Content uploaded by Fadi Abu-Amara
Author content
All content in this area was uploaded by Fadi Abu-Amara on Aug 19, 2022
Content may be subject to copyright.
ORIGINAL RESEARCH
Robot and virtual reality-based intervention in autism:
a comprehensive review
Fadi Abu-Amara
1
•Ameur Bensefia
1
•Heba Mohammad
1
•Hatem Tamimi
1
Received: 26 September 2020 / Accepted: 15 July 2021
Bharati Vidyapeeth’s Institute of Computer Applications and Management 2021
Abstract Autism Spectrum Disorder is a neurological and
developmental disorder. Children diagnosed with this dis-
order have persistent deficits in their social-emotional
reciprocity skills, nonverbal communication, and develop-
ing, maintaining, and understanding relationships. Besides,
autistic children usually have motor deficits that influence
their imitation and gesture production ability. The present
study aims to review and analyze the current research
findings in using robot-based and virtual reality-based
intervention to support the therapy of improving the social,
communication, emotional, and academic deficits of chil-
dren with autism. Experimental data from the surveyed
works are analyzed regarding the target behaviors and how
each technology, robot, or virtual reality, was used during
therapy sessions to improve the targeted behaviors. Fur-
thermore, this study explores the different therapeutic roles
that robots and virtual reality were observed to play.
Finally, this study shares perspectives on the affordances
and challenges of applying these technologies.
Keywords Autism Spectrum Disorder Robot Virtual
reality Autism Augmented reality
1 Introduction
Autism Spectrum Disorder (ASD) is a neurological and
developmental disorder. Children diagnosed with ASD
have persistent deficits in developing, maintaining, and
understanding relationships. Autistic children also have
deficits in social-emotional reciprocity and nonverbal
communication skills. Besides, they usually have motor
deficits that influence their imitation and gesture produc-
tion ability. There are two categories of motor deficits: the
basic motor deficit and the praxis performance, including
children with difficulty in social, communication, and
behavioral skills. Although all autistic children share
common characteristics, the precise degree to which each
child is affected significantly differs [1].
Different therapeutical methods exist to communicate
with autistic children and improve their skills. However,
choosing a suitable method imposes a great challenge since
autism needs a comprehensive intervention to aid diverse
autistic children most appropriately. Besides, teaching
autistic children is usually associated with high therapy
costs [2]. Recently, the usage of new technologies, such as
the robot, virtual reality (VR), and augmented reality (AR),
in autism therapy has been considered, motivated by the
fact that children with ASD prefer to engage socially with
objects [3]. Moreover, the preference for interaction with
these technologies is due to their predictable, enjoyable,
interactive, and straightforward environment [4]. Further,
during the learning process, no need for children to con-
sider socio-emotional expectations, which should decrease
their social anxiety. Finally, the current advancement in
these technologies offers eye contact, non-verbal commu-
nication skills, and self-initiated interactions, to list a few.
&Fadi Abu-Amara
fabuamara@hct.ac.ae
Ameur Bensefia
abensefia@hct.ac.ae
Heba Mohammad
hmohammad@hct.ac.ae
Hatem Tamimi
htamimi@hct.ac.ae
1
Computer and Information Sciences Department, Higher
Colleges of Technology, P.O. Box: 25026, Abu Dhabi, UAE
123
Int. j. inf. tecnol.
https://doi.org/10.1007/s41870-021-00740-9
Table 1 Experimental data summary
Targeted
behavior
Participants Method Challenges Robot
Eye contact
Imitation
Repetition
words
7–12 years old
Mild autism
Three game modules executed by NAO
that cover drawing a number, math game,
and emotional gesture
Each module starts with NAO introductory
rapport
A teacher/guardian must be
present
Room setup such as lighting
NAO [24]
Talking
Literary reading
No ASD sample is chosen Multimodal picture book recommendation
framework
Textual info extraction, an image info
extraction, and a multimodality info
integration module
Finding research on topic
obtainment
Providing parents with training
courses
NAO [26]
Gestural
recognition
and production
Twelve children aged
6–12 years
Low-functioning autism
Use NAO robot as a teaching agent for
gestures
NAO engaged in daily life conversations
and demonstration
Determining suitable learning
environments
NAO [27]
Attention
Visual
stimulation
Seven children aged
5–14 years
Moderate to severe autism
Tangram puzzle game
Graded cueing feedback
NAO asked for help during his turn to
stimulate the child’s cooperation
Participants interested in the robot
decreased with time due to
habituation to it
NAO [22]
Attention
Gazing
Three children aged
6–12 years
Language disabilities and
poor communicative
skills
A network of sensors to capture the
children gazing
Dog and cat different pictures placed at
two opposite sides of the room
Availability of complete data
about children
Limited degrees of freedom
NAO [21]
Interaction
Early detection
of ASD
1–14 years An observation-based autism screening
system
Six questions were adapted to the robotic
context
The robot collected response directly from
toddlers
Data extrapolation NAO [28]
Interaction
Emotion
One typically developed
child
Partially observable Markov decision
process to model the diagnostic protocol
Manual setup per each child
Open-space environment
NAO [30]
Eye contact Three children aged
6–13 years
High functioning ASD
An AI system based on robot-assisted
treatment of autism
Detects the child’s non-verbal signals
Exercises focus on: eye contact, joint
attention, body imitation, facial
imitation, facial expression imitation
The therapist has to assign each
exercise to a child according to
the functioning levels
NAO [31]
Interaction Five children aged
5–8 years
moderate to severe autism
Face recognition
Eye gaze detection
The two degrees of freedom PABI
has
PABI [33]
Interaction No ASD sample is chosen Applied Behavioral Analysis therapy
Aid diagnosing autistic children
PABI has only 8 degrees of
freedom
PABI [20]
Interaction Eighteen children aged
4–6 years
Moderate ASD
Complex and semi-unstructured
interaction in ASD therapy
Fully autonomous robot
Maintain a fully autonomous robot ASTRO [1]
Tactile
interaction
Eight boys aged 6–9 years Teaching how to identify human body
parts
Teaching triadic relationships
KASPAR is a semi-autonomous
robot
Kaspar [36]
Imitation
Attention
Turn-taking
Seventy-three
professionals and adults
with ASD
Evaluate robot, end-user, environment, and
practical implementation requirements
Manual analysis of collected data
The sample is limited to one
country
Kaspar [37]
Int. j. inf. tecnol.
123
Robots’ use as therapeutic tools for individuals with
ASD can be considered according to three angles: aca-
demic, industry, and legal development [5].
•Academic development: In 1995, artificial, social
intelligence, and autonomous robots were proposed.
Social intelligence robots can complete different tasks
without any help. Further, different hypothetical sce-
narios were suggested for robot-human and robot-robot
communication and cooperation. At that time, robots
were controlled in a well-structured environment.
However, few robots showed a degree of intelligence
concerning robustness, autonomy, flexibility, and
adaptability [5].
•Industry development: the literature indicates a grow-
ing industry and a significant demand for educational
robot technology [6]. Robokind, Aldebaran, and
Origami Robotics are among the leading companies
in manufacturing robots for autistic people. The first
company, Robotkind, provides the market with
Table 1 continued
Targeted
behavior
Participants Method Challenges Robot
Facial
expression
recognition
Emotions
Forty-five children
High functioning ASD
Game-based scenario
Storytelling to improve emotions
High manufacturing costs ZECA [38]
Cognitive
flexibility
Interaction
Forty-one autistic children
aged 4–13 years
Forty typically developed
children aged 4–7 years
Each participant goes through robot
interaction and human interaction
Probabilistic reversal learning task
The robot may distract the ASD
children
Repeated measures that may
interfere with results
Keepon [10]
Social and self-
regulation
skills
Nineteen children with
ASD
Student and Milo engage through social
narratives
Milo delivers lessons
Some children have a newfound
ability to communicate
Milo [42]
Interaction
Communication
and social
skills
Three children with ASD Specially designed educational scenarios
Enhance short-term and long-term memory
The therapist manually evaluate
the observation sheets
Pepper [84]
Attention
Facial
expression
skill
Fifteen parents of children
with low-level autism
aged 6–15 years
Six autistic children aged
7–12 years
Train autistic children on recognizing
facial expressions
Parents answer survey questions
Limited robot’s skills ifbot [45]
Interaction Eleven autistic children
aged 5–8 years
Individual interactive session
Children engage in self-directed
exploratory play
The experimenter asked a series of pre-
established questions
Simple mechanical toy dog AIBO [46]
Attention
Imitation
Four autistic children aged
4–5 years
Low functioning autism
Imitation of body movements
Evaluation sheet that lists child’s
expressions/reactions observed
Observing imitation for low
functioning autistic children
Tito [47]
Social
interaction
Sixty autistic children
aged 5–16 years
Free play and structured games
Single autistic subjects
The robot starts with a free play period and
then switch to a structured game session
The robot requires three
interventions from developers
TeoG [85]
Emotional and
social skills
Four children with ASD Educational games and activities
Identify interactive behaviors
Measure amount of social engagement
Integrating quantitative data with
qualitative data
Aisoy [48]
Communication
skills
Interaction
Twenty children with
ASD
Introduce each child to the robot, talk
freely, and play
Conversation and playtime
The robot has limited skills and
degrees of freedom
Zoomorphic
[86]
Int. j. inf. tecnol.
123
advanced social robots such as the Milo robot [7]. The
Aldebaran company develops educational and thera-
peutical tools for autistic children with the NAO robot
as the main product. The Origami company manufac-
tures the Romibo robot, which offers expressive eyes
and can track eye contact.
•Legal development: in 1975, the Individuals with
Disabilities Education Act (IDEA) was proposed as a
federal statute to formulate special-need education and
determine the feasibility of educational robotic systems
for public schools of the United States [8]. The IDEA
was revised in 2004. According to the Assistive
Technology clause, long experience and research can
improve special-education children’s learning experi-
ence by utilizing assistive technology devices and
assistive technology services.
A strange phenomenon is worth discussing regarding the
use of robots as therapeutic and educational tools for aut-
ism. The best practices of using the robot in the autism field
are mainly determined by industry professionals who
mostly lack academic training or clinical experience.
External input should improve any system. However, we
end up having autistic therapists and educational practi-
tioners reacting to assistive technology instead of proac-
tively aiding its development [5]. The correct order of
influence on the autism field should be clinical therapists,
followed by educational practitioners, and followed by the
industry’s direction.
In this paper, we review and analyze the current autism
works, mainly based on robots, and to a lesser extent,
virtual reality and augmented reality in supporting the
therapy of improving the social, communication, emo-
tional, and academic deficits of children with autism.
Furthermore, experimental data from the surveyed works
are analyzed regarding the targeted behaviors and how
each technology was used during therapy sessions to
improve the targeted behaviors. Moreover, this study
explores the different therapeutic roles that robots and
virtual reality were observed to play. Finally, this study
shares perspectives on the affordances and challenges of
applying these technologies.
2 Search and selection method
An online search had been conducted using systematic
literature search procedures. The review focused on the
online scientific databases relevant to the addressed sub-
ject, such as PubMed for medical science, Embase for
biomedical literature, PsycINFO for behavioral science and
mental health, Scopus, and Web of Science. In addition, we
searched interdisciplinary databases that include
international conference proceedings and journal papers
related to science, information technologies, and health
science. The selection of the articles was focused on the
following:
•The academic peer-reviewed papers were published in
English journals and conference proceedings from 2002
to 2021.
•The search query considered the following terms:
autism, autistic children, virtual reality, augmented
reality, robot, robot-based intervention, social behavior,
or virtual education.
•The selection was limited to the articles associated with
studies focusing on using robots and virtual reality for
autistic children. Studies that focus on other technolo-
gies and in different languages had been excluded.
•Each author had to review papers and focus during the
reading process on the title, abstract, methodology, and
conclusion sections.
•The inclusion criteria of the selected articles consid-
ered: (1) the robot-based intervention or virtual reality-
based intervention as well as human-based intervention
and robot-based intervention (2) Targeted participants
are children diagnosed with autism or typically devel-
oped children.
•The exclusion criteria considered papers covering other
technologies not related to robots or virtual reality.
3 Robot-based intervention
The choice of the robot to be used in autism therapy is a
crucial factor. Indeed, several considerations must be
considered, such as the robot’s interaction quality and
structure. The interaction quality explores the level of
social interaction between children and robot. For example,
some robots offer a straightforward form of social inter-
action, while others provide different forms of advanced
social interaction. On the other hand, the interaction
structure explores the type of activities offered during the
interaction. For example, some robots offer unstructured
activities, such as free play, while other robots offer highly
structured activities.
Robot autonomy is another critical point to consider
when developing therapies for autism. Some robots are
fully autonomous, while other robots are remotely con-
trolled during the interaction. Remotely controlled robots
adopt the Wizard-of-Oz (WoZ) method, where a human
operator remotely controls a robot [9]. Many researchers
adopted this approach since human perception is used to
overcome autonomous robots’ perceptual challenges. Fur-
ther, a remotely controlled robot offers productive inter-
actions and divers’ social behaviors. It is worth mentioning
Int. j. inf. tecnol.
123
that remotely controlled robots provide broad features and
interactions that vary from simple ones where the robot can
emotionally respond to a child through engagement like the
Keepon robot [10] and Pleo robot [11]; to other complex
interactions like playing turn-taking games or imitation
such as the Kaspar robot [12]. Throughout scenarios, a
therapist meditates a child during interaction with a robot.
The WoZ setup may require more than one therapist, apart
from the robot’s technical staff. Further, human operators’
responsibilities increase, which demands them to process
many inputs and handle complex actions since the inter-
actions with children become more complicated. There-
fore, more human operators are needed to provide a rich
and multimodal interaction for an autistic child [1].
Nevertheless, the long-term dependence of the WoZ
method on human operators makes it a costly solution for
autism therapy [13]. In a previous work involving auton-
omous robots, two additional clusters, mainly differ in the
child-robot interaction, were identified. The first cluster
covered autonomous robots that included works that
introduced robots like Bubbleblower [14] and Roball [15],
to list a few. These robotic systems allow only the simplest
form of interaction and exhibit no social interaction.
Experimental results indicate that children interacted with
robots during their free play, in an unstructured interaction,
where these robots behave just little more than sophisti-
cated toys [14,15]. The other cluster is marked as auton-
omous social robots. It includes works that use robots like
NAO [13,16], Infanoid [17], and iRobiQ/CARO [18], to
list a few. It is worth mentioning that robots that are
dependable enough to interact with humans while operat-
ing autonomously over an extended period were rarely
reported in the literature.
Several guidelines were suggested for the ASD domain
as a roadmap to robot-mediated intervention and evidence-
based practice in autism [19]. The guidelines describe
intervention objectives, participants, dependent and inde-
pendent variables, research design, and training procedure.
In addition to the listed considerations, researchers identi-
fied that selecting a robot for the autism field could be
driven by [20–22]:
•Shape: robots can take different forms such as android,
human-like, and animal-shaped, to list a few. This may
distress the children during the therapy sessions.
•Embodiment: social robots’ physical presence allows
them to do physical explorations and interactions with
their environment, using gestures and touch to com-
municate with children.
•Complexity: using robots can dramatically decrease the
complexity of interactions between children and peo-
ple. The robots can be programmed to focus on a few
skills and only one interaction aspect.
In this section, an exhaustive list of the most relevant
research that adopted social robots in autism studies is
presented based on the type of used robot.
3.1 NAO robot
The NAO robot introduces an ideal platform to improve
different skills among autistic children. The NAO is
designed to look approachable and to express emotions like
a toddler. NAO has various features such as autonomous,
19 languages, fully programmable, grasping small objects,
understanding spoken words and confine sounds, walking,
wandering, and distinguishing people. The NAO is equip-
ped with different hardware such as cameras, microphones,
motors, legs, head, hands, and arms. All the conversations
between NAO and children are recorded for future analy-
sis. NAO is the most used robot-based intervention for
research, education, and healthcare [2]. The NAO achieved
a higher gazing time with kids than a human therapist [23].
Besides, children who interacted with NAO achieved
higher scores on communication behavior tests. Many
researchers used NAO to improve the communication and
social skills of autistic children [24]. For example, picture
books can improve reading, understanding, expressing, role
reversing, and emotional skills [25]. Each picture book
contains image information and textual information.
The NAO robot was used to develop a multimodal
picture book recommendation framework based on the
conversation content [26]. The proposed framework con-
sisted of textual information extraction, image information
extraction, and a multimodality information integration
module. However, the proposed framework needs
improvement with more competent association functions.
Besides, therapists and parents should be involved in the
experiments.
A study compared the learning outcomes of NAO robot-
based intervention against human-based intervention in
teaching autistic children the use of intransitive gestures
[27]. The study used production assessments that evaluated
children’s attention, fine motor skills, and visual percep-
tion. Two NAO robots were programmed to offer fourteen
different intransitive gestures and utilized to perform role-
play. Results indicated that autistic children from both
groups successfully recognized the various gestures and
produced them accurately. However, more research is
required to determine suitable learning environments and
investigating nonverbal communication skills.
The tangram puzzle game was proposed to assist chil-
dren [22]. The tangram puzzle was made up of seven
geometrical pieces of different shapes used by children
during therapy sessions. In addition, the graded cueing
feedback was implemented, where children facing diffi-
culties during the game were given gradual cues or
Int. j. inf. tecnol.
123
prompts. Occasionally, the NAO robot asked for help
during his turn to stimulate the child’s cooperation. All
participants showed considerable interest in the robot
during the first sessions. However, this interest decreased
over time due to habituation.
In [21], the dynamics of joint attention for children with
ASD were investigated. A network of sensors was used to
capture the children gazing, displacement, and kinematic
energy. The NAO robot and the sensors were placed on a
table to keep them at the same level as the child’s head.
Two different pictures, dog and cat, were placed at two
opposite sides of the room. The robot tried to draw the
children’s attention to any of the photos in the room by
increasing the amount of information and using different
modalities such as gazing and head movement. Results
showed that the participants with ASD spend less time
focusing on the target picture than the tardive dyskinesia
(TD) participants in terms of gazing. Also, in terms of
displacement, the ASD participants had a higher amplitude
on the ground.
In [28], The NAO robot was used to develop an obser-
vation-based autism screening system. Six questions from
the Quantitative Checklist for Autism in Toddlers were
adapted to the robotic context. The robot collected
responses to the six questions directly from toddlers rather
than from their parents. Another work was conducted to
study the possibility of utilizing NAO and Pepper robots as
a robot-assisted diagnostic protocol [29]. The partially
observable Markov decision process was used to model the
diagnostic protocol tasks to automate robot actions. Results
showed definite signs of the robot’s interaction and
observation of children’s behavior. Another study was
done on a typically developed child, instead of an autistic
child, since he can express emotions and handle stress [30].
The NAO robot was used to detect the eye contact
behavior of three children diagnosed with high functioning
ASD [31]. Each session lasted for 20 min, where the eye
contact exercise was performed 15 times. The eye-detec-
tion algorithm was based on the Viola–Jones detector.
Results indicate that children need help to complete all
levels. In [32], four single-subject experiments were con-
ducted to compare the social engagement level of autistic
children with NAO robot versus humans. All sessions, used
motor imitation exercise and joint attention, indicated that
children showed interest in NAO, especially at the begin-
ning. This interest diminished with time, especially with
the severe autism case. The study concluded that NAO is
considered a better facilitator for moderate autism level
children.
3.2 PABI robot
The PABI is a cartoonish robot specifically designed to
provide applied behavioral analysis (ABA) therapy to
children with autism using the discrete trial training
method [20]. Due to its penguin-like shape, it is called
Penguin for Autism Behavioral Intervention (PABI). PABI
provides all ABA therapy levels, considering the children’s
psychological underpinnings. The PABI robot’s eyes
include stereo cameras for tracking the child’s movements
and modeling eye contact through monitoring children’s
attention to the therapy sessions and chart their progress.
In [33], a PABI-based intervention was developed. The
proposed system consisted of two phases: face recognition
and eye-gaze detection. The histogram of oriented gradi-
ents extracted features invariant to lighting, object occlu-
sion, and orientation changes for the face detection phase.
Then, a linear classifier was used to classify positive and
negative feature samples. Finally, a multi-scale sliding
window search was used to detect skewed images. In [20],
PABI-based intervention achieved better interaction results
than human-based intervention.
3.3 Astro robot
Unlike teleoperated robots, the Astro robot system works in
a fully autonomous way. Manufacturers state that Astro
offers enhanced social interaction capabilities than auton-
omous social robots. For example, the developed system
promotes and explains different therapeutical activities,
provides feedback on various tasks, asks for assistance, and
offers a rich set of activities wherein the child can move
freely around the therapy room [1]. It is worth mentioning
that allowing the child to move freely around a room
presents significant challenges in terms of perception. This
is due to the need to monitor the whole room space and
autonomously differentiate the child from all other objects.
The child performs different activities during a therapy
session, such as finding hidden balls, solving a geometric
puzzle, playing a turn-taking game, and assisting the robot
in moving from one place to another. The robot encourages
a child to perform certain activities, explains how to per-
form activities, and rewards through positive
reinforcements.
To develop an autonomous robot platform capable of
interacting with a human, the robot must act in situations
where neither the human nor the robot can fully complete
the task without the other’s assistance. The Astro robot can
perform activities such as saying hello, giving an intro-
duction, tracking the child’s collection of colored balls
hidden in the room, improving a child’s empathy and social
reciprocity by asking for a child’s help in removing some
obstacles from the robot’s path, and improving social-
Int. j. inf. tecnol.
123
communicative behavior skill by encouraging children to
make requests.
3.4 Kaspar robot
The Kaspar robot has different features such as singing,
playing games and tambourine, combing its hair, and
imitating eating [34]. In addition, the Kaspar robot is
equipped with touch sensors to differentiate between soft
and harsh touch and consequently respond [35]. Kaspar is a
semi-autonomous humanoid robot that acts as a social
companion to improve the communication difficulties of
autistic children. The Kaspar robot allows three distinct
operation modes. An example of the autonomous control
mode is when a child activates the robot’s sensors, such as
arm and feet sensors. In the controlled operating mode, a
therapist, or professional controls the robot via a remote-
control device. Finally, the semi-autonomous mode
includes a combination of the two modes.
The Kaspar robot was used to teach children how to
identify body parts and appropriate physical interaction
[36]. In [37], 73 autistic adults participated in different
focus groups to draw out requirements for robot-assisted
intervention. Moreover, 22 participants, children’s parents,
professionals, and autistic adults, attended sessions to
generate ideas for interventions. It is recommended to add
to Kaspar the ability to fetch, grasp, manipulate, and hold
objects [37].
3.5 ZECA robot
The Zeno Engaging Children with Autism (ZECA) robot is
a humanoid robot that can walk and has a human-inspired
character face and gestural body that enables the robot to
do facial expressions [38]. In [38], a game-based scenario
was used, where 45 children diagnosed with high func-
tioning ASD had been divided equally and randomly into
three groups. The first group was exposed to robot-inter-
vention settings. The second group was exposed to human-
intervention settings only. The third group was considered
a control group; they only have to do pre-test and post-test
sessions. The research consisted of four phases: familiar-
ization phase, pre-test phase, practice phase, and post-test
phase. Results showed that the robot could partially con-
tribute to development of facial expression recognition
skills.
3.6 Keepon robot
The Keepon robot was used to study the interpersonal
coordination and social development skills of autistic
children [39]. In [10], the Keepon robot was used to
examine how 41 autistic children and 40 typically
developed children perform cognitive flexibility tasks.
During an interaction with the robot, autistic children
showed more task engagement and enjoyment than human
interaction. Further, there was no significant difference in
the robot-based intervention and human-based interven-
tion’s cognitive flexibility performance. Finally, during the
learning stage, children’s performance was improved dur-
ing robot sessions.
3.7 Milo robot
The Milo robot is developed to aid autistic children in
improving their communication skills [40]. Milo has
advanced social features, uses facial and vocal expressions
to interact with children, and can listen or tell a story. The
Milo robot can be used to improve feelings and sentiments.
It also provides calming aptitude and two-sided discussion
[41]. The study in [42] showed that the Milo robot had
improved self-regulation, social, and academic skills. Stu-
dents enjoyed interaction and dancing with the Milo, which
was a reward for good behavior. The technology can be
used to improve social, communication, and emotion reg-
ulation skills. However, using a computer-based or tablet-
based solution resulted in autistic children who interacted
with technology only and avoided interaction with humans
[43]. Thus, Milo was utilized to deliver verbal interaction
and social narratives resulting in reduced repetitive actions
[43].
3.8 Other robots
The Pepper robot has different features such as a high-level
human interaction, advanced capabilities, attractive face
design, emotional robot, advanced voice recognition ability
of several variations in the human voice, cameras, ability to
use body language, and the ability to recognize 20 lan-
guages, to list a few. The Pepper robot was used to improve
the emotions and learning of special-need children [44]. A
study evaluated different robot features based on their
impact on improving the social and communication skills
of autistic children [45]. The ifbot, a small sphere-shaped
robot, was used in the experiments. Results indicated that
the robot’s face and moving limb drew children’s attention
and improved facial expression skill. However, it did not
improve other social and communication skills. Further,
the robot’s verbal communication feature achieved better
interaction with low-functioning autistic children than
human interaction. The AIBO Dog robot improved mutual
interaction, verbal engagement, and authentic interaction
[46]. Another work utilized the Tito Mobile robot with two
autistic children. Robot-based intervention results indicated
improved attention in visual contact and physical proximity
and imitating smiling compared to the human-based
Int. j. inf. tecnol.
123
intervention [47]. In [48], the Aisoy1 is an interactive robot
that performs educational games and activities to improve
emotional, communication, self-confidence, self-stability,
and kinetic and cognitive skills [48].
3.9 Discussion
Despite the presented research using robots with autistic
children, the evidence of robot-based treatment’s effec-
tiveness is still limited. Indeed, this is mainly because
many studies did not cover all types of deficits related to
autism, such as social, communication, emotional, educa-
tional, behavior, imitation, and gesture production abilities.
Further, many papers based their developed methods on the
fact that autistic children share common characteristics.
However, these papers did not pay much attention to the
fact that each child is affected by autism to a different
degree. This makes it extremely difficult to develop a
comprehensive robot-based intervention solution that aids
diverse autistic children in the most suitable way. Also,
using an appropriate number of robots, and the dependence
on human operators to control the robots are usually
associated with high therapy costs and other overheads.
Further, the robot’s integration as a worldwide educa-
tional tool for students is still early. Also, the best practices
of using the robot in the autism field are mainly determined
by industry professionals rather than autistic therapists and
educational practitioners. In addition, guidelines were
suggested as a roadmap to robot-mediated intervention but
still not widely standardized. Moreover, not all studies
involved autistic therapists and parents in the experiments.
We summarized the experiments discussed in the pre-
vious sections [1,10,20–22,24,26–28,30,
31,33,36–38,42,45–48,85,86] in Table 1, and in Fig. 1
in terms of performances. Figure 1A shows statistics of the
targeted skills in the surveyed papers. As the figure shows,
the most targeted skills are Interaction, Attention, eye
contact, and facial expressions. Figure 1B shows statistics
of the targeted autistic children ages. As the figure shows,
the most targeted ages are 5–8 years. Further, the average
number of targeted children per experiment is 17.2, where
some works targeted one child while others targeted 73
children.
4 Virtual reality-based intervention
Virtual reality (VR) offers a virtual environment and
interactive video gaming, based on computer systems,
where sound and sight are the primary sensory stimuli.
Virtual reality is considered an interactive intervention tool
in the autism field, where actions received from autistic
children are partially reflected in the artificial environment.
The VR utilizes various technologies, such as monoscopic
display, stereoscopic display, user technologies, and aug-
mented reality, to integrate the virtual world with the nat-
ural world [49].
The literature reported using virtual reality as a treat-
ment tool in different areas such as diagnosis [50], reha-
bilitation [51], training on surgery [52], and raising
patient’s emotional wellbeing [53], to name a few. More-
over, virtual reality can treat mental disorders such as post-
traumatic stress disorders, phobia, autism spectrum disor-
der, and obsessive–compulsive disorders. In addition, VR
is also used to improve various social interactions, com-
munication, emotional response, social skills, and execu-
tive functions.
The use of virtual reality in treating autism spectrum
disorder includes many benefits over traditional interven-
tions, such as offering a controlled and safe training
environment, adapting the environment to the skills of the
autistic child, allows emulating everyday life skills, and
providing a realistic training environment [49]. Further-
more, the VR offers the ability to receive different mea-
sures of the child’s performance, offer suitable real-time
feedback, and adapt the scenarios [54].
If we consider a sample of literature papers, the targeted
autistic children consisted of 142 boys and 22 girls aged
between four and seventeen years with an average age of
10 years [55–75]. In this sample, five papers utilized vir-
tual reality scenarios, environments, and avatars to improve
emotional skills [55,56,59,69,70]. Another article used
virtual reality scenarios and driving modules to enhance
daily living skills such as shopping in a supermarket and
driving [76]. Another work utilized virtual reality and
augmented reality in improving the communication skills
of autistic children [62,73]. In [56], virtual reality sce-
narios and mobile object identification were used to
improve attention skills. Other work utilized virtual reality
devices, such as SenseW armband, BodyMedia, and
immersive stereoscopic surround-screen, in improving
children’s physical activity [74]. Another experiment used
a blue room virtual reality environment as an intervention
tool to reduce phobia [75].
One of the virtual reality applications in the autism field
is reducing the habit of unconsciously copying other peo-
ple’s actions. Studies show that autistic children usually
copy the observed activity’s goal with differences in their
mimicry [77]. A two-dimensional VR environment was
used to induce the mimicry where 26 autistic children
played an imitation game with two avatars; socially
engaged and socially disengaged [77]. The virtual reality
graphics were displayed on a projector screen. Autistic
children used two electromagnetic markers to record their
finger movements and to establish eye contact with avatars.
Results indicate that participants mimicked avatar motions,
Int. j. inf. tecnol.
123
with minor differences, to achieve the observed action.
Results also suggest that VR can be used as a training and
diagnostic tool.
In [78], an interactive scenario-based VR system, made
up of five modules, was developed to improve the com-
munication skills of autistic children. These modules are
the display and trackers module, speech recognition mod-
ule, gesture recognition module, virtual environment and
avatars module, and auto-navigation module. Further, a VR
classroom greeting scenario was developed, where an
interaction between the avatar and the children took place.
Two groups of children were used; ASD children and
typically developed children. Two scenarios of 20 min
each, on two separate days, were proposed to each partic-
ipant on three different setups of the VR system: desktop
environment, computer augmented reality environment
(CAVE), and head-mounted display (HMD). Results
showed that for both ASD and TD children, a satisfaction
rate exceeds 82% for CAVE and 78% for HMD. For the
desktop environment, the performances were averaged at
60%. Also, the children with ASD showed more interest in
desktop environment comparing to TD children.
The study in [66] used virtual reality technology as a
therapeutic tool to treat phobias in autistic children. Results
showed a decrease in phobia intensity in 25% of the chil-
dren over two therapy sessions. Another work developed a
tutoring system based on integrating the NAO robot and
virtual reality [65]. In this work, a virtual teacher taught
sight words to three autistic children and the NAO robot
that emulated a peer. Experimental results indicated that
autistic children could memorize and absorb all the sight
words explicitly as instructed. Further, children learned
about 94% of the sight words exclusively taught to the
NAO robot. In [79], an interactive learning environment
was developed for behavioral training for children with
autism based on the picture exchange communication
system and face-to-face interviews. The participants of the
study consisted of children and their parents. A virtual
toilet was designed for behavioral training per request from
parents. Results indicated that younger parents were more
nervous and depressed while dealing with their children,
while older parents showed a better understanding.
In [80], a VR environment was developed for autistic
children to improve their communication and social skills.
The VR environment used a fish shop game as an inter-
active environment, where the autistic children had to use a
wearable Oculus Rift VR headset and a microphone. The
fish shop game was chosen since it had fewer social skills
and behavior challenges where the context was related to
pets and animals like dogs, cats, and fish. In addition, the
system was built to track the child’s eye contact during the
conversation to record visual attention in a safe
environment.
Despite the presented research in using VR with autistic
children, the evidence of VR-based treatment’s effective-
ness is still limited. This limitation is because many studies
did not use a control group of autistic children with the
same characteristics that received other intervention-based
treatments compared to VR-based intervention [49].
Besides, the same questionnaire should be conducted at the
same time during the intervention program. Also, most of
the studies used a low number of autistic children in their
experiments. Furthermore, some works researched male
autistic children only while others did not maintain the 3:1
male to female autistic children ratio in the chosen sample
[81]. Finally, some papers conducted their research on
high-performance autistic children. Therefore, results and
conclusions obtained from these papers cannot be gener-
alized to the autism field.
5 Augmented reality-based intervention
Children with autism suffer from a lack of understanding
and interpretation of social situations and generate the
appropriate response in the proper format. Therefore,
augmented reality (AR) has been considered to increase the
motivation of children with ASD. The authors in [82]
proposed implementing a social story through an aug-
mented reality environment to improve autistic children’s
Fig. 1 A statistics of targeted skills and Bstatistics of the targeted age
Int. j. inf. tecnol.
123
skills. A social story is a short story describing situations,
skills, interactions, and behaviors, in a written or a visual
format that ASD children can understand. The experi-
mentations had been conducted with three children
accompanied by their therapists. The therapists evaluated
the system’s performance, who filled out a questionnaire
about the level of conformity of the system. The Likert
scale and the obtained results showed that 71% of
respondents agree that the system can support social story
therapy.
In [83], a novel collaborative virtual environment (CVE)
platform using various technologies was developed to
enable autistic children to communicate and interact in a
naturalistic approach. The two remotely and separated
players should synchronize and coordinate their hand
movements to complete the game. The system included an
application module to manage game connection and exe-
cution, a communication module, and an eye tracker
module. A leap motion detector had been used to track the
player’s hand movement and gestures to manipulate the
virtual objects. Further, the CVE had three collaborative
games: puzzle game, collect game, and delivery game. The
proposed system was tested with 12 typically developing
(TD) children and 12 children with ASD. The 12 groups,
one TD child and one ASD child, were used to imitate real-
life interaction settings. In addition, a training session had
been conducted with six different collaborative games. The
study tested the acceptability of the system among autistic
children using five-Likert surveys. Results showed that
autistic children enjoyed the games, and their interaction
progressively improved.
6 Conclusion
Autistic children prefer to interact with new technologies,
such as robots, virtual reality, and augmented reality, since
these technologies offer predictable, enjoyable, interactive,
and straightforward environments. Further, during the
learning process, there is no need for autistic children to
consider socio-emotional expectations, which should
decrease their social anxiety. The current advancement in
these technologies offers eye contact, non-verbal commu-
nication skills, and self-initiated interactions, to list a few.
In this paper, which represents the first phase of a multi-
year project, the recent research findings in utilizing robots
and virtual reality as intervention tools in supporting the
therapy of improving different skills among autistic chil-
dren are investigated. Despite the presented research using
the robot and virtual reality interventions for autistic chil-
dren, the evidence of treatment’s effectiveness is still
limited. Many studies did not cover all types of deficits,
based their developed methods on the fact that autistic
children share common characteristics, followed best
practices of the industry professional instead of autistic
therapists and educational practitioners, or did not involve
autistic therapists and parents in the experiments.
To build a networked robot system capable of effec-
tively participating in different activities, several techno-
logical challenges must be addressed, such as the robot’s
ability to perceive persons and the objects in the room,
robot motion and head rotation to apply the mind-reading
mechanism, robot’s decision-making system to consider
the existing state of the environment and plan the subse-
quent actions, and additional features such as an LCD for
facial expressions. These features aim at attracting and
focusing the child’s attention as well as improving the
interaction. Finally, symbiotic interaction should also be
considered in case the robot cannot perform some actions.
The second phase of this project should focus on
developing a comprehensive technology-based intervention
considering that each child is affected by a different degree
of autism. The developed solution should also reduce
therapy costs, reduce dependence on human operators,
follow autistic therapists and educational practitioners’
practices, involve parents in the experiments, and involve
more symbiotic interactions.
Funding The authors would like to acknowledge Higher Colleges of
Technology for their financial support to this research grant.
References
1. Melo FS, Sardinha A et al (2019) Project INSIDE: towards
autonomous semi-unstructured human–robot social interaction in
autism therapy. Artif Intell Med 96:198–216
2. Yousif JH, Kazem HA, Chaichan MT (2019) Evaluation imple-
mentation of humanoid robot for autistic children: a review. Int J
Comput Appl Sci 6(1):412–420
3. Srinivasan S, Eigsti I, Neelly L, Bhat A (2016) The effects of
embodied rhythm and robotic interventions on the spontaneous
and responsive social attention patterns of children with autism
spectrum disorder (ASD): a pilot randomized controlled trial. Res
Autism Spectr Disord 27:54–72
4. Dautenhahn K (2000) Design issues on interactive environments
for children with autism. In: Proceedings of 3rd international
conference on disability, virtual reality and associated technolo-
gies, pp. 1–8. http://web.mit.edu/16.459/www/Dautenhahn.pdf.
Accessed 2020
5. Wood Z (2018) Synthesizing and restructuring the conversation
around the use of robots in the education and therapy of indi-
viduals with Autism Spectrum Disorders, master thesis, Colum-
bia University
6. Chen A (2015). The Ever-Growing Ed-Tech Market. The
Atlantic. www.theatlantic.com/education/archive/2015/11/quanti
fying-classroom-tech-market/414244/. Accessed Sep 2020
7. Advanced social robots. http://www.robokindrobots.com.
Accessed Sep 2020
Int. j. inf. tecnol.
123
8. Your child’s rights: autism and school. https://www.autism
speaks.org/what-autism/your-childs-rights. Accessed 22 Sep
2020
9. Kelley J (1984) An iterative design methodology for user-friendly
natural language office information applications. ACM Trans Inf
Syst 2(1):26–41
10. Costescu CA, Vanderborght B, David DO (2015) Reversal
learning task in children with autism spectrum disorder: a robot-
based approach. J Autism Dev Discord 45(11):3715–3725
11. Kim ES, Berkovits LD, Bernier EP, Leyzberg D, Shic F, Paul R,
Scassellati B (2013) Social robots as embedded reinforcers of
social behavior in children with autism. J Autism Dev Disord
43(5):1038–1049
12. Robins B, Dautenhahn K, Dickerson P (2009) From isolation to
communication: a case study evaluation of robot assisted play for
children with autism with a minimally expressive humanoid
robot. In: 2009 second international conferences on advances in
computer-human interactions, pp. 205–211
13. Esteban P, Baxter P, Belpaeme P, Billing E, Cai H, Cao H et al
(2017) How to build a supervised autonomous system for robot-
enhanced therapy for children with autism spectrum disorder.
J Behav Robot 8(1):18–38
14. Feil-Seifer D, Mataric M (2009) Toward socially assistive
robotics for augmenting interventions for children with autism
spectrum disorders. In: Proceedings of 11th international sym-
posium on experimental robotics, vol 54, pp 201–210
15. Michaud F, The
´berge-Turmel C (2002) Mobile robotic toys and
autism: observations of interaction. Kluwer Academic Publishers,
London
16. Greczek J, Kaszubski E, et al (2014) Graded cueing feedback in
robot-mediated imitation practice for children with autism spec-
trum disorders. In: Proceedings of 23rd IEEE international
symposium on robot and human interactive communication,
pp. 561–566
17. Kozima H, Nakagawa C, Kawai N, Kosugi D, Yano Y (2004) A
humanoid in company with children. In: Proceedings of 4th
IEEE/RAS international conference on humanoid robots,
pp. 470–477
18. Yun SS, Kim H, Choi J, Park S (2015) A robot-assisted beha-
vioural intervention system for children with autism spectrum
disorders. Robot Auton Syst 76:58–67
19. Begum M, Serna RW, Yanco HA (2016) Are robots ready to
deliver autism interventions? A comprehensive review. Int J Soc
Robot 8(2):157–181
20. Dickstein-Fischer L, Fischer GS (2014) Combining psychological
and engineering approaches to utilizing social robots with chil-
dren with autism. In: Engineering in medicine and biology
society, pp. 792–795
21. Anzalone SM, Xavier J et al (2019) Quantifying patterns of joint
attention during human-robot interactions: an application for
autism spectrum disorder assessment. Pattern Recogn Lett
118:42–50
22. Bernardo B, Alves-Oliveira P, Santos MG, Melo FS, Paiva A
(2016) An interactive tangram game for children with auism.
Lecture notes in computer. Springer, New York
23. Ting Z (2013) Application in the intervention program of autism
based on NAO robot. Syst Simul Technol 9:327–331
24. Miskam MA, Masnina NF et al (2014) Encouraging children with
autism to improve social and communication skills through the
game-based approach. Proc Comput Sci 42:93–98
25. Miranda LS, Mary ET, Amy PA (2017) Using Children’s picture
books about autism as resources in inclusive classrooms
26. Yang X, Shyu ML et al (2019) Integrating image and textual
information in human-robot interactions for children with autism
spectrum disorder (ASD). IEEE Trans Multimed 21(3):746–759
27. So W, Wong MK et al (2019) Who is a better teacher for children
with autism? comparison of learning outcomes between robot-
based and human-based interventions in gestural production and
recognition. Res Dev Disabil 86:62–75
28. Romero-Garcı
´a R et al (2021) Q-CHAT-NAO: a robotic
approach to autism screening in toddlers. J Biomed Inform
118:103797–103797
29. Petric F, Miklic D, Kovacic Z (2017) Robot-assisted autism
spectrum disorder diagnostics using POMDPs. In: ACM/IEEE
international conference on human-robot interaction, pp 369–370
30. Salleh MHK, Miskam MA, Yussof H, Omar AR (2017) HRI
assessment of ask NAO intervention framework via typically
developed child. Proc Comput Sci 105:333–339
31. Palestra G, Carolis BD, Esposito F (2017) Artificial Intelligence
for robot-assisted treatment of autism. In: Workshop on artificial
intelligence with application in health, pp 1–8
32. Tapus A, Peca A, Aly A, Pop C et al (2012) Children with autism
social engagement in interaction with nao, an imitative robot—a
series of single case experiments. Interact Stud 13(3):315–334
33. Dickstein-Fischer LA, Pereira RH, Gandomi KY, Fathima AT,
Fischer GS (2017) Interactive tracking for robot-assisted autism
therapy. In: ACM/IEEE international conference on human-robot
interaction, pp. 107–108
34. Bharatharaj J, Huang L, Kra
¨geloh C, Elara MR, Al-Jumaily A
(2018) Social engagement of children with autism spectrum
disorder in interaction with a parrot-inspired therapeutic robot.
Proc Comput Sci 133:368–376
35. Ricks DJ, Colton MB (2010) Trends and considerations in robot-
assisted autism therapy. In: 2010 IEEE international conference
on robotics and automation, pp 4354–4359
36. Costa S, Lehmann H, Dautenhahn K, Robins B, Soares F (2015)
Using a humanoid robot to elicit body awareness and appropriate
physical interaction in children with autism. Int J Soc Robot
7(2):265–278
37. Huijnen CAGJ, Lexis MAS et al (2017) How to implement robots
in interventions for children with autism? A co-creation study
involving people with autism, parents and professionals. J Autism
Dev Disord 47:3079–3096
38. Hanson D, Baurmann S, Riccio T, Margolin R, Dockins T,
Tavares M, Carpenter K (2009) Zeno: a cognitive character. In:
AI Magazine, and special Proceedings Of AAAI national con-
ference, Chicago, pp 9–11
39. Peca A, Simut R, Cao HL, Vanderborght B (2016) Do infants
perceive the social robot Keepon as a communicative partner?
Infant Behav Dev 42:157–167
40. Chalmers C (2018) Robotics and computational thinking in pri-
mary school. Int J Child-Comput Interact 17:93–100
41. Richard M (2018) Parent perspective: the greatest investment any
school can make. Robots4autism. https://robots4autism.com/
author/richard/. Accessed 22 Sep 2020
42. IESD case study: children with asd make gains with robot-
s4autism in a Dallas, Texas Charter School (2017), pp 1–4
43. IESD case study: children on the autism spectrum show
improvement with robot4autism in Spartanburg, South Carolina
(2016), pp 1–4
44. Kamps HJ (2016) The autism solutions bot helps autistic kids.
https://techcrunch.com/2016/09/11/autism-solutions/. Accessed
22 Sep 2020
45. Lee J, Takehashi H et al (2012) Which robot features can stim-
ulate better responses from children with autism in robot-assisted
therapy? Int J Adv Robot Syst 9(72):1–5
46. Stanton CM, Kahn PH, Severson RL, Ruckert JH, Gill BT (2014)
Robotic animals might aid in the social development of children
with autism. In: Proceedings of the 3rd ACM/IEEE international
conference on Human-robot interaction, pp 271–278
Int. j. inf. tecnol.
123
47. Duquette A, Michaud F, Mercier H (2008) Exploring the use of a
mobile robot as an imitation agent with children with low-func-
tioning autism. Auton Robot 24(2):147–157
48. Albo-Canals J, Yanez C, Barco A, Angulo Baho
´n C, Heerink M
(2016) Modelling social skills and problem solving strategies
used by children with ASD through cloud connected social robots
as data loggers: first modelling approach. In: Conference pro-
ceedings New Friends 2015: the 1st international conference on
social robots in therapy and education, Netherlands, pp 1–2
49. Mesa-Gresa P, Gil-Gomez H, Lozano-Quilis JA et al (2018)
Effectiveness of virtual reality for children and adolescents with
autism spectrum disorder: an evidence-based systematic review.
Sensors (Basel) 18(8):1–15
50. Areces D, Rodrı
´guez C et al (2018) Efficacy of a continuous
performance test based on virtual reality in the diagnosis of
ADHD and its clinical presentations. J Atten Disord
22(11):1081–1091
51. Albiol-Perez S, Gil-Gomez JA et al (2017) The effect of balance
training on postural control in patients with Parkinson’s disease
using a virtual rehabilitation system. Methods Inf Med
56(2):138–144
52. Pulijala Y, Ma M, Pears M, Peebles D, Ayoub A (2018) Effec-
tiveness of immersive virtual reality in surgical training—a ran-
domized control trial. J Oral Maxillofac Surg 76(5):1065–1072
53. Bekels K, Calnan D et al (2017) Effect of an immersive preop-
erative virtual reality experience on patient reported outcomes: a
randomized controlled trial. Ann Surg 265(6):1068–1073
54. Mesa-Gresa P, Lozano JA, Llo
´rens R, Alcan
˜iz M, Navarro MD,
Noe
´E (2011) Clinical validation of a virtual environment test for
safe street crossing in the assessment of acquired brain injury
patients with and without neglect. Lect. Notes Comput. Sci.
Springer, Berlin, pp 44–51
55. Ip HHS, Wong SWL, Chan DFY et al (2018) Enhance emotional
and social adaptation skills for children with autism spectrum
disorder: a virtual reality enabled approach. Comput Educ
117:1–15
56. Thomas M, Padmavathi S, Tamilselvi D (2018) Rehabilitation
Therapy for autism spectrum disorder using virtual reality. Smart
secure systems—IoT and analytics perspective, 1st edn. Springer,
Singapore
57. Bekele E, Wade J, et al (2016) Multimodal adaptive social
interaction in virtual environment (MASI-VR) for children with
Autism spectrum disorders (ASD). In: 2016 IEEE virtual reality,
pp. 121–130
58. Didehbani N, Allen T, Kandalaft M, Krawczyk D, Chapman S
(2016) Virtual reality social cognition training for children with
high functioning autism. Comput Hum Behav 62:703–711
59. Ip HH, Wong SW, et al (2016) Virtual reality enabled training for
social adaptation in inclusive education settings for school-aged
children with autism spectrum disorder (ASD). In: Proceedings of
the international conference on blending learning, pp 94–102
60. Ke F, Lee S (2015) Virtual reality based collaborative design by
children with high-functioning autism: design-based flexibility,
identity, and norm construction. Interact Learn Environ
24:1511–1533
61. Cheng Y, Huang CL, Yang CS (2015) Using a 3D immersive
virtual environment system to enhance social understanding and
social skills for children with autism spectrum disorders. Focus
Autism Other Dev Disabl 30(4):222–236
62. Parsons S (2015) Learning to work together: designing a multi-
user virtual reality game for social collaboration and perspective-
taking for children with autism. Int J Child-Comput Interact
6:28–38
63. Bekele E, Crittendon J et al (2014) Assessing the utility of a
virtual environment for enhancing facial affect recognition in
adolescents with autism. J Autism Dev Disord 44(7):1641–1650
64. Stichter JP, Laffey J et al (2014) iSocial: delivering the social
competence intervention for adolescents (sci-a) in a 3d virtual
learning environment for youth with high functioning autism.
J Autism Dev Disord 44(2):417–430
65. Saadatzi MN, Pennington RC et al (2018) Small-Group tech-
nology-assisted instruction: virtual teacher and robot peer for
individuals with autism spectrum disorder. J Autism Dev Disord
48(11):3816–3830
66. Rogers S (2019) How virtual reality can help those with autism.
https://www.forbes.com/sites/solrogers/2019/04/03/how-virtual-
reality-can-help-those-with-autism/#409bf5fc198e. Accessed 1
Oct 2019
67. Alcorn A, Pain H, Rajendran G, Smith T et al (2011) Social
communication between virtual characters and children with
autism. Artificial intelligence in education. Lecture Notes in
Computer Science 6738. Springer, Berlin, pp 7–14
68. Milne M, Luerssen MH, et al (2010) Development of a virtual
agent based social tutor for children with autism spectrum dis-
orders. In: Proceedings of the 2010 international joint conference
on neural networks (IJCNN), pp 1–9
69. Chen CH, Lee IJ, Lin LY (2016) Augmented reality-based video-
modeling storybook of nonverbal facial cues for children with
autism spectrum disorder to improve their perceptions and
judgments of facial expressions and emotions. Comput Hum
Behav 55:477–485
70. Lorenzo G, Lledo
´A, Pomares J, Roig R (2016) Design and
application of an immersive virtual reality system to enhance
emotional skills for children with autism spectrum disorders.
Comput Educ 98:192–205
71. Lamash L, Klinger E, Josman N (2017) Using a virtual super-
market to promote independent functioning among adolescents
with Autism Spectrum Disorder. In: 2017 international confer-
ence on virtual rehabilitation, pp 1–7
72. Wade J, Zhang L, Bian D, Fan J et al (2016) A Gaze-Contingent
adaptive virtual reality driving environment for intervention in
individuals with autism spectrum disorders. ACM Trans Interact
Intell Syst 6:1–23
73. Taryadi BR, Kurniawan I (2017) The improvement of autism
spectrum disorders on children communication ability with PECS
method Multimedia Augmented Reality-Based. J Phys Conf Ser
947:1–7
74. Finkelstein S, Barnes T, Wartell Z, Suma AA (2013) Evaluation
of the exertion and motivation factors of a virtual reality exercise
game for children with autism. IEEE Virtual Real. https://doi.org/
10.1109/VAAT.2013.6786186
75. Maskey M, Lowry J, Rodgers J, McConachie H, Parr JR (2014)
Reducing specific phobia/fear in young people with autism
spectrum disorders (ASDs) through a virtual reality environment
intervention. PLoS ONE 9(7):e100374. https://doi.org/10.1371/
journal.pone.0100374
76. Adjorlu A, Høeg E, et al (2017) Daily living skills training in
virtual reality to help children with autism spectrum disorder in a
real shopping scenario. In:16th IEEE international symposium on
mixed and augmented reality, pp 294–302
77. Forbes PAG, Pan X, Hamilton AFC (2016) Reduced mimicry to
virtual reality avatars in autism spectrum disorder. J Autism Dev
Disord 46(12):3788–3797
78. Halabi O, El-Seoud S et al (2017) Design of immersive virtual
reality system to improve communication skills in individuals
with autism. Int J Emerg Technol Learn (IJET) 12:50–64
79. Ramachandiran CR, Jomhari N, Thiyagaraja S, Maria M (2015)
Virtual reality based behavioural learning for autistic children.
Electron J e-Learn 13(5):357–365
80. Rosenfield NS, Lamkin K, Re J, Day K, Boyd L, Linstead E
(2019) A virtual reality system for practicing conversation skills
for children with autism. Multimodal Technol Interact 3(2):28
Int. j. inf. tecnol.
123
81. Loomes R, Hull L, Mandy WPL (2017) What Is the male-to-
female ratio in autism spectrum disorder? A systematic review
and meta-analysis. J Am Acad Child Adolesc Psychiatry
56(6):466–474
82. Syahputra MF, Arisandi D et al (2017) Augmented reality social
story for autism spectrum disorder. J Phys Conf Ser. https://doi.
org/10.1088/1742-6596/978/1/012040
83. Zhao H, Swanson AR et al (2018) Hand-in-hand: a communi-
cation-enhancement collaborative virtual reality system for pro-
moting social interaction in children with autism spectrum
disorders. IEEE Trans Hum Mach Syst 48(2):136–148
84. Efstratiou R et al (2021) Teaching daily life skills in autism
spectrum disorder (ASD) interventions using the social robot
pepper. Robotics in education. Advances in intelligent systems
and computing, vol 1316. Springer, Cham
85. Brivio A, Rogacheva K, Lucchelli M, Bonarini A (2021) A soft,
mobile, autonomous robot to develop skills through play in
autistic children. Paladyn, J Behav Robot 12(1):187–198
86. Niderla K, Maciejewsk M (2021) Construction of a zoomorphical
robot for rehabilitation of autistic children. J Phys Conf Ser
1782:012024. https://doi.org/10.1088/1742-6596/1782/1/012024
Int. j. inf. tecnol.
123
