished…

by posting information on Facebook pages, discussion

forums and mailing lists related to game design. Seven par-
the aesthetics of the game!
den from gaming design.

ticipants volunteered to take part in the evaluation. They

were also asked to provide recommendations of other experts
Coverage seems fine

to be invited to participate. The demographic information of

incorporated!

experts involved in the evaluation is shown in Table 15.

Agreement

13

760 Universal Access in the Information Society (2020) 19:739–781

Table 15  Experts’ demographic information useful, while details are sufficient and readable. This can
Id Gender Country Role Experience be seen through the green line or line with circle markers
for which the value of each marker is at least 4, which indi-
1 Male Australia Game designer 4 cates that more than 50% of the respondents selected ‘agree’.
2 Male Pakistan Game designer 2 These were the technical people who are going to be design-
3 Male Malaysia Game designer 5 ing the game, and according to them, all components are
4 Male Malaysia Game designer 4 very much needed in game design.
5 Male Canada Researcher 4
6 Male Malaysia Researcher 3 7.2.2 Descriptive responses provided by the researchers
7 Female United Arab Researcher 3 and game designers
Emirates

A total of 56 descriptive responses were collected from

7.2 Results
7.2.1 Frequency of opinions provided by the game
designers and researchers
This section presents the cumulative frequencies of
responses for components in the input, process and output
phases of the framework, and they are presented in Figs. 16,
17 and 18, respectively. The results in the figures show that
responses across the 7 experts are consistent and all com-
ponents in input, process and output phases are important,
the feedback of game designers and researchers, including
agreement, concerns and suggestions. No disagreement was
found.
N = 12, 21% of the overall descriptive responses corre-
spond to an agreement; these responses are related to having
sufficient detail and the way these details are presented in the
descriptions of components.
N = 15, 27% of responses indicate the concerns by the
experts regarding information asked in the question. These
concerns are related to (1) providing more details on the
ASD related components (autism behaviour, instruc-
tion methods, strategies) and GBLAs; (2) similarities and

Fig. 16  Cumulative frequency of responses for the components in the input phase

Fig. 17  Cumulative frequency of responses for the components in the process phase

13
Universal Access in the Information Society (2020) 19:739–781 761
Fig. 18  Cumulative frequency of responses for the components in the
output phase
differences between some component like GBLAs and strat-
egies, GBLAs and game mechanics, learning outcomes and
capabilities, and reflection and debriefing; (3) use of game
dynamics; and (4) the purpose of storytelling.
N = 29, 52% of responses indicate the suggestions by the
experts regarding information asked in the question. Sugges-
tions include the following: provide examples in components
like autism behaviours and symptoms, intended learning out-
comes, instruction method, using a more descriptive name of
components such as capabilities and strategies, and reducing
the number of components by merging similar components
together.
7.3 Changes incorporated in the framework
Analysis from the quantitative and qualitative responses
revealed that more details of the components need to be
incorporated in the framework. The following is a list of
changes that have been added:
(1) Autism behaviours Examples have been added in the
description to show behaviours related to vocabulary
learning that can be used for the intervention;
(2) Instruction method Examples have been added in the
description to show instruction methods for ASD that
can be used for the intervention;
(3) GBLAs examples have been added in the description to
show attributes that can be used for the intervention;
(4) The component ‘capabilities’ is renamed to ‘Desirable
capabilities’;
(5) The component ‘strategies’ is renamed to ‘Instructional
strategies’;
(6) The descriptions of components, namely GBLAs, strat-
egies, game mechanics, learning outcomes, capabili-
ties, reflection and debriefing, has been edited to ensure
that they all are difficult;
(7) The use of the component ‘game dynamics’ has been
clarified in the description of the component;
(8) The purpose of ‘storytelling’ has been clarified in the
description of the component.
8 Evaluation of serious game design
framework—applicability test followed
by applicability survey
This section describes the applicability evaluation conducted
with the same experts from study two (Sect. 7). It begins by
providing information on the instruments used and proce-
dures carried out. Finally, the analysis of the results is pre-
sented, leading to the revision of the proposed framework.
The methods used to analyse quantitative and qualitative
responses are the same as of methods used in studies 1 and
2 (Sects. 6, 7).
8.1 Instruments used
The instruments used throughout the evaluation are briefly
explained below.
8.1.1 Framework document
This document contains a revised version of the proposed
framework which has been updated after conducting two
expert review studies.
8.1.2 Game design document template
One of the responsibilities of game designers as a part of
their job is to produce a series of documents including
game design, characters design and level design documents
among others to tell others about their game design. In this
research, the focus is on a serious game design document
as an end result of using the proposed framework; the con-
tents and information presented in this design document vary
from company to company, project to project or designer to
designer. However, it does follow a common thread. Com-
mon information presented in such documents includes:
• An overview of the game to be developed:
• Title of the game;
• Description of the game;
• Target audience;
• Age group;
• The platform on which it would work;
• Number of players that can simultaneously play the
game;
13

762 Universal Access in the Information Society (2020) 19:739–781

Table 16  Categories, criteria and questions used in the questionnaire

Category Criteria Q#

Understandability Able to understand the components and phases 1

Learnability Able to learn the framework and its components 2
Adaptability Able to adapt the framework and design games for these children to learn vocabulary 3
Usability Able to use the framework and its components to produce design of serious game 4
Memory load Able to design serious game with less stress and efforts 5
Expectations Able to meet all the expectations associated with the game design 6
Usefulness Able to fulfil the needs of game design 7
Complexity Ease with which the framework and its components can be used 8
Development Able to develop a game based on the design generated through the framework 9, 10
Willingness/future use Able to use the framework and design games for these children to learn vocabulary in future 11

• Specifications of the game components used in the design
of serious games.
The same pattern of common information is also used in
the template of the document that was sent to the experts;
the template consists of two parts: Part A gives an overview
of the serious games to be designed for children with ASD
for learning vocabulary and Part B divides the specification
of the game components divided according to the phases of
the framework, i.e. input, process and output.
8.1.3 Applicability survey
The questionnaire consists of ten different categories; each
category contains one or more questions. The first ten cat-
egories are based on the quality model by ISO/IEC 25010—
Systems and software engineering—Systems and software
Quality Requirements and Evaluation (SQuaRE)—System
and software quality models ISO/IEC [76] and Purdue usa-
bility testing questionnaire (PUTQ) by Lin et al. [77]. These
categories, their associated criteria and the question numbers
that fall in each category are presented in Table 16.
The list of questions, indicated as ‘QS#’ in the above
table, are as follows:
Q1. Is the framework step-by-step understandable follow-
ing phases and components which are part of it?
Q2. Is the framework easy to learn?
Q3. Is the framework easy to adapt and design serious
game?
Q4. Is the framework easy to use?
Q5. Is using framework require minimal memory load to
design serious game?
Q6. Did the framework meet your expectation to produce
the design of the serious game?
Q7. Is the framework useful to design serious game for
these children to learn vocabulary?
Q8. Is the framework less complex to use?
Q9. Do you think the information presented in your
design document can provide insight into the game to the
members of the team?
Q10. Do you think the information presented in your
design document can be used to develop the game?
Q11. Would you be willing to use the framework and
design serious game for these children to learn vocabulary?
8.2 Study protocol
The experts were sent two instruments: (1) the framework
document and (2) the game design document. They were
asked to work on the game design document and send the
same once they have provided all the necessary details in the
document. Once the document was received, they were sent
another email with one attachment of applicability survey
instrument and were asked to revert back with the completed
survey. An email of appreciation was lastly sent for sparing
their valuable time to be part of both studies (studies 2 and
3) and providing feedback to improve the framework.
8.3 Results
This section presents the results of the applicability test per-
formed with the experts.
8.3.1 Game design documents
Each design document submitted by the experts was indi-
vidually analysed in terms of specifications of components
provided by the experts; this can highlight the most com-
monly used components and the least commonly used com-
ponents across the designs. Two experts were working at the
same place so they produced and submitted one design docu-
ment together. Table 17 provides a summary of the specifi-
cations for each component provided by the experts in the
designs. The components are divided into two categories.
The components specific to ASD are categorised as ASD

13
Universal Access in the Information Society (2020) 19:739–781 763
components, while the remaining components are catego-
rised as game components (GCs). The specifications of all
the components are presented in the third to eight columns.
ASD components
All the identified behaviours have been targeted in differ-
ent designs by the experts though receptive and expressive
skills are the most commonly used. Two instructional meth-
ods have been used, namely discrete trial training (DTT)
and incidental teaching, of which DTT has been used in
almost all the designs. Different strategies have been used
by the experts; however, multimedia instruction and explicit
instruction are the two most commonly used. The experts
have used basic modalities in the designs.
Game components (GCs)
The experts used cognitive, psychomotor and affective
skills across the design documents. The learning outcomes
have been defined based on the instructional contents and
learning outcomes. The context targeted by the experts
includes both the classroom and also general, which includes
classroom, home or other settings. The game genres used
across the designs were educational and simulation among
others. The educational genre was most commonly used.
Each game design consists of a number of game mechan-
ics and game dynamics. Each design uses different GBLAs
from each of learning and psychology theories. The design-
ers used various attributes in the user profile; one commonly
used attribute among these designs is the real or nickname
of the child. The game story is used in three designs. Each
story in these designs consists of one or more of the com-
ponents and is used to tell a part of the overall story to the
user. The storytelling and narratives are individually used in
one design each; they are the minimally used components
across the designs. The characters are used in four designs;
among these designs, experts have used animated characters
for the interaction and to provide support to the child, while
they are playing the game. The recording of information
related to user achievements is described across the design
and debriefing and also used in the games.
8.3.1.1  Specifications of  one game design document The
details of a game design document produced by one designer
based on the proposed framework are described in the fol-
lowing.
Part B: Specification of game components according to
IPO format
Part B-1: Game components from the input phase:
(1) Autism behaviours and symptoms One common behav-
iour, i.e. ‘struggle to receptively identify objects’
among these children is targeted in the design of the
serious game;
(2) Desirable capabilities By the end of the game, it is
expected that the child can identify the objects when
they listen to the names of objects and recall the same
objects over the period of time;
(3) Intended learning outcomes The specific learning out-
comes to be achieved in these children by the end of
this game are: (1) identifying objects and (2) recalling
objects over a period of time;
(4) Instructional contents The extensive learning of content
to be provided in the game includes fruits, birds, foods,
numbers and the alphabet.
Part B-2: Game components from the process phase:
(5) Reflection This informs the child about the purpose of
undergoing any activity, rectifies errors made during
activity and provides necessary suggestions to over-
come these problems in future;
(6) Instruction method The instruction method to be used
in the game is discrete trial training (DTT);
(7) Instructional strategies The specific strategies of
explicit instruction and multimedia instruction used
in the game are described in Table 18;
(8) Game genre This is an educational game which pro-
vides an interactive learning environment for different
types of vocabulary items;
(9) Game mechanics and dynamics The game mechanics
and the game dynamics associated with this game are
mentioned in Table 19;
(10) Storytelling The story is centred on a child who faces
difficulty in the identification of the object when the
name of the object is called;
(11) Narrative no narratives;
(12) Non-player characters (NPC) One helping character is
provided in the game for the interaction with the child.
They can interact with the character to seek guidance
when they are not sure what to do next. This character
also monitors the game activities and communicates
with the child accordingly; for instance, if it does not
recognise any activity (action performed through
mouse) in 10 s, then an auto-popup in the middle of
screen will ask the child whether everything is alright
or whether they are stuck and looking for help;
(13) Game-based learning attributes A subset of game-
based learning attributes used in the game is presented
in Table 20 along with their underlying theories;
(14) Game story The design contains one NPC to provide
support to the user throughout the game and using the
user’s perspective of the storytelling;
(15) Modalities The modalities incorporated in the game
include text, images, animated character, animation
and sound.
13


Table 17  Summary of specifications in game design documents

764

Type Components Design 1 Design 2 Design 3 Design 4 Design 5 Design 6

13
ASD components Autism behaviours Receptive skills, con- Difficulty in identify- Receptive and expres- Receptive and expres- All behaviours Receptive and expres-
and symptoms tent and grammar is ing nouns, inappro- sive skill sive skill sive skill, difficulty
delayed, rigid under- priate use of verbs in identifying nouns,
standing of words reverse pronouns,
inappropriate use of
verbs
Instruction method DTT DTT DTT Incidental teaching DTT DTT
Instructional strate- Explicit, multimedia Multimedia Explicit, multimedia Explicit, multimedia Explicit, multimedia, Multimedia, associa-
gies capacity, association tion
Modalities Basic Basic Basic Basic Basic Basic
Game components Desirable capabilities Cognitive Cognitive, affective Affective Cognitive, psycho- Cognitive, psychomo- Cognitive, affective
(GCs) motor tor, affective
Intended learning ✓ ✓ ✓ ✓ ✓ ✓
outcomes
Instructional contents Animals, birds, fruits, Noun, verbs Numbers, alphabets, Quran Various ­contents* Noun, pronouns, verbs,
vegetables shapes, colours, people, vehicle, bath-
shapes and colours room items, kitchen
items
Context Classroom General Classroom Classroom, general General Classroom, general
Game genre Educational Educational Simulation Educational Multiple Educational
Game mechanics Multiple Multiple Multiple Multiple Multiple Multiple
Game dynamics Multiple Multiple Multiple Multiple Multiple Multiple
GBLAs (behaviour- FDK, PNP, RWD, PND, RWD, PFL FDK, RWD, IEP FDK, RNG, RNG RWD, IEQ, CNF, PND, RWD, PFL
ism) RNG LNC, PND
GBLAs (cognitivism) MTV INL, RNG, ATS, MTV, ATS MTV. INL, TLS INL, LNR, ATS, SCF, INL, RNG, ATS, MTV
MTV TLS
GBLAs (constructiv- PRG, ADP LCN, PRG SUR, SCF LCN, PRG, SCF LCN, LOC, SUR, LCN, PRG
ist) ADP, PRG
GBLAs (psychology) ASS, SAL CHL, FDK, ASS, IFD, CHL ALS, ASS, FDK, ASS, CHL, ALS, CHL, FDK, ASS, SAL
SAL CHL REP, PNP, FAN,
MYS
Game story ✓ ✓ ✓
Storytelling ✓
Narratives ✓
Non-player characters ✓ ✓ ✓ ✓
Universal Access in the Information Society (2020) 19:739–781
 Universal Access in the Information Society (2020) 19:739–781 765

Part B-3: Game components from the output phase:

tion (interpersonal), ISC interaction (social), LNC: language/communication, LOC location, MYS mystery, PNP pieces or players, PRG progress, SUR surprise, REP representation, RNG rules/
INL incremental learning, LNR linearity, ATS attention span, SCF scaffolding, TLS transfer of learnt skills, LCN learner control, PND practice and drill, FDK feedback, RWD rewards, SAL situ-
ated and authentic learning, ALS accommodating to the learner’s styles, ADP adaptation, ASS assessment, CHL challenge, CNF conflict, FAN fantasy, IEQ interaction (equipment), IIN interac-
Name, gender, date of
User profile and achievements: Each user profile includes
name, gender, email address, date of birth and all the
birth, image achievements made in the game.
Design 6

8.3.1.2  Mapping of components on to the prototype Fig-

✓
ure 19 shows different screens of the prototype which have
✓
been created based on the above design document. Each
Name, gender, email

birth, phone#, par-

screen has used one or more components from the frame-

ents information
address, date of

work; the number in each circle corresponds to one of the

components presented in Table  21, while the number in
Design 5

each rounded rectangle represents the screen number.

✓
✓

8.4 Applicability survey

Figure 20 presents the cumulative frequency of responses

of the applicability survey carried out by the experts after
Name, image

carrying out the applicability test of the framework (i.e. after

Design 4

they have experienced the framework and produced design

document).
✓
✓

It can be seen that all the experts have mentioned that

the framework is useful and has met their expectations.
goals, SFT safety, SNS sensory stimuli, PFL positive feelings, INT intensity, MTV motivation, HNS help and support

According to N = 5 of 7 experts, the framework and its com-

Name, image

ponents are easy to understand, learn, adapt and use; further,

the design documents generated can be used to develop a
Design 3

prototype and they are also willing to use the framework

✓
✓

in the future as well. The framework is also less complex

as mentioned by N = 5 of 7 experts. A mixed opinion was
Name, gender, email

recorded regarding the use of framework require minimal

address, date of

memory; N = 3 of 7 mentioned that it would involve a mini-

mal workload, while the same number of experts were not
Design 2

sure whether it would require less or more workload.

birth

✓
✓

8.5 Changes incorporated in the framework

Figure 21 shows the revised version of the framework based

on the changes incorporated from Sects. 6.6 and 7.3. The
Nickname

changes incorporated in the revised version of the frame-

Design 1

work include the addition of a new component called ‘con-

text’ which is highlighted with a red dash dot bounded rec-
✓
✓

tangle and the renaming of components highlighted with a

red solid dash bounded rectangle. To reduce the redundancy
User achievements

and confusion of the reader, the name and the description

Components

of all the components presented in Sect. 3 already reflect

User profile

Debriefing

the changes recommended by the experts in the studies pre-

Basic: text, images, audio, video

sented in Sects. 6, 7 and 8.

Table 17  (continued)

9 Evaluation of a serious game prototype

The game design document presented in Sect. 8.3.1.1 was

transformed into a prototype shown in Sect. 8.3.1.2. The
Type

detailed information on the development and evaluation of a

13

766 Universal Access in the Information Society (2020) 19:739–781

Table 18  Specific strategies
Explicit instruction Multimedia instruction

Students are given a definition or other attributes associated with word Children are taught vocabulary by going beyond text and use other
to be learned media, for instance, still and animated images, sound and animated
Repeated and multiple exposures of words lead to gain in vocabulary characters, among others
Pictures, stimulus materials or instructional examples used in teacher- Children who cannot read yet will associate printed words with spoken
directed instruction are used in the game. words; Furthermore, autistic children sometimes have difficulty
Pictures, stimulus materials or instructional examples found in the age- processing spoken words, especially those from TVs, and children
appropriate general education classroom are used in the game who can read may benefit from being able to see the words as well as
hear them
Instructions are provided in the form of written words, pictures where
needed and the verbal instructions are also simultaneously played to
help them in understanding; these instructions are provided in small
steps

Table 19  Game mechanics and Game mechanics Game type Game dynamics

game dynamics
Aim and select Easy If correct object is identified
Name of the object is shown underneath object
Name of the object is played for child to listen and memorise
Increase points by 10
If incorrect object is identified
Highlight all the correct objects present on the screen for 1 s to provide hint
Medium If correct object is identified
Name of the object is shown underneath object
Name of the object is played for child to listen and memorise
Increase points by 10
If incorrect object is identified
Highlight one of the correct objects present on the screen random for 1 s to
provide hint
Decrease points by 5
Hard If correct object is identified
Increase points by 10
If incorrect object is identified
No hint is provided like other levels
Decrease points by 5
Decrease time by 5 s

Table 20  Game-based learning Behaviourism Cognitivism Constructivist Psychology

attributes used in serious game
design 1. Reward 6. Incremental learning 10. Scaffolding 14. Situated and authentic learning
2. Practice and drill 7. Attention span 11. Learner control 15. Intermittent feedback
3. Feedback 8. Transfer of learnt skill 12. Progress 16. Assessment
4. Rules/goals 9. Motivation 13. Surprise 17. Challenge
5. Positive feelings

serious game prototype is discussed in other research papers
[18, 19, 78] though the gist of the evaluation is provided
here. The effectiveness of the prototype was examined using
experimental evaluation based on the single-subject research
design (SSRD), the aim of which is to measure the improve-
ment in the receptive identification of vocabulary items
among autistic children over the period of time, i.e. before
and after using the prototype. The CADS Enhancement
Centre which is located in Kuala Lumpur, Malaysia, was
invited to take part in the evaluation. Five participants with
difficulties in learning different categories of vocabulary
were recruited through the help of the teacher nominated
by the principal of the centre. Parental consent was sought
in writing before the recruited participants were involved
in the evaluation of the serious game prototype. The paren-
tal consent was prepared in consultation with the princi-
pal and sent to each parent with the recruited participant
and also received from them. The parents of each recruited

13
Universal Access in the Information Society (2020) 19:739–781 767

Fig. 19  Mapping of the com-

ponents from the serious game
design to the serious game
prototype

Table 21  Components of serious game design framework proposed in evaluation of the prototype, one category of vocabulary, i.e.
this research birds, was identified with the help of a teacher working with
S# Component S# Component these children. The category was identified keeping a view in
mind that the recruited participants have very minimal or no
1 Autism behaviours and symptoms 11 Game dynamics knowledge of the birds. A common problem in most of the
2 Desirable capabilities 12 Game genre serious games is the availability of limited activities; there
3 Intended learning outcomes 13 Reflection
is no option for the teachers, parents or caregivers to create
4 Instructional contents 14 Game story
customised activities for their students or children based on
5 Learning activities 15 Non-player characters
their needs. To overcome this problem, the prototype pro-
6 Game-based learning attributes 16 Narratives
vides a facility for the teachers working with children with
7 Modalities 17 Story
ASD, parents and caregivers to use their imagination and
8 Instruction method 18 Debriefing
create a customised activity by specifying values of the fol-
9 Instructional strategies 19 User achievements
lowing parameters as shown in Fig. 22:
10 Game mechanics 20 User profile
(1) Contents to be tested;
(2) Distractors to be shown;
participant selected an option to allow their child to take (3) Duration of an activity in the form of minutes and
part in the evaluation and signed the consent form. The pro- seconds;
totype currently supports learning of a total of 209 vocabu- (4) Activity type in the form of easy, medium and hard;
lary items which are divided into eleven categories including (5) Number of test items;
fruits, vegetables, animals, birds, shapes, colours, colours (6) Number of distractor items;
and shapes, people, vehicles, numbers and alphabets. For the

Fig. 20  Cumulative frequency of responses by the experts

13

768
Fig. 21  Proposed framework to design serious games for children with ASD
(7) Number of items to be shown including test and dis-
tractor items;
(8) Background image and transparency level of the
image;
(9) Moveable objects to be incorporated into an activity;
(10) Background music.
Once the values of the above-mentioned parameters have
been specified, they can save it as a new activity. The typi-
cal options, like loading, updating and deleting an existing
activity, are also supported by the prototype. The ‘start activ-
ity’ operation will create an environment by loading all the
parameters and start an activity for an individual (child with
ASD) to play.
The receptive learning of vocabulary among these
children was measured in terms of the number of correct
responses given by the participants and the number of
attempts made to identify each correct response. The meas-
urement was performed during baseline, i.e. before they used
the prototype for learning, during the intervention, i.e. while
they were using the prototype for learning, and during main-
tenance, i.e. after they had used the prototype. An overall
comparison of learning outcomes for the participants was
made, from the baseline, intervention and maintenance. The
summary of tasks performed in each session for the evalua-
tion is presented in Table 22.
The percentage of correct responses and the number of
attempts made for all five participants during baseline, inter-
vention and the maintenance are shown in Table 23. The
13
Universal Access in the Information Society (2020) 19:739–781
results show that the percentage of correct responses across
the participants increased from 54% in the baseline to 93%
in the intervention and 94% during the maintenance at the
end of weeks 1 and 2 following the withdrawal of inter-
vention. Similarly, the number of attempts made across the
participants lowered from 1.2 in the baseline to 1.1 in the
intervention and remained the same during the maintenance
at the end of weeks 1 and 2 following the withdrawal of
intervention. This shows that the serious game prototype
was effective for the learning of birds and can be used for
the learning of other categories of vocabulary by children
with ASD.
10 Conclusion
To the best of our knowledge, there exist no specialised
frameworks that can be employed to design serious games
to help autistic children in the learning of vocabulary, nor
any generalised framework that can be adapted to serve that
purpose. So far, the research of designing serious games
for autistic children has addressed inclusive design using
experts, children and game designers to develop games [36],
components to keep the children motivated in playing the
game [31] and closed-loop therapeutic intervention systems
[34, 35]. However, certain elements seem to be missing from
the research, such as components that can constitute seri-
ous games and the structure to support the design of games
for autistic children. To address this shortcoming, a serious
Universal Access in the Information Society (2020) 19:739–781 769
Fig. 22  Activity selector screen [78]
Table 22  Summary of the tasks Week Days Session
performed during each session
[78] 0 2 Presession
1 3 Baseline
1–4 15 Intervention
5 2 Maintenance 1
5 2 Maintenance 2
game design framework (SGDF) was presented in this paper
to develop serious games that can help autistic children learn
various vocabularies. The components of the framework are
based on three perspectives, i.e. children with ASD, typi-
cal children and game design in general. The components
related to ASD include necessary components to support
children with ASD in terms of learning vocabulary. An
existing IPO-based structure was used in the research pre-
sented; the structure was modified by incorporating all the
components found so that the original structure is tailored
to meet the needs of children with ASD.
The SGDF proposed was first evaluated through expert
reviews by the academic experts (who evaluate/use/design
such frameworks on a regular basis) focusing on the com-
ponents used, relationships among the components and
structure of the proposed framework. The revised version
of the framework was then further evaluated through expert
reviews by the game designers and researchers working in
the area of serious games focusing on their feedback on the
Participants Tasks
All Introduction to serious game ‘vocab builder’ and how to use it
All Initial assessment
All Learn vocabulary
All Test at the end of week one after the intervention was over
All Test at the end of week two after the intervention was over
components used in the framework (who work with these
components in real life). Hence, the structure, components
and their details of the proposed framework have been effec-
tively refined.
The refined framework proceeded with the applicability
test and six design documents were produced by the experts.
Each document contains specifications of game components.
The analysis of the components used in their design docu-
ments demonstrated the following:
(1) Experts used most of the components which shows that
the framework and its components are easy to under-
stand and can be used to fulfil design needs, i.e. a seri-
ous game for these children to learn vocabulary. There
were variations in the specifications of game compo-
nents among the designers. These variations reflect the
designers’ most valuable assets which are imagination
and creativity. Due to the adequacy of the components
13

770 Universal Access in the Information Society (2020) 19:739–781

Table 23  Percentage of correct Participant Correct responses Attempts made

responses and the number
of attempts made for all five Baseline (%) Interven- Mainte- Baseline Intervention Maintenance
participants tion (%) nance (%)

P1 72 97 100 1.4 1.0 1.0

P2 53 94 92 1.8 1.1 1.1
P3 59 81 88 1.6 1.3 1.2
P4 42 96 100 2.4 1.1 1.0
P5 42 89 88 2.3 1.1 1.1
Mean 54 93 94 1.9 1.1 1.1

provided by the framework, such creativity is supported
and not constrained;
(2) The specifications of game components in the design
documents were structured according to the IPO, i.e.
input, process and output. The underlying structure of
the proposed framework will aid in organising specifi-
cations which are both feasible and easy to follow.
The applicability survey conducted following their hands-
on experience with framework positively reaffirms its practi-
cal use. The experts expressed their willingness to use the
framework.
Future research can be carried out in terms of the pro-
posed SGDF as well as using it as a basis to develop the
prototypes. Future research directions are briefly discussed
in the following subsections.
10.1 Evaluation of SGDF
In terms of the research on the proposed framework, the
components incorporated from the ASD perspective were
based on the review of the literature on the vocabulary learn-
ing of children with ASD and the existing SGDFs for chil-
dren with ASD. Although the components from the vocabu-
lary learning seem specific, they can be considered generic
as the researchers can identify specific behaviours associated
with the learning of one or more skills related to children
with ASD to be targeted and the suitable instruction methods
that seem appropriate to teach the targeted skills in the seri-
ous game. However, the details of these components need
to be enriched so that they can be utilised for the develop-
ment of games for various other skills or evaluate existing
serious games to identify the missing components so that
serious games can be improved. The details may include
identification of all the behaviours related to the skills to be
targeted in the design of the serious game. It also requires
identification of all the evidence-based instruction methods
that have proved to be effective to provide an intervention to
these children for the targeted skills.
The researchers can also investigate to what extent the
components from the proposed SGDF are used in the seri-
ous games for children with ASD. This would allow the
researchers to identify the missing components in the pro-
posed SGDF and adapt the proposed framework by incorpo-
rating the missing components to fulfil their needs.
10.2 Development of new prototypes
Possibilities for future research towards the development of
new prototypes for children with ASD to learn difficult skills
may include:
(1) Targeting multiple behaviours in the design which
would allow an individual child with ASD to learn,
practise, improve and overcome their difficulties in one
or more behaviours targeted in the prototype. The range
of children which may be catered is highly dependent
on the behaviours targeted in its design;
(2) Considering a number of different categories of the
vocabulary items or the vocabulary items not supported
by the prototype discussed in this research;
(3) Investigating the impact of new modalities towards the
learning, generalisation and maintenance of the instruc-
tional content;
(4) Developing virtual reality (VR)-, augmented reality
(AR)- and mixed reality (MR)-based environments,
where children with ASD can be exposed to real-world
scenarios to support the generalisation of their learning.

13
Universal Access in the Information Society (2020) 19:739–781 771

Appendix

Appendix A
Note: The name and description of each component below are before the expert evaluation
of the framework was carried out. The name, description or both of the components were
modified based on the experts’ feedback. Therefore, it is recommended you to have a look
at the manuscript for the names and descriptions of components in the framework.

Part A – Demographic information of an expert

Gender: [ ] Male [ ] Female

Qualification: ____________________________________________________________

Designation: ____________________________________________________________

University/institution/company: ____________________________________________________________

Country: ____________________________________________________________

Research interests: ____________________________________________________________

Years of experience: ____________________________________________________________

Part B – importance, description and details of components, and their adequacy (missing,
removing) of components used in the proposed framework

The purpose of this section is to get feedback on the importance of each component in the game,
determine if the description is readable and the details in the description are sufficient or require
more information to be added and if there are components that need to be removed from or
added to the framework.

Component #: Autism behaviours

Description: Each child with ASD possess different set of behaviours. It is essential to identify the
behaviours that needs to be targeted in the serious game. The selection of these behaviour will help in the
selection and creation of appropriate instructional contents and learning activities.
Q1 Is this component important in the design of the game? [ ] Disagree [ ] Not sure [ ] Agree
Note:
If you ‘agree’, proceed from Q1a through Q1c and Q2
If you ‘disagree’ or ‘not sure’, proceed to Q2
A Does the name of component clearly state its use in the [ ] Disagree [ ] Not sure [ ] Agree
design of the game?
B Is the description of component readable? [ ] Disagree [ ] Not sure [ ] Agree
C Are the details provided in the description sufficient? [ ] Disagree [ ] Not sure [ ] Agree
Q2 Remarks (if any): this will help us to improve a specific aspect of component for which the answer is
either ‘disagree’ or ‘not sure’.

Component #: Capability
Description: Capability refers to the cognitive, psychomotor, and possibly affective skills which the learner is
to develop as a result of playing the game. Cognitive skills include the capabilities of recall, analysis,
synthesis, and evaluation. Psychomotor skills include the capabilities of well-timed, fluid execution.

13

772 Universal Access in the Information Society (2020) 19:739–781

Affective skills include the capabilities of identifying, adopting, and valuing appropriate attitudes and points
of view.
Q1 Is this component important in the design of the game? [ ] Disagree [ ] Not sure [ ] Agree
Note:
If you ‘agree’, proceed from Q1a through Q1c and Q2
If you ‘disagree’ or ‘not sure’, proceed to Q2
A Does the name of component clearly state its use in the [ ] Disagree [ ] Not sure [ ] Agree
design of the game?
B Is the description of component readable? [ ] Disagree [ ] Not sure [ ] Agree
C Are the details provided in the description sufficient? [ ] Disagree [ ] Not sure [ ] Agree
Q2 Remarks (if any): this will help us to improve a specific aspect of component for which the answer is
either ‘disagree’ or ‘not sure’.

Component #: Learning outcomes

Description: Learning outcomes are the goals to be achieved from playing the serious game. An intended
learning outcome is a particular combination of capability and subject matter. For example, the learner
should be able to recall the date of the battle of Hastings or should be able to analyse whether a particular
bird is a raptor. In an attempt to synthesize the work of Gagne (1984), Anderson (1982), and others, Kraiger,
Ford, and Salas (1993) proposed several broad categories of learning outcomes: skill-based, cognitive, and
affective outcomes. Skill-based learning outcomes address technical or motor skills. Cognitive learning
outcomes include three subcategories of declarative knowledge, procedural knowledge, and strategic
knowledge. Affective learning outcomes refer to attitudes.
Q1 Is this component important in the design of the game? [ ] Disagree [ ] Not sure [ ] Agree
Note:
If you ‘agree’, proceed from Q1a through Q1c and Q2
If you ‘disagree’ or ‘not sure’, proceed to Q2
A Does the name of component clearly state its use in the [ ] Disagree [ ] Not sure [ ] Agree
design of the game?
B Is the description of component readable? [ ] Disagree [ ] Not sure [ ] Agree
C Are the details provided in the description sufficient? [ ] Disagree [ ] Not sure [ ] Agree
Q2 Remarks (if any): this will help us to improve a specific aspect of component for which the answer is
either ‘disagree’ or ‘not sure’.

Component #: Instructional contents

Description: The instructional contents are the subject matter that players should learn. The details of the
actual subject matter to be learnt, or the type of contents that the players learn, could be an exhaustive list.
Gilbert and Gale (2007) state that contents can be classified into four types: facts, procedures, concepts, and
principles.
Q1 Is this component important in the design of the game? [ ] Disagree [ ] Not sure [ ] Agree
Note:
If you ‘agree’, proceed from Q1a through Q1c and Q2
If you ‘disagree’ or ‘not sure’, proceed to Q2
A Does the name of component clearly state its use in the [ ] Disagree [ ] Not sure [ ] Agree
design of the game?
B Is the description of component readable? [ ] Disagree [ ] Not sure [ ] Agree
C Are the details provided in the description sufficient? [ ] Disagree [ ] Not sure [ ] Agree

13
Universal Access in the Information Society (2020) 19:739–781 773

Q2 Remarks (if any): this will help us to improve a specific aspect of component for which the answer is
either ‘disagree’ or ‘not sure’.

Component #: Reflection
Description: Reflection is where the learner thinks about the purpose of the learning activities that have been
undertaken, and decides the strategy to apply during the next activity.
Q1 Is this component important in the design of the game? [ ] Disagree [ ] Not sure [ ] Agree
Note:
If you ‘agree’, proceed from Q1a through Q1c and Q2
If you ‘disagree’ or ‘not sure’, proceed to Q2
A Does the name of component clearly state its use in the [ ] Disagree [ ] Not sure [ ] Agree
design of the game?
B Is the description of component readable? [ ] Disagree [ ] Not sure [ ] Agree
C Are the details provided in the description sufficient? [ ] Disagree [ ] Not sure [ ] Agree
Q2 Remarks (if any): this will help us to improve a specific aspect of component for which the answer is
either ‘disagree’ or ‘not sure’.

Component #: Instruction method

Description: children with ASD require individually designed interventions that meet their needs. Instruction
methods for ASD are the evidence-based methods that have proven to be effective for some individuals on
the spectrum as there is no universal method which has been identified as effective method for all the
children on spectrum.
Q1 Is this component important in the design of the game? [ ] Disagree [ ] Not sure [ ] Agree
Note:
If you ‘agree’, proceed from Q1a through Q1c and Q2
If you ‘disagree’ or ‘not sure’, proceed to Q2
A Does the name of component clearly state its use in the [ ] Disagree [ ] Not sure [ ] Agree
design of the game?
B Is the description of component readable? [ ] Disagree [ ] Not sure [ ] Agree
C Are the details provided in the description sufficient? [ ] Disagree [ ] Not sure [ ] Agree
Q2 Remarks (if any): this will help us to improve a specific aspect of component for which the answer is
either ‘disagree’ or ‘not sure’.

Component #: Strategies
Description: The strategies are used to support instruction methods for ASD. They are the techniques which
facilitate instructors to get the attention of children, stimulate their interest towards the learning of contents,
keep them engaged in the learning process, develop their thinking skills and increase interaction so that they
remain focused on the learning. The strategies can also be used to connect idea to the real-world which can
provide opportunity to these children to generalize the contents are one situation to another.
Q1 Is this component important in the design of the game? [ ] Disagree [ ] Not sure [ ] Agree
Note:
If you ‘agree’, proceed from Q1a through Q1c and Q2
If you ‘disagree’ or ‘not sure’, proceed to Q2
A Does the name of component clearly state its use in the [ ] Disagree [ ] Not sure [ ] Agree

13

774 Universal Access in the Information Society (2020) 19:739–781

design of the game?

B Is the description of component readable? [ ] Disagree [ ] Not sure [ ] Agree
C Are the details provided in the description sufficient? [ ] Disagree [ ] Not sure [ ] Agree
Q2 Remarks (if any): this will help us to improve a specific aspect of component for which the answer is
either ‘disagree’ or ‘not sure’.

Component #: Learning activities

Description: Learning activities are the set of activities designed to keep players actively involved when they
are playing the game. The effective design of these activities ensures that players stay engaged and immersed
without getting bored. It is essential to ensure that the learning material used is appropriate and challenging
for players who are seeking improvement in their competency is slightly above their current level of
competency James (2003). These activities make use of the material from the contents to be used in the
games.
Q1 Is this component important in the design of the game? [ ] Disagree [ ] Not sure [ ] Agree
Note:
If you ‘agree’, proceed from Q1a through Q1c and Q2
If you ‘disagree’ or ‘not sure’, proceed to Q2
A Does the name of component clearly state its use in the [ ] Disagree [ ] Not sure [ ] Agree
design of the game?
B Is the description of component readable? [ ] Disagree [ ] Not sure [ ] Agree
C Are the details provided in the description sufficient? [ ] Disagree [ ] Not sure [ ] Agree
Q2 Remarks (if any): this will help us to improve a specific aspect of component for which the answer is
either ‘disagree’ or ‘not sure’.

Component #: Game genre

Description: Game genre is used to categorise games based on the interaction between the players’ so-called
gameplay rather than differences in visual or narratives. It is defined by a set of gameplay challenges. These
genres range from action, adventure to a combination of action-adventure, role-playing, simulations,
strategy, and sports among others.
Q1 Is this component important in the design of the game? [ ] Disagree [ ] Not sure [ ] Agree
Note:
If you ‘agree’, proceed from Q1a through Q1c and Q2
If you ‘disagree’ or ‘not sure’, proceed to Q2
A Does the name of component clearly state its use in the [ ] Disagree [ ] Not sure [ ] Agree
design of the game?
B Is the description of component readable? [ ] Disagree [ ] Not sure [ ] Agree
C Are the details provided in the description sufficient? [ ] Disagree [ ] Not sure [ ] Agree
Q2 Remarks (if any): this will help us to improve a specific aspect of component for which the answer is
either ‘disagree’ or ‘not sure’.

Component #: Game mechanics

Description: Game mechanics are a set of actions, behaviours and control mechanisms given to a player
within the context of a game. They define the game as a rule-based system, specifying what there is, how
everything behaves, and how the player can interact with the game world.

13
Universal Access in the Information Society (2020) 19:739–781 775

Q1 Is this component important in the design of the game? [ ] Disagree [ ] Not sure [ ] Agree
Note:
If you ‘agree’, proceed from Q1a through Q1c and Q2
If you ‘disagree’ or ‘not sure’, proceed to Q2
A Does the name of component clearly state its use in the [ ] Disagree [ ] Not sure [ ] Agree
design of the game?
B Is the description of component readable? [ ] Disagree [ ] Not sure [ ] Agree
C Are the details provided in the description sufficient? [ ] Disagree [ ] Not sure [ ] Agree
Q2 Remarks (if any): this will help us to improve a specific aspect of component for which the answer is
either ‘disagree’ or ‘not sure’.

Component #: Game dynamics

Description: Dynamics are the emergent behavior that arises from gameplay, when the game
mechanics are put in use. This provides more fun, enjoyment and engagement in the games for the player.
Q1 Is this component important in the design of the game? [ ] Disagree [ ] Not sure [ ] Agree
Note:
If you ‘agree’, proceed from Q1a through Q1c and Q2
If you ‘disagree’ or ‘not sure’, proceed to Q2
A Does the name of component clearly state its use in the [ ] Disagree [ ] Not sure [ ] Agree
design of the game?
B Is the description of component readable? [ ] Disagree [ ] Not sure [ ] Agree
C Are the details provided in the description sufficient? [ ] Disagree [ ] Not sure [ ] Agree
Q2 Remarks (if any): this will help us to improve specific aspect of component for which the answer is
either ‘disagree’ or ‘not sure’.

Component #: Storytelling
Description: There are two perspectives on storytelling in games, the designer’s story and the player’s story
(Rouse, 2004, pp. 216-218). The designer’s story is the storytelling that is designed into the game. The
designer’s story can be used to set the stage, provide purpose and engagement, and convey content, among
other things. The setting, character design, and narrative are the designers primary design tools.
The storytelling that occurs during play combines the designer’s story with the interactions and choices the
player makes. The resulting experience crafts the player’s story. Some games have stronger designer stories,
such as adventure and role-playing games, while others have little to no designer story, such as classic arcade
games like Pacman and puzzle games like Tetris. However, all games have a player’s story, which at the
very least reflects the story of the gameplay challenges encountered by the player and how the player
addressed them. When approaching a design, the designer must first decide on what type of stories he or she
wants the player to be able to experience and design the setting, character design, and narrative to achieve
this.
Q1 Is this component important in the design of the game? [ ] Disagree [ ] Not sure [ ] Agree
Note:
If you ‘agree’, proceed from Q1a through Q1c and Q2
If you ‘disagree’ or ‘not sure’, proceed to Q2
A Does the name of component clearly state its use in the [ ] Disagree [ ] Not sure [ ] Agree
design of the game?
B Is the description of component readable? [ ] Disagree [ ] Not sure [ ] Agree
C Are the details provided in the description sufficient? [ ] Disagree [ ] Not sure [ ] Agree
Q2 Remarks (if any): this will help us to improve a specific aspect of component for which the answer is
either ‘disagree’ or ‘not sure’.

13

776 Universal Access in the Information Society (2020) 19:739–781

Component #: Narrative
Description: A narrative (or story) is any report of connected events, presented in a sequence of written or
spoken words, and/or in a sequence of moving pictures. Narrative is typically used to establish the setting
and initial motivation of the game, but often it is not the main focus of gameplay. Narratives often serve as a
means to explore the future. As Edward Branigan (2006) has mentioned: one of the purposes of seeing and
perceiving narratively is to weigh how certain effects that are desired may be achieved, how desire is linked
to possibilities for being, how events may proceed. In this way, perceiving narratively operates to draw the
future into desires expressed in the present as well as demonstrates how the present was caused by the past
and how the present may have effects in the future (p.32).
Q1 Is this component important in the design of the game? [ ] Disagree [ ] Not sure [ ] Agree
Note:
If you ‘agree’, proceed from Q1a through Q1c and Q2
If you ‘disagree’ or ‘not sure’, proceed to Q2
A Does the name of component clearly state its use in the [ ] Disagree [ ] Not sure [ ] Agree
design of the game?
B Is the description of component readable? [ ] Disagree [ ] Not sure [ ] Agree
C Are the details provided in the description sufficient? [ ] Disagree [ ] Not sure [ ] Agree
Q2 Remarks (if any): this will help us to improve a specific aspect of component for which the answer is
either ‘disagree’ or ‘not sure’.

Component #: Characters
Description: These characters are a non-player characters (NPC) sometimes known as a non-person
characters or non-playable characters which are controlled by the computer through artificial intelligence
(AI) than controlled by the player in game.
Q1 Is this component important in the design of the game? [ ] Disagree [ ] Not sure [ ] Agree
Note:
If you ‘agree’, proceed from Q1a through Q1c and Q2
If you ‘disagree’ or ‘not sure’, proceed to Q2
A Does the name of component clearly state its use in the [ ] Disagree [ ] Not sure [ ] Agree
design of the game?
B Is the description of component readable? [ ] Disagree [ ] Not sure [ ] Agree
C Are the details provided in the description sufficient? [ ] Disagree [ ] Not sure [ ] Agree
Q2 Remarks (if any): this will help us to improve a specific aspect of component for which the answer is
either ‘disagree’ or ‘not sure’.

Component #: Game attributes

Description: The learning through serious games needs to be validated against current teaching practices to
ensure it is at the same level as that of traditional classroom learning. It is essential to incorporate certain
aspects of current teaching practices in serious games that have proven to be successful in the classroom.
These successful aspects that support learning and engagement in serious game are called game attributes.
The game attributes are based on four theories namely behaviourism, cognitivism, constructivism and
psychology theory.
Q1 Is this component important in the design of the game? [ ] Disagree [ ] Not sure [ ] Agree

13
Universal Access in the Information Society (2020) 19:739–781 777

Note:
If you ‘agree’, proceed from Q1a through Q1c and Q2
If you ‘disagree’ or ‘not sure’, proceed to Q2
A Does the name of component clearly state its use in the [ ] Disagree [ ] Not sure [ ] Agree
design of the game?
B Is the description of component readable? [ ] Disagree [ ] Not sure [ ] Agree
C Are the details provided in the description sufficient? [ ] Disagree [ ] Not sure [ ] Agree
Q2 Remarks (if any): this will help us to improve a specific aspect of component for which the answer is
either ‘disagree’ or ‘not sure’.

Component #: Modalities
Description: Nigay and Coutaz (1993) have defined modality as “...the type of communication channel used
to convey or acquire information. It also covers the way an idea is expressed or perceived, or the manner an
action is performed”. Games that make use of just one modality are known as unimodal, whereas, games in
which two modalities are used are known as bimodal. Finally, games that integrate more than two modalities
are known as multimodal games. The combination of two or more modalities for input is known as
multimodal fusion. Likewise, the partitioning of information into two or more communication modalities is
also known as multimodal fission. Examples of input modalities include speech, gestures, gait, facial
expression among others, whereas, text, graphics, animation, video model, virtual character, and force
feedback are few examples of output modalities.
Q1 Is this component important in the design of the game? [ ] Disagree [ ] Not sure [ ] Agree
Note:
If you ‘agree’, proceed from Q1a through Q1c and Q2
If you ‘disagree’ or ‘not sure’, proceed to Q2
A Does the name of component clearly state its use in the [ ] Disagree [ ] Not sure [ ] Agree
design of the game?
B Is the description of component readable? [ ] Disagree [ ] Not sure [ ] Agree
C Are the details provided in the description sufficient? [ ] Disagree [ ] Not sure [ ] Agree
Q2 Remarks (if any): this will help us to improve a specific aspect of component for which the answer is
either ‘disagree’ or ‘not sure’.

Component #: User profile and achievements

Description: Game must allow users to set up their own profiles that contain their basic information. Once
created, this profile will be updated regularly with all the accomplishments as users progress through various
activities of the games. This way, users can also look back at all the achievements. It can also allow them to
practise one of the previously learnt skills to improve their learning. Achievement in the activities can be
shown in the form of scores, number of resources gathered, or the time taken to accomplish that task.
Q1 Is this component important in the design of the game? [ ] Disagree [ ] Not sure [ ] Agree
Note:
If you ‘agree’, proceed from Q1a through Q1c and Q2
If you ‘disagree’ or ‘not sure’, proceed to Q2
A Does the name of component clearly state its use in the [ ] Disagree [ ] Not sure [ ] Agree
design of the game?
B Is the description of component readable? [ ] Disagree [ ] Not sure [ ] Agree
C Are the details provided in the description sufficient? [ ] Disagree [ ] Not sure [ ] Agree
Q2 Remarks (if any): this will help us to improve a specific aspect of component for which the answer is

13

778 Universal Access in the Information Society (2020) 19:739–781

either ‘disagree’ or ‘not sure’.

Q3a. Are there any new components that need to be added?

Q3b. How adding components mentioned in response to the above question (Q3a) will help in
the design of the game?

Part C – structure of the framework

The purpose of this section is to validate the logical division of components to the three phases
corresponding to input, process and output of IPO and the structure of the framework.

C1 – components in the input phase

Q4 Is the distribution of the following components in the ‘input’ phase of IPO coherent?
A Autism behaviours [ ] Disagree [ ] Not sure [ ] Agree
B Capabilities [ ] Disagree [ ] Not sure [ ] Agree
C Learning outcomes [ ] Disagree [ ] Not sure [ ] Agree
D Instruction contents [ ] Disagree [ ] Not sure [ ] Agree
E If any of your answers in the previous question was ‘disagree’ or ‘not sure’, can you elaborate on how
the distribution of those components can be improved?

C2 – components in the process phase

Q5 Is the distribution of the following components in the ‘process’ phase of IPO coherent?
A Refection [ ] Disagree [ ] Not sure [ ] Agree
B Instruction method [ ] Disagree [ ] Not sure [ ] Agree
C Strategies [ ] Disagree [ ] Not sure [ ] Agree
D Game genre [ ] Disagree [ ] Not sure [ ] Agree
E Game mechanics [ ] Disagree [ ] Not sure [ ] Agree
F Game dynamics [ ] Disagree [ ] Not sure [ ] Agree
G Scene [ ] Disagree [ ] Not sure [ ] Agree
H Story [ ] Disagree [ ] Not sure [ ] Agree
I Narratives [ ] Disagree [ ] Not sure [ ] Agree
J Characters [ ] Disagree [ ] Not sure [ ] Agree
K Game attributes [ ] Disagree [ ] Not sure [ ] Agree
L Modalities [ ] Disagree [ ] Not sure [ ] Agree
M Multimodal fusion [ ] Disagree [ ] Not sure [ ] Agree
N Multimodal fission [ ] Disagree [ ] Not sure [ ] Agree
O If any of your answers in the previous question was ‘disagree’ or ‘not sure’, can you elaborate on how
the distribution of those components can be improved?

C3 – components in the output phase

Q6 Is the distribution of the following components in the ‘output’ phase of IPO coherent?

13
Universal Access in the Information Society (2020) 19:739–781 779

A User achievements [ ] Disagree [ ] Not sure [ ] Agree

B User profile [ ] Disagree [ ] Not sure [ ] Agree
C If any of your answers in the previous question was ‘disagree’ or ‘not sure’, can you elaborate on how
the distribution of those components can be improved?

C4 – IPO based structure

Q7 Is the IPO based structure easy to understand? [ ] Disagree [ ] Not sure [ ] Agree
Q8 Does IPO based structure represents the structure of [ ] Disagree [ ] Not sure [ ] Agree
typical games?
Q9 Can you easily make use of this structure to design [ ] Disagree [ ] Not sure [ ] Agree
vocabulary based games to facilitate these children?

Part D – relations between components

The purpose of this section is to ascertain whether the relations between components are logical
or it needs to be fixed and to determine the missing relations (if any) so that framework can be
improved accordingly.

10 Are the following relations between components (connected to each other through an arrow) logical?

A Autism behaviors capabilities [ ] Disagree [ ] Not sure [ ] Agree

B Capabilities learning outcomes [ ] Disagree [ ] Not sure [ ] Agree
C Learning outcomes instruction contents [ ] Disagree [ ] Not sure [ ] Agree
D Learning outcomes learning activity [ ] Disagree [ ] Not sure [ ] Agree
E Game attributes (behaviorism, constructivism, [ ] Disagree [ ] Not sure [ ] Agree
cognitivism and psychology)
F Learning activity user achievements [ ] Disagree [ ] Not sure [ ] Agree
G If any of your answers in the previous questions (Q10A through Q10F) was ‘disagree’, can you
elaborate on how these relations can be improved?

11a Are there any relations (other than relations mentioned in Q10) among the components which you
think are missing?

11b How adding the new relations mentioned in the response of the above question (Q11a) will improve
the overall structure of the framework?

Part D – comprehensiveness of the framework

The purpose of this section is to get your suggestions or recommendations for the improvement
of the framework.

Q12. Do you have any other suggestions/recommendations for the improvement of the
framework?

References 6. Nation, I.S.P.: Teaching and Learning Vocabulary, 1st edn. Heinle
ELT, Boston (1990)
7. Siriwan, M.: English vocabulary learning strategies employed by
1. Chiang, H.M., Lin, Y.H.: Reading comprehension instruction for
Rajabhat University students. Unpublished Doctoral Dissertation,
students with autism spectrum disorders a review of the literature.
Suranaree University of Technology, Thailand (2007)
Focus Autism Other Dev. Disabil. 22(4), 259–267 (2007)
8. Barbu, E., Martın-Valdivia, M.T., Urena-López, L.A.: Open book:
2. Khowaja, K., Salim, S.S.: A systematic review of strategies and
a tool for helping ASD users’ semantic comprehension. In: Paper
computer-based intervention (CBI) for reading comprehension of
presented at the Proceedings of the 2th Workshop of Natural Lan-
children with autism. Res. Autism Spectrum Disord. 7(9), 1111–
guage Processing for Improving Textual Accessibility (NLP4ITA)
1121 (2013)
(2013)
3. Fukkink, R.G., Hulstijn, J., Simis, A.: Does training in second-
9. Ramdoss, S., Mulloy, A., Lang, R., O’Reilly, M., Sigafoos, J.,
language word recognition skills affect reading comprehension?
Lancioni, G., EL Zein, F.: Use of computer-based interventions
An experimental study. Mod. Lang. J. 89(1), 54–75 (2005)
to improve literacy skills in students with autism spectrum dis-
4. Wilkins, D.A.: Linguistics in language teaching: E. Arnold, 1973
orders: a systematic review. Res. Autism Spectrum Disord. 5(4),
(1972)
1306–1318 (2011)
5. Thornbury, S.: How to Teach Vocabulary. Pearson Education Lim-
10. Whalon, K.J., Al Otaiba, S., Delano, M.E.: Evidence-based read-
ited, London (2006)
ing instruction for individuals with autism spectrum disorders.

13

780
Focus Autism Other Dev. Disabl. 24(1), 3–16 (2009). https​://doi.
org/10.1177/10883​57608​32851​5
11. Iuppa, N.V., Borst, T.: Story and Simulations for Serious Games:
Tales from the Trenches. Focal Press, Waltham (2006)
12. Zyda, M.: From visual simulation to virtual reality to games. Com-
puter 38(9), 25–32 (2005)
13. Sawyer, B: Ten Myths About Serious Games. Escapist Magazine
(2007)
14. Kiili, K.: Digital game-based learning: towards an experiential
gaming model. Internet High. Educ. 8(1), 13–24 (2005)
15. Noor, M., Shahbodin, F., Pee, C.: Serious Game for Autism Chil-
dren: Review of Literature. In: Paper presented at the World Acad-
emy of Science, Engineering and Technology (2012)
16. Zakari, H.M., Ma, M., Simmons, D.: A Review of Serious Games
for Children with Autism Spectrum Disorders (ASD) Serious
Games Development and Applications, pp. 93–106. Springer,
Berlin (2014)
17. Khowaja, K., Salim, S.S.: Design components for serious games
of children with autism spectrum disorder (ASD). In: Paper pre-
sented at the Serious Games Conference 2014 Kintex Convention
Centre, South Korea (2014)
18. Khowaja, K., Al-Thani, D., Salim, S.S.: Vocabulary learning of
children with autism spectrum disorder (ASD): from the develop-
ment to an evaluation of serious game prototype. In: Paper Pre-
sented at the 12th European Conference on Games Based Learn-
ing, Sophia Antipolis, France (2018)
19. Khowaja, K., Salim, S.S., Al-Thani, D.: Components to design
serious games for children with autism spectrum disorder (ASD)
to learn vocabulary. In: Paper Presented at the 2018 IEEE 5th
International Conference on Engineering Technologies and
Applied Sciences (ICETAS) (2018)
20. E-Games: Game Development Process (n.d.). Retrieved from
http://www.e-games​.tech.purdu​e.edu/GameD​evPro​cess.asp
21. Hayes, G.R., Hirano, S., Marcu, G., Monibi, M., Nguyen, D.H.,
Yeganyan, M.: Interactive visual supports for children with
autism. Pers. Ubiquitous Comput. 14(7), 663–680 (2010)
22. Garris, R., Ahlers, R., Driskell, J.E.: Games, motivation, and
learning: a research and practice model. Simul. Gaming 33(4),
441–467 (2002)
23. Winn, B.: The design, play, and experience framework. Handb.
Res. Eff. Electron. Gaming Educ. 3, 1010–1024 (2008)
24. Yusoff, A.: A conceptual framework for serious games and its
validation. University of Southampton (2010)
25. Arachchilage, N.A.G., Love, S.: A game design framework for
avoiding phishing attacks. Comput. Hum. Behav. 29(3), 706–714
(2013)
26. Denis, G., Jouvelot, P.: Motivation-driven educational game
design: applying best practices to music education. In: Paper Pre-
sented at the Proceedings of the 2005 ACM SIGCHI International
Conference on Advances in Computer Entertainment Technology
(2005)
27. Järvinen, A.: Game design for social networks: interaction design
for playful dispositions. In: Paper Presented at the Proceedings of
the 2009 ACM SIGGRAPH Symposium on Video Games (2009)
28. Ijsselsteijn, W., Nap, H.H., de Kort, Y., Poels, K.: Digital game
design for elderly users. In: Paper Presented at the Proceedings of
the 2007 conference on Future Play (2007)
29. Moreno-Ger, P., Burgos, D., Martínez-Ortiz, I., Sierra, J.L.,
Fernández-Manjón, B.: Educational game design for online edu-
cation. Comput. Hum. Behav. 24(6), 2530–2540 (2008)
30. Park, J.H., Abirached, B., Zhang, Y.: A framework for designing
assistive technologies for teaching children with ASDs emotions.
In: Paper Presented at the CHI’12 Extended Abstracts on Human
Factors in Computing Systems (2012)
31. Whyte, E.M., Smyth, J.M., Scherf, K.S.: Designing serious
game interventions for individuals with autism. J. Autism Dev.
13
Universal Access in the Information Society (2020) 19:739–781
Disord. 45(12), 3820–3831 (2015). https​://doi.org/10.1007/s1080​
3-014-2333-1
32. Baranowski, T., Buday, R., Thompson, D.I., Baranowski, J.: Play-
ing for real: video games and stories for health-related behav-
ior change. Am. J. Prev. Med. 34(1), 74–82 (2008). https​://doi.
org/10.1016/j.amepr​e.2007.09.027
33. Kapp, K.M.: The Gamification of Learning and Instruction:
Game-Based Methods and Strategies for Training and Education.
Wiley, New York (2012)
34. Bono, V., Narzisi, A., Jouen, A.L., Tilmont, E., Hommel, S.,
Jamal, W., Group, M.S.: GOLIAH: a gaming platform for home-
based intervention in autism–principles and design. Front Psychia-
try 7(70), 70 (2016). https​://doi.org/10.3389/fpsyt​.2016.00070​
35. Jouen, A.-L., Narzisi, A., Xavier, J., Tilmont, E., Bodeau, N.,
Bono, V., Group, M.S.: GOLIAH (Gaming Open Library for Inter-
vention in Autism at Home): a 6-month single blind matched con-
trolled exploratory study. Child Adolesc. Psychiatry Ment. Health
11, 17 (2017). https​://doi.org/10.1186/s1303​4-017-0154-7-&gt
36. Malinverni, L., Mora-Guiard, J., Padillo, V., Valero, L., Hervás,
A., Pares, N.: An inclusive design approach for developing video
games for children with autism spectrum disorder. Comput. Hum.
Behav. 71, 535–549 (2016)
37. Miles, M.B., Huberman, A.M.: Qualitative Data Analysis: An
Expanded Sourcebook. Sage Publications, Thousand Oaks (1994)
38. Maxwell, J.A.: Qualitative Research Design: An Interactive
Approach. Sage Publications, Thousand Oaks (2012)
39. Kitchenham, B.: Procedures for performing systematic reviews.
Keele University Technical Report TR/SE-0401 and NICTA
Technical Report 0400011T.1. http://www.it.hiof.no/~haral​dh/
misc/2016-08-22-smat/Kitch​enham​-Syste​matic​-Revie​w-2004.
pdf (2004)
40. Case-Smith, J., Weaver, L.L., Fristad, M.A.: A systematic review
of sensory processing interventions for children with autism
spectrum disorders. Autism 19(2), 133–148 (2015). https​://doi.
org/10.1177/13623​61313​51776​2
41. Dawson, G., Watling, R.: Interventions to facilitate auditory,
visual, and motor integration in autism: a review of the evidence.
J. Autism Dev. Disord. 30(5), 415–421 (2000)
42. Rocky Point Academy: Characteristics of autism (1997). Retrieved
from http://www.calga​ryaut​ism.com/chara​cteri​stics​.htm
43. American Psychiatric Association: Diagnostic and Statistical
Manual of Mental Disorders: DSM-5, 5th edn. American Psychi-
atric Association, Washington, D.C (2013)
44. Bosseler, A., Massaro, D.W.: Development and evaluation of a
computer-animated tutor for vocabulary and language learning
in children with autism. J. Autism Dev. Disord. 33(6), 653–672
(2003)
45. Johnson, C.P.: Early Clinical Characteristics of Children with
Autism Autistic Spectrum Disorders in Children, pp. 83–121.
CRC Press, Boca Raton (2004)
46. Centers for Disease Control and Prevention [CDC], Signs and
symptoms of ASD. (n.d.). Retrieved from http://www.cdc.gov/
ncbdd​d/autis​m/signs​.html
47. Bloom, B.S.: Taxonomy of Educational Objectives Book 1: Cog-
nitive Domain, 2nd edn. Longman, London (1971)
48. Simpson, E.: Educational Objectives in the Psychomotor Domain:
The Psychomotor Domain, vol. 3. Gryphon House, Washington,
DC (1971)
49. Harrow, A.J.: A Taxonomy of the Psychomotor Domain. A Guide
for Developing Behavioral Objectives. David McKay, New York
(1972)
50. Dave, R.H.: Psychomotor levels. In: Armstrong, R.J. (ed.) Devel-
oping and Writing Educational Objective, pp. 33–34. Educational
Innovators Press, Tucson (1970)
51. Krathwohl, D.R., Bloom, B.S., Masia, B.B.: Taxonomy of Edu-
cational Objectives, the Classification of Educational Goals:
Universal Access in the Information Society (2020) 19:739–781 781
Handbook II: Affective Domain, vol. 2. David McKay Company,
Philadelphia (1967)
52. Garris, R., Ahlers, R.: A game-based training model develop-
ment, application, and evaluation. In: Paper Presented at the The
Interservice/Industry Training, Simulation & Education Confer-
ence (I/ITSEC) (2001)
53. Kennedy, D., Hyland, Á., Ryan, N.: Writing and Using Learn-
ing Outcomes: A Practical Guide. University College Cork, Cork
(2006)
54. Gagne, R.M.: The conditions of learning, 4th edn. Holt, Rinehart
& Winston, New York (1985)
55. Gilbert, L., Gale, V.: Front-end Analysis Principle of E-Learning
Systems Engineering, pp. 77–101. Chandos Publishing, Oxford
(2008)
56. Simon, H.A.: Learning and problem solving. In: Kolb, D.A.,
Rubin, I.M., McIntyre, J.M. (eds.) Organizational Psychology:
A Book of Readings, pp. 36–58. Prentice-Hall, Englewood Cliffs
(1979)
57. National Autism Center [NAC]: National Standards Report:
the national standards project-addressing the need for evi-
dence based practice guidelines for Autism spectrum disorders
(2009). Retrieved from http://www.natio​nalau​tismc​enter​.org/pdf/
NAC%20Sta​ndard​s%20Rep​ort.pdf
58. Gee, J.P.: What Video Games Have to Teach Us about Learning
and Literacy. Palgrave Macmillan, Basingstoke (2007)
59. Hunicke, R., LeBlanc, M., Zubek, R.: MDA: a formal approach
to game design and game research. In: Paper Presented at the
Proceedings of the AAAI Workshop on Challenges in Game AI,
San Jose, California, United States of America (2004)
60. Schreiber, I.: Level 5: Mechanics and Dynamics (2009). Retrieved
from https​://gamed​esign​conce​pts.wordp​ress.com/2009/07/13/
level​-5-mecha​nics-and-dynam​ics/
61. Sinclair, K.: The art of storytelling in gaming (2014). Retrieved
from http://www.sunda​nce.org/blogs​/progr​am-spotl​ight/the-art-
of-story​telli​ng-in-gamin​g
62. Rouse III, R.: Game design: Theory and practice. Jones & Bartlett
Learning, Burlington (2010)
63. Adams, E.: Fundamentals of game design, 2nd edn. Pearson Edu-
cation, London (2010)
64. Aarseth, E.: A narrative theory of games. In: Paper presented at
the Proceedings of the International Conference on the Founda-
tions of Digital Games (2012)
65. Techopedia: Bot (2016). Retrieved from https​://www.techo​pedia​
.com/defin​ition​/10459​/bot-gamin​g
66. Kienzle, J., Denault, A., Vangheluwe, H.: Model-based design of
computer-controlled game character behavior. In: Paper Presented
at the International Conference on Model Driven Engineering
Languages and Systems (2007)
67. Bedwell, W.L., Pavlas, D., Heyne, K., Lazzara, E.H., Salas, E.:
Toward a taxonomy linking game attributes to learning an empiri-
cal study. Simul. Gaming 43(6), 729–760 (2012)
68. Murphy, C.: Why games work and the science of learning. Paper
presented at the Interservice, Interagency Training, Simulations,
and Education Conference (2011)
69. Wilson, K.A., Bedwell, W.L., Lazzara, E.H., Salas, E., Burke,
C.S., Estock, J.L., Conkey, C.: Relationships between game attrib-
utes and learning outcomes review and research proposals. Simul.
Gaming 40(2), 217–266 (2009)
70. Nigay, L., Coutaz, J.: A design space for multimodal systems:
concurrent processing and data fusion. In: Paper Presented at the
Proceedings of the INTERACT’93 and CHI’93 conference on
Human factors in computing systems (1993)
71. Oviatt, S.: Multimodal interfaces. In: Jacko, J.A. (ed.) Human-
Computer Interaction Handbook: Fundamentals, Evolving Tech-
nologies, and Emerging Applications, pp. 286–304. Lawrence
Erlbaum, New Jersey (2012)
72. Khowaja, K., Salim, S.S., Asemi, A., Ghulamani, S., Shah, A.:
A systematic review of modalities in computer-based interven-
tions (CBIs) for language comprehension and decoding skills of
children with autism spectrum disorder (ASD). Universal Access
in the Information Society. (2019). https​://doi.org/10.1007/s1020​
9-019-00646​-1
73. Lederman, L.C.: Debriefing: a critical reexamination of the post-
experience analytic process with implications for its effective use.
Simul. Games 15(4), 415–431 (1984)
74. Seaman, C.B.: Qualitative methods in empirical studies of soft-
ware engineering. IEEE Trans. Softw. Eng. 25(4), 557–572 (1999)
75. de Freitas, S., Jarvis, S.: A framework for developing serious
games to meet learner needs. In: Paper Presented at the The
Interservice/Industry Training, Simulation & Education Confer-
ence (I/ITSEC) (2006)
76. ISO/IEC 25010—Systems and software engineering—systems and
software Quality Requirements and Evaluation (SQuaRE)—sys-
tem and software quality models ISO/IEC (2010). Retrieved from
https​://www.iso.org/obp/ui/#iso:std:iso-iec:25010​:ed-1:v1:en
77. Lin, H.X., Choong, Y.-Y., Salvendy, G.: A proposed index of usa-
bility: a method for comparing the relative usability of different
software systems. Behav. Inf. Technol. 16(4–5), 267–277 (1997)
78. Khowaja, K., Salim, S.S.: Serious game for children with autism to
learn vocabulary: an experimental evaluation. Int. J. Hum.-Com-
put. Interact. 35(1), 1–26 (2018). https​://doi.org/10.1080/10447​
318.2017.14200​06
79. Basil, C., Reyes, S.: Acquisition of literacy skills by children with
severe disability. Child Lang. Teach. Ther. 19(1), 27–48 (2003)
80. Hetzroni, O.E., Shalem, U.: From logos to orthographic symbols:
a multilevel fading computer program for teaching nonverbal
children with autism. Focus Autism Other Dev. Disabil. 20(4),
201–212 (2005)
81. Hetzroni, O.E., Tannous, J.: Effects of a computer-based interven-
tion program on the communicative functions of children with
autism. J. Autism Dev. Disord. 34(2), 95–113 (2004)
82. Massaro, D.W., Bosseler, A.: Read my lips: the importance of
the face in a computer-animated tutor for vocabulary learning by
children with autism. Autism 10(5), 495–510 (2006). https​://doi.
org/10.1177/13623​61306​06659​9
83. Moore, M., Calvert, S.: Brief report: vocabulary acquisition for
children with autism: teacher or computer instruction. J Autism
Dev Disord 30(4), 359–362 (2000).
84. Whalen, C., Moss, D., Ilan, A.B., Vaupel, M., Fielding, P., Mac-
donald, K., Cernich, S., Symon, J.: Efficacy of TeachTown: basics
computer-assisted intervention for the intensive comprehensive
autism program in Los Angeles Unified School District. Autism
14(3), 179–197 (2010). https:​ //doi.org/10.1177/136236​ 13103​ 6328​
2
Publisher’s Note Springer Nature remains neutral with regard to
jurisdictional claims in published maps and institutional affiliations.
13