A review of the applicability of robots in education

Abstract

Robots are becoming an integral component of our society and have great potential in being utilized as an educational technology. To promote a deeper understanding of the area, we present a review of the field of robots in education. Several prior ventures in the area are discussed (post-2000) with the help of classification criteria. The dissecting criteria include domain of the learning activity, location of the activity, the role of the robot, types of robots and types of robotic behaviour. Our overview shows that robots are primarily used to provide language, science or technology education and that a robot can take on the role of a tutor, tool or peer in the learning activity. We also present open questions and challenges in the field that emerged from the overview. The results from our overview are of interest to not only researchers in the field of human–robot interaction but also administration in educational institutes who wish to understand the wider implications of adopting robots in education.

Full text

Technology for Education and Learning, 2013
A REVIEW OF THE APPLICABILITY
OF ROBOTS IN EDUCATION
Omar Mubin,
∗
Catherine J. Stevens,
∗∗
Suleman Shahid,
∗∗∗
Abdullah Al Mahmud,
∗∗∗∗
and Jian-Jie Dong
∗∗∗∗∗
Abstract
Robots are becoming an integral component of our society and have
great potential in being utilized as an educational technology. To
promote a deeper understanding of the area, we present a review of
the field of robots in education. Several prior ventures in the area
are discussed (post-2000) with the help of classification criteria. The
dissecting criteria include domain of the learning activity, location
of the activity, the role of the robot, types of robots and types of
robotic behaviour. Our overview shows that robots are primarily
used to provide language, science or technology education and that
a robot can take on the role of a tutor, tool or peer in the learning
activity. We also present open questions and challenges in the field
that emerged from the overview. The results from our overview
are of interest to not only researchers in the field of human–robot
interaction but also administration in educational institutes who
wish to understand the wider implications of adopting robots in
education.
Key Words
Social robotics, pedagogy, human–robot interaction, educational
robotics, educational robots
1. Introduction
Robots are slowly being incorporated in our society and the
number of service robots has in 2008 already outnumbered
industrial robots [1]. Robots are slowly beginning a process
of seamless integration in everyday lives both at home
and at school. This impact of social robotics is even
more crucial for children and teenagers, where robots can
be used for their development and intellectual growth.
As a consequence, greater attention must be levied onto
∗
School of Computing, Engineering and Mathematics and
MARCS Institute, University of Western Sydney, Australia;
e-mail: o.mubin@uws.edu.au
∗∗
MARCS Institute and School of Social Sciences and Psy-
chology, University of Western Sydney, Australia; e-mail:
kj.stevens@uws.edu.au
∗∗∗
Tilburg University, The Netherlands; e-mail: s.shahid@uvt.nl
∗∗∗∗
Delft University of Technology, The Netherlands; e-mail:
a.almahmud@tudelft.nl
∗∗∗∗∗
National Chia-Yi Girl’s Senior High School, Taiwan; e-mail:
djj6.tw@gmail.com
Recommended by Dr. R. Kuo
(DOI: 10.2316/Journal.209.2013.1.209-0015)
how educational robots can be better integrated into the
lives of young people. With the continuous advent of
technology, it is worthwhile to understand the potential
of robots as effective add-ons to learning. Robots can
be an entertaining platform to learn about computers,
electronics, mechanical engineering and languages. It has
been shown [2] that young children performed better on
post-learning examinations and generated more interest
when language learning took place with the help of a robot
as compared to audiotapes and books. Educational robots
are a subset of educational technology, where they are used
to facilitate learning and improve educational performance
of students. Robots provide an embodiment and the ability
to add social interaction to the learning context and hence
an advancement on purely software-based learning. Not
all educational robots require social interaction and this is
discussed further in our review.
In this article, we present an analytical overview of
the prevailing field of robots in education. The aim of
this overview is fourfold: firstly, to provide an overview
because a comprehensive overview of the field of robots
in education does not exist in robotics literature. The
only exceptions are [3] and [4]. The former was con-
ducted in 1996 when the field of robotics was in its in-
fancy. The latter, albeit recent, only considers 10 stud-
ies (hence is not meta level); looks at high school chil-
dren only; does not mention the use of robots to impart
non-technical education and only LEGO Mindstorms are
discussed. Comprehensive reviews exist for robotics (in
particular social robotics [5]), where educational robotics
has been touched upon but not in great detail. Secondly,
our overview will enable researchers in educational robotics
to evaluate their research in an analytical manner and to
place it in appropriate dimensions, for example, choos-
ing an appropriate research question, choosing robotic
behaviour for their learning activity or choosing the appro-
priate robotic kit. Thirdly, research in the area covers a
vast space across cultures, student age groups, robot types
and subject domain. Therefore, a meta-level approach
would help in understanding the field better. Lastly, this
overview will attempt to identify research opportunities for
new research in educational robotics.
A systematic approach was followed to compile rel-
evant articles. Research that was conducted within the
last 10 years was considered (to ensure that we would
1

only overview relatively recent research on the topic). In
addition, an attempt was made to contemplate research
across a variety of host countries such as Japan [6], [7],
Korea [8], [9], Australia [10], Germany [10], USA [11], and
Holland [12], [13]. This also gave us a mix of robots.
Moreover, articles were short listed from not only Robotics
conferences such as HRI [14], IROS
1
, RO-MAN
2
but also
educational technology sources such as the conference on
educational robots
3
and on educational technology. The
paper is structured as follows: initially we describe the
results from a bottom up review of the literature. The
analysis of prior literature resulted in emerging themes of
the use of robots in education and then we consolidated our
findings under those themes. We then discuss future av-
enues of research and, in conclusion, we present a “looking
ahead” angle on the field.
2. The Major Dimensions that Classify Research on
Robots in Education
Review of the literature yielded classification criteria that
could act as thematic placeholders of current and future
research in the area of robots in education. The criteria
refer to questions of the type: what is studied, when is it
studied, how is it studied. We now list the dimensions and
give examples from the literature.
2.1 What Is the Domain or Subject of the Learning
Activity?
The very first criterion that we present is the subject of
learning. The two main yet quite broad categories are
robotics and computer education (a general instilment of
the awareness of technology that could be referred as tech-
nical education) and non-technical education (science and
language). Technical education is the notion of giving
students the knowledge of robots and technology. In most
cases this is done with an aim to introduce computer sci-
ence and programming and to familiarize undergraduate
students with technology [11] and in [12] where Dutch high
school students were gradually exposed to technical sub-
jects using robots. A lesson plan usually involves first an
initial introduction to programming the robot (introduc-
tion phase) and then the students apply their knowledge
practically by making their robots work (intensive phase)
[15]. The introduction phase usually helps when the stu-
dents or even the educational setup is unfamiliar with the
use of robots in education. As the students also build
the robot such activities are usually quite hands on. The
activity of building one’s own robot has been shown to
provide a strong sense of ownership and enhanced interest
in students [12] as students can take their robots home,
tinker with them during free time,
etc.
The second observed domain in the area of robots in
education are non-technical subjects (such as the sciences),
where we witness the employment of robots as an interme-
diate tool to impart some form of education to students,
1
http://www.iros2013.org/
2
http://www.kros.org/ro-man2013/
3
http://www.rie2013.eu/
such as mathematics [10] and geometry [16]. In such scenar-
ios, the movement of the robot is typically the main princi-
ple upon which the learning is based. For example, in [10]
(project executed in Macquarie University Australia) chil-
dren discuss the concept of rotations and transformations
based on the movement of the robot. In [16], the path tra-
jectory of the iRobot is used to interpret angles and geome-
try. Other examples of non-technical applications of robots
education are areas such as kinematics [17] and music
orchestration using the Tiro robot [8] for Korean children.
The third common domain in current literature is the
use of robots to teach a second language. For example En-
glish was taught to Japanese children by the Robovie robot
in [6] by researchers from the robotics laboratory ATR,
Kyoto and in [9] English was taught to Korean children
using the Tiro robot. The implications of using robots
to teach a second language have been well documented
[18] by computer science researchers in Taiwan, where it
is stated that children are not as hesitant to speak to
robots in a foreign language as they are when talking to
a human instructor. In addition, robots can easily behave
in a repetitive manner while students are talking to them
allowing the students to practice without the problem of a
human instructor getting tired. Moreover, in [18] the em-
bodiment of a robot and its social capabilities is discussed
as an important aspect of teaching language. The analysis
in [18] is well presented but lacks empirical evidence. For
example, to verify their claims a study would need to be
run to compare language instruction by a human, a robot
and a computer. Another critical issue is that language
instruction requires accurate speech recognition and that
is one of the hurdles in acknowledging the use of robots
for language instruction [12]. This is precisely why some
researchers use wizard-of-oz techniques (a human wizard
controls the robot behind the scene) to run their experi-
ments [7]. For some of the aforementioned studies of using
robots to teach language, one finds it hard to reach con-
fident validation. The studies were conducted over a few
weeks and therefore a large component of the language was
not learnt.
A fourth domain is the field of assistive robotics, where
robots are used for the cognitive development of children
or teenagers. We will not elaborate further on this, as
it is outside the scope of this paper and extensive survey
articles already exist in the area [19].
2.2 Where Does the Learning Take Place?
The second dimension is the location of the learning activ-
ity. The use of robots in education is either intra-curricular
or extra-curricular. Intra-curricular activities are those
that are part of the school curriculum and a formal part
of the syllabus. One could even include some robot com-
petitions as part of formal learning, as they take place
towards the end of the learning activity and are a form of
assessment-based learning [20]. Extra-curricular learning
takes place after school hours at the school itself as work-
shops under the guidance of instructors, at home under the
guidance of parents or at other designated locations, such
as public places and events. Extra-curricular activities are
2

Table 1
Example Case Studies Across Different Roles of an Educational Robot
Tutor Peer Tool
Language The robot helps students When a student pronounces a word A student learns certain phrases
in remembering correctly, the robot says well in a non-native language by
vocabulary [13] done [9] playing a game with a robot [12]
Science The robot adapts the The robot and the student Sensors and actuators in the robot
arithmetic exercises based collaboratively solve exercises enable the students to learn
on the performance of the in a science class [30] about physics [31]
student [29]
Technology The robot discusses the The robot plays a happy animation The students use LEGO
difficulty of the sound when the students Mindstorms NXT to learn about
programming task with successfully program the programming [32]
the students robot [10]
generally more relaxed, allow for deviations and therefore
easier to setup and organize. There are several exam-
ples of educational robotics curriculum in formal settings
[11]. One of the most well-documented example of infor-
mal robotics education are the Thymio robot workshops
conducted in EPFL, Switzerland [21]. One of the main
advantages of running informal sessions with educational
robots over formal curriculum advancements is that they
are short term, require minimum curriculum design and
because the robot experts can be there “onsite” minimum
training is required for the teaching staff. However, infor-
mal sessions are usually one-off and hence one can question
their longitudinal impact.
2.3 What Role and Behaviour Does the Robot Have
During Learning?
The robot can take on a number of different roles in
the learning process, with varying levels of involvement
of the robot in the learning task. The choice depends
on the content, the instructor, type of student and the
nature of the learning activity. Firstly, on the one hand
the robot can take a passive role and be used as a learning
tool/teaching aid. This would especially apply to robotics
education, where students would be building, creating and
programming robots. On the other hand, the robot can
take the role of co-learner, peer or companion and have
active spontaneous participation [7] (where the focus was
on cooperative learning with the Asimo robot) or even
care receiver [22] (where children learnt English along the
way as they taught the Nao robot). The role of a robot
as a mentor has also been discussed in [23]. However,
it is apparent that before the robot can take on the role
of an autonomous mentor, technological advancements are
necessary in the perceptive abilities of social robots.
In summary, we can define three main categories of the
role of a robot during the learning activity: tool, peer or tu-
tor (see Table 1 for examples). A similar dissection (albeit
named differently as learning about, with and from robots)
has also been discussed in [24]. Upon analysis of prior liter-
ature, it is evident that a clear mapping needs to be drawn
out linking the learning activity to the interaction style of
the robot. For example, for basic learning tasks a coopera-
tive robot was preferred compared to an instructional robot
[7] but for language learning a tutoring style was preferred
[13]. This decision is also governed by the perception of the
students. It has been shown that younger children were
content with robots behaving as peers in the learning pro-
cess while older children thought of robots more as teaching
tools/aids [24]. The degree of social behaviour of the robot
is more or less linked to what role the robot plays during
the learning activity, to the subject domain and to the age
of the students. In [7], it was ascertained that the children
preferred a human-like behaviour and voice for the Asimo
robot. Other attributes such as maintaining eye contact
have also been discussed [25] to engage students. For lan-
guage learning and cognitive development, social interac-
tion is imperative as suggested by [26] (it may not be essen-
tial for technical education), where a survey of two robots
was conducted regarding 4 weeks of usage at home and
school. The two robots were an emotionless humanoid and
an animated robotic dog. The conclusion was that both
children and their parents preferred the robotic dog. Simi-
lar results have been obtained in [27], where a social agent
was found to generate much more interest as compared to
a less social agent and in [13], where a more social robot led
to higher post-test scores while teaching a foreign language
to primary school children in a study conducted by HRI
researchers in Holland with the iCat robot. The constraint
of the way a robot “looks” physically is more flexible; for
example, a humanoid robot could potentially be used to
teach any subject. Nevertheless, prior research has looked
into physical attributes of a robotic teacher. In [28], a quan-
titative analysis of the preferred dimensions of the physical
features of educational humanoid robots is presented.
2.4 What Types of Robots Are Used in Education?
The embodiment of the robot is also a critical factor in the
learning activity. There exist numerous robotic kits, rang-
ing from low-cost single function kits to LEGO Mindstorms
to humanoid robots costing thousands of dollars. To ex-
plore the various options, we may consider a hypothetical
progressive scale of embodiment. On one end of the scale
3

Table 2
Choice of Robots Across Subject Domains and Across Background Knowledge Required in Computing (Darker the Colour
the more Computing Knowledge Is Required to Use/Interact with the Robot in that Cell)
Subject Type Electronic Robotic Kit Mechanical Robotic Kit Humanoid Robot
Language LEGO Mindstorms robots Robovie the
teach ROILA by playing humanoid robot
games with children [12] teaches English [6]
Science Students learn the Students use the Students learn physics
principles of electronics accelerometer of the by understanding the
by creating robots using Thymio robot to processes involved in
the Boebot multi understand gravity [21] humanoid kicking a
function kit [35] ball [43]
Technology Students learn Students enjoy a tangible University students
programming while method to learn learn about computer
creating robots using programming using the vision while
Arduino [44] Mindstorm robots [37] implementing the
Nao robot [45]
there could be the low-cost single function mechanical kits
that are typically used to illustrate only one function, such
as following a line or reacting to the source of sound, e.g.,
OWI-9910 Weasel [33]. Further down the scale, we have
kits which provide the option of educating about not only
robotics but also electronics, e.g., Arduino [34], Parallex
BoeBot [35]. Such kits are fully programmable and stu-
dents can also build robots and upload scripts onto them.
If the kits allow more mechanical freedom and flexibility
with the robot design we step into the category of robots
such as used LEGO Mindstorms [36]. Mindstorm robots
have been shown to teach a wide array of subjects ranging
from language [12], computer science/programming [37],
physics [31], engineering design [38] and robotics [32].
Further, we move to fully embodied robots/agents used
in both formal and informal education such as the Nao
humanoid robot [22], robots embodied as pet animals or
toy characters (Pleo the dinosaur [39]). These robots have
the ability to engage in social interaction, by virtue of
being able to talk and exhibit facial expressions. In most
situations, such robots are used to teach non-technical
subjects such as language, music, which require the robot
to engage in some form of social interaction with the
student. Not all robotic kits will appeal to all kinds of
students. For example, we cannot expect young children to
build complex robots or even use them. On the contrary,
to attract young children, the robot must have animated
features [40], one example being the BeeBot robot [40].
The BeeBot is a colourful bug like robot that can move
but does not have the ability to display expressions or
verbally express itself. The BeeBot is neither physically
manipulative as for example the Mindstorms robot. It is
therefore suitable to teach subjects such as mathematics
[10] and programming [41] to young children.
In general, educational robots should be designed to
take into account the age and the requirements of the stu-
dents or they must be adaptable in real time. For example,
as shown in [42], robotic technology was developed that
enabled a robot “Asobo” to adapt its behaviour based on
the prevalent mood of young children. However, real time
and subjective evaluations are yet to be conducted for such
state of the art technology. Ultimately, the choice of which
robot to utilize in the learning activity depends on various
factors: cost, subject domain, age of the students. We
provide certain examples on the type of robots employed
for each subject domain in Table 2. The background colour
of each cell indicates the computing skills required by the
students to use the robot.
2.5 Which Pedagogical Theories Underpin Re-
search on Robots in Education?
In this section, we discuss a few pedagogical theories
that are the most prevalent within the domain of educa-
tional robotics. The work of Papert [46] on the program-
ming language Logo to introduce turtle geometry has been
well grounded in robotics literature under the
theory of
constructionism. Initially within the field of robots in
education there was a gradual shift from the theory of
constructivism as suggested by Piaget [47] to the modern
educational method of Papert. This shift has been well
explained in [48], including why the paradigm of Papert
best fits the field. The theory of constructivism states that
learnt knowledge is shaped by what the learners know and
experience. Papert adds to this by introducing the notion
of constructionism, which states that learning occurs when
a student constructs a physical artefact and reflects on
his/her problem solving experience based on the motiva-
tion to build the artefact. Research in robots in educa-
tion lends itself well to the constructionism theory and is
by far the most adopted in robotics curricula [49], [50].
Most robotics curricula are hands-on, encourage students
to think and be creative and are based on problem solving.
Robots also act as a bridge in enabling students to
understand humans. For example, students can learn
how speech is processed by humans by considering how
robots recognize speech [12]. This fits with the aspect
of constructionism where learning is a function of what
4

students know in the real world and what they infer in the
virtual world. The connection to biology via the linking of
human sensors to robotic sensors has also been discussed
in [50]. Analogous to the theory of constructionism lie the
principles of active learning [51] and learning by design
[52] that advocate a hands-on approach to increase the
motivation of students. Such paradigms are well suited
to the field because by their very nature “most” robots
are tangible and require to be physically manipulated as
part of the learning activity. Interacting with tools and
artefacts also accords with the concept of the extended
mind [53]. Lastly, we would like to mention the notion of
social constructivism as proposed by Vygotsky [54] which
generally applies to most peer or tutor-based methodologies
of robotics education. The theory of Vygotsky gave rise
to the principle of scaffolding, i.e., breaking up of complex
tasks into smaller tasks, a common occurrence in robotics
education [55].
3. Discussion: Challenges and Open Questions in
the Area of Robots in Education
One of the primary goals of our overview has been to
identify relatively unaddressed avenues and challenges in
the area of robots in education, which we now enlist briefly.
3.1 Exploring the Impact of Robots in Collabora-
tive Learning
It has been established in pedagogy that collaborative
learning is more beneficial than individual learning [56].
It would be interesting to see if the trend replicates while
evaluating and comparing the learning processes of a child
learning alone against learning with a robot and against
learning with another child. If the results prove that
collaborative learning with a robot is as efficient or not
significantly worse than learning with a human peer then
this will be the first step in wider acknowledgement of
the integration of robots in education. Similar research
has been conducted in measuring the game experience
of children while playing alone, with a friend or with a
robot [57].
3.2 Understanding the Role of Teachers in Robots
for Education
One of the major shortcomings in the area is the absence
of well-defined curriculum and learning material for teach-
ers. Robotics education is still seen as an extra-curricular
activity and a part of informal education. As we discussed
earlier, informal education does not require well-defined
curricula
per se. Efforts must be devoted not only to the
development of robotic hardware and software for educa-
tion but also to the design of learning material and appro-
priate curriculum and to the role of the teacher. In theory,
the role of a teacher is directly linked to the role the robot
plays in the learning activity. If the robot acts as the main
focal entity in the learning activity (i.e., as a teaching tool,
e.g., in the case of teaching about robots), the teacher
takes on the role of a facilitator [58]. If the robot takes on a
passive role then the onus is on the teacher to transfer base
knowledge (e.g., using the robot to teach language). In
such situations, training of teaching staff on robotics and
how to conduct robotics curricula is imperative. Looking
ahead it is clear that work needs to be done before robots
can be fully integrated into our schools and support must
be gained from the teachers. In a survey [59], teachers were
more critical of robots in schools than parents and students
were. Teachers need to be reassured that the intention is
not to replace them with robots but rather provide them
with a teaching tool/aid that can complement the learning
experience and motivate of the students.
3.3 Adapting Robotic Behaviour and Curricula to
the Learner
Another important aspect to consider in research on educa-
tional robots is the character of learners and typically not
much work has been undertaken on it. This would include
various attributes of the learner such as the age, gender,
background knowledge of robotics and computer science,
and social and cultural profile. This is where adaptable
tools such as the LEGO Mindstorms are useful as they
cater for learners from diverse technical backgrounds by
providing various programming options (script based or
more advanced languages such as Java/C++). We can
also see examples on considering the gender of students as
in the Roberta project, where an effort is made to engage
girls in technical subjects [60].
3.4 Designing a Socially Acceptable Educational
Robot
Moreover, improvements are also required in the design
of robots. For students to have a satisfying user expe-
rience with robots, efforts must be devoted to improving
the speech understanding capabilities of robots and re-
producing human-like behaviour (in light of the uncanny
valley [61]).
4. Conclusions
This review paper has presented a summary of impor-
tant and recent works in the area of robots in education.
We believe that not only are robots built on advanced
technology but they also provide a tangible and physical
representation of learning outcomes: a valuable aspect of
employing them in education. An outcome of the review
is to encourage pedagogical experts to further understand
the practical aspects of the utilization of robots in educa-
tion. We have tried not to delve too much into pedagogical
theoretical aspects and have attempted to focus more on
issues related to Student–Robot Interaction, unlike prior
reviews in the field [4]. Moreover, since there is a large
volume of research in the area we may have neglected cer-
tain works, such as reports from educational institutions,
which are not readily and widely available. To conclude,
our message is that we do not intend that robots replace
human teachers but highlight the added value that robots
can bring to the classroom in the form of a stimulating,
engaging and instructive teaching aid.
5

Acknowledgements
The first author acknowledges Pierre Dillenbourg who sup-
ported the development of the paper.
References
[1] IFR, Statistical Department. World Robotics Survey 2008.
[2] J. Han, M. Jo, V. Jones, and J.H. Jo, Comparative study on
the educational use of home robots for children, Journal of
Information Processing Systems
, 4(4), 2008, 159–168.
[3] D. Lees and P. LePage, Robots in education: the current state
of the art,
Journal of Educational Technology Systems, 24(4),
1996, 299–320.
[4] F.B.V. Benitti, Exploring the educational potential of robotics
in schools: a systematic review,
Computers and Education,
58(3), 2012, 978–988.
[5] T. Fong, I. Nourbakhsh, and K. Dautenhahn, A survey of
socially interactive robots, Robotics and Autonomous Systems,
42(3–4), 2003, 143–166.
[6] T. Kanda, T. Hirano, D. Eaton, and H. Ishiguro, Interactive
robots as social partners and peer tutors for children: a field
trial, Human–Computer Interaction, 19(1), 2004, 61–84.
[7] S.Y. Okita, V. Ng-Thow-Hing, and R. Sarvadevabhatla, Learn-
ing together: asimo developing an interactive learning part-
nership with children,
Proc. ROMAN, 2009, 1125–1130.
[8] J.H. Han, D.H. Kim, and J.W. Kim, Physical learning activities
with a teaching assistant robot in elementary school music
class,
Journal of Convergence Information Technology, 5(5),
2009, 1406–1410.
[9] J. Han and D. Kim, R-Learning services for elementary school
students with a teaching assistant robot,
Proc. HRI, 2009,
255–256.
[10] K. Highfield, J. Mulligan, and J. Hedberg, Early mathematics
learning through exploration with programable toys, Proc.
Joint Conference Psychology and Mathematics
, 2008, 17–21.
[11] T. Balch, J. Summet, D. Blank, D. Kumar, et al., Design-
ing personal robots for education: hardware, software, and
curriculum,
IEEE, Pervasive Computing, 7(2), 2008, 5–9.
[12] O. Mubin, C. Bartneck, L. Feijs, H. Hooft van Huysduynen,
et al., Improving speech recognition with the robot interac-
tion language, Disruptive Science and Technology, 1(2), 2012,
79–88.
[13] M. Saerbeck, T. Schut, C. Bartneck, and M.D. Janse, Expres-
sive robots in education: varying the degree of social supportive
behavior of a robotic tutor,
Proc. CHI, 2010, 1613–1622.
[14] C. Bartneck, The end of the beginning: a reflection on the
first five years of the HRI conference, Scientometrics, 86(2),
2011, 487–504.
[15] A. Chiou, Teaching technology using educational robotics,
Queensland University, 2004, 9–14.
[16]
Tertl, Studos. Tertl Studos, http://tertl.com/.
[17] R. Mitnik, M. Nussbaum, and A. Soto, An autonomous
educational mobile robot mediator, Autonomous Robots, 25(4),
2008, 367–382.
[18] C.W. Chang, J.H. Lee, P.Y. Chao, C.Y. Wang, et al., Exploring
the possibility of using humanoid robots as instructional tools
for teaching a second language in primary school,
Educational
Technology and Society
, 13(2), 2010, 13–24.
[19] A. Tapus, M.J. Mataric, and B. Scassellati, Socially assistive
robotics, IEEE Robotics and Automation Magazine, 14(1),
2007, 35.
[20] L.B. Almeida, J. Azevedo, C. Cardeira, P. Costa, et al.,
Mobile robot competitions: fostering advances in research,
development and education in robotics,
Proc. CONTROLO,
2000, 592–597.
[21] F. Riedo, P. Retornaz, L. Bergeron, N. Nyffeler, et al., A two
years informal learning experience using the thymio robot,
Proc. Advances in Autonomous Mini Robots, 2012, 37–48.
[22] F. Tanaka and S. Matsuzoe, Children teach a care-receiving
robot to promote their learning: field experiments in a class-
room for vocabulary learning,
Journal of HRI, 1(1), 2012.
[23] M.A. Goodrich and A. Schultz, Human robot interaction: a
survey,
Foundations and Trends in Human–Computer Inter-
action, 1(3), 2007, 203–275.
[24] N. Shin and S. Kim, Learning about, from, and with robots:
students’ perspectives, Proc. RO-MAN, 2007, 1040–1045.
[25] W.L. Johnson, J.W. Rickel, and J.C. Lester, Animated peda-
gogical agents: face-to-face interaction in interactive learning
environments,
International Journal of Artificial Intelligence
in Education, 11(1), 2000, 47–78.
[26] N. Shin and S. Lee, The effects of appearance and interface
design on user perceptions of educational robots,
Proc. URAI
2008, 2008.
[27] R. Moreno, R.E. Mayer, H.A. Spires, and J.C. Lester, The case
for social agency in computer-based teaching: do students learn
more deeply when they interact with animated pedagogical
agents?
Cognition and Instruction, 19(2), 2001, 177–213.
[28] H.J. Ryu, S.S. Kwak, and M.S. Kim, A study on external
form design factors for robots as elementary school teaching
assistants,
Proc. ROMAN, 2007, 1046–1051.
[29] J.B. Janssen, C.C. van der Wal, M.A. Neerincx, and R. Looije,
Motivating children to learn arithmetic with an adaptive robot
game,
Proc. Social Robotics, 2011, 153–162.
[30] T. Hashimoto, H. Kobayashi, A. Polishuk, and I. Verner,
Elementary science lesson delivered by robot, Proc. HRI, 2013,
133–134.
[31] W. Church, T. Ford, N. Perova, and C. Rogers, Physics with
robotics: using lego mindstorms in high school education,
Proc. Advancement of Artificial Intelligence Spring Symposium,
2010, 47–49.
[32] A.J. Hirst, J. Johnson, M. Petre, B.A. Price, et al., What
is the best programming environment/language for teaching
robotics using lego mindstorms?
Artificial Life and Robotics,
7(3), 2003, 124–131.
[33] G. McComb, Getting kids into robotics, Robotics Re-
sources, Servo magazine, 10, 2008, 73–75, http://www. servo-
magazine.com/ uploads/issue_downloads/950/Getting_Kids_
into_Robotics-SV200810.pdf
[34] R. Balogh, Educational robotic platform based on arduino.
Proc. Conference on Educational Robotics, 2010, 119–122.
[35] H. Mukai and N. McGregor, Robot control instruction for
eighth graders, IEEE, Control Systems, 24(5), 2004, 20–23.
[36] LEGO, Lego Mindstorms Nxt. http://mindstorms.lego.com/
en-us/Default.aspx.
[37] K. Powers, P. Gross, S. Cooper, M. McNally, et al., Tools
for teaching introductory programming: what works?
ACM
SIGCSE Bulletin, 38(1), 2006, 560–561.
[38] J.V. Ringwood, K. Monaghan, and J. Maloco, Teaching engi-
neering design through lego mindstorms, European Journal of
Engineering Education
, 30(1), 2005, 91–104.
[39] M. Heerink, M. Daz Boladeras, J. Albo-Canals, C. Angulo
Bahn, et al., A field study with primary school children on
perception of social presence and interactive behavior with a
pet robot,
Proc. ROMAN, 2012.
[40] P. Janka, Using a programmable toy at preschool age: why
and how? Proc. SIMPAR, 2008, 112–121.
[41] K. Stoeckelmayr, M. Tesar, and A. Hofmann, Kindergarten
children programming robots: a first attempt, Proc. Robotics
in Education
, 2011, 185–192.
[42] P. Ruvolo, I. Fasel, and J. Movellan, Auditory mood detection
for social and educational robots,
Proc. Conference on Robotics
and Automation, 2008, 3551–3556.
[43] S. Carpin, M. Lewis, J. Wang, S. Balakirsky, et al., Usarsim:
a robot simulator for research and education,
Proc. Robotics
and Automation, 2007, 1400–1405.
[44] S. Alers and J. Hu, Admoveo: a robotic platform for teach-
ing creative programming to designers, Learning by Playing.
Game-Based Education System Design and Development
, 2009,
410–421.
[45] D.S. Touretzky, Preparing computer science students for the
robotics revolution, Communications of the ACM, 53(8), 2010,
27–29.
[46] S. Papert, Mindstorms: children, computers, and powerful
ideas
(Basic Books, Inc., 1980).
[47] J. Piaget and B. Inhelder, The psychology of the child, vol.
5001 (Basic Books (AZ), 1969).
6

[48] T.A. Mikropoulos and J. Bellou, Educational robotics as
mindtools, Earth Lab UOI, 2008.
[49] M. Kabatova and J. Pekarova, Lessons learnt with lego mind-
storms: from beginner to teaching robotics,
Proc. Conference
on Educational Robotics, 2010, 51–56.
[50] M.A. Yousuf, Robots in education, Chapter to be published in
the Encyclopedia of Artificial Intelligence, (Eds.) J.R. Rabual,
J. Dorado and A. Pazos, 2008, 978–981.
[51] M. Harmin and M. Toth, Inspiring active learning: a complete
handbook for today’s teachers (Association for Supervision &
Curriculum Development, 2006).
[52] R. Goldman, A. Eguchi, and E. Sklar, Using educational
robotics to engage inner-city students with technology,
Proc.
Conference on Learning Sciences, 2004, 214–221.
[53] A. Clark and D. Chalmers, The extended mind, Analysis, 58,
1998, 7–19.
[54] L. Vygotsky, Mind in society (Harvard University Press, 1978).
[55] J.T. Doswell and P.H. Mosley, An innovative approach to
teaching robotics, Proc. Advanced Learning Technologies, 2006,
1121–1122.
[56] P. Dillenbourg, What do you mean by collaborative learn-
ing? in
Collaborative Learning: Cognitive and Computational
Approaches (Elsevier, 1999), 1–19.
[57] S. Shahid, E. Krahmer, and M. Swerts, Child–robot interaction:
playing alone or together? Proc. CHI, 2011, 1399–1404.
[58] D. Alimisis, Robotics in education & education in robotics:
shifting focus from technology to pedagogy.
[59] E. Lee, Y. Lee, B. Kye, and B. Ko, Elementary and mid-
dle school teachers, students and parents: perception of
robot-aided education in Korea,
Proc. Conference on Educa-
tional Multimedia, Hypermedia and Telecommunications, 2008,
175–183.
[60] A. Bredenfeld and T. Leimbach, The roberta initiative, Proc.
SIMPAR 2010
, 2010, 558–567.
[61] M. Mori, The Uncanny Valley, Energy, 7(4), 1970, 33–35.
Biographies
Omar Mubin is currently a lec-
turer in human–computer interac-
tion at the School of Computing,
Engineering and Mathematics at
the University of Western Sydney,
Australia. He obtained his Ph.D.
qualification in human–robot in-
teraction from the Eindhoven
University of Technology, The
Netherlands. His research inter-
ests include educational robots,
user centred design and empirical
research in the human perception of social robotics.
Catherine J. Stevens investi-
gates the psychological processes
in creating, perceiving and per-
forming music and dance, and
applies experimental methods
to evaluate complex systems
and human–computer interac-
tion. She holds B.A. (Hons) and
Ph.D. degrees from the University
of Sydney. She is professor in
Psychology and leads the Music
Cognition and Action research
program in the MARCS Institute at the University of
Western Sydney (http://marcs.uws.edu.au/).
Suleman Shahid is currently
an assistant professor at the
Tilburg centre for Cognition and
Communication, Department of
Communication and Information
Sciences, Tilburg University. He
obtained his Ph.D. in communica-
tion sciences in 2011. Before join-
ing Tilburg University in 2007,
he finished his Professional Doc-
torate in Engineering (P.D.Eng.)
degree from the Eindhoven Uni-
versity of Technology and during his stay in Eindhoven he
also spent almost a year at the Philips Research, Eind-
hoven. He has a background in Media Computing, but in
the last few years he has been involved in interaction design
and social aspects of affective computing, particularly in a
cross-cultural setting.
Abdullah Al Mahmud is a post-
doctoral research fellow at the
faculty of Industrial Design En-
gineering, Delft University of
Technology, The Netherlands.
He obtained his Ph.D. degree
in human–computer interaction
design from the Department of
Industrial Design, Eindhoven Uni-
versity of Technology (TU/e),
The Netherlands. He also earned
a Professional Doctorate in Engi-
neering (P.D.Eng.) degree from the same university and
worked as a visiting researcher at the Philips Research, The
Netherlands. His research interests are method adaptation
(i.e., design and evaluation) for different purposes/user
groups, socially responsible interaction design, etc. More-
over, he is very much interested in all sorts of field trials
including reliable data gathering techniques.
Jian-Jie Dong is a part time lec-
turer of National Formosa Univer-
sity and a technology teacher of
National Feng-Yuan Senior High
School in Taiwan. His research
interest is about educational tech-
nology application.
7