Improving a Mobile Learning Companion for Self-regulated Learning

using Sensors

Haeseon Yun

, Albrecht Fortenbacher

and Niels Pinkwart

University of Applied Science in Berlin, Wilhelminenhofstr 75a, 12459 Berlin, Germany

Humboldt University of Berlin, Rudower Chaussee 25, 12489 Berlin, Germany

Keywords:

Intelligent Tutoring System, Learning Companion, Self-regulated Learning, Mobile Learning, Sensor-based

Learning, Pervasive Learning.

Abstract:

The ability to efﬁciently manage learning is linked to positive learning experience and outcome. However, at-

taining the ability of self-regulating is not a matter of course for students and it requires an external assistance.

A learning companion stemming from intelligent tutoring system (ITS) has non-authoritative, co-present ef-

fect as a learning support and with available sensor technology, self-regulated learning can be better promoted.

Sensor enhanced learning companion can detect learning states, learning behaviours and context and provide

valuable feedback to learners to increase their awareness of learning progress and to effectively manage their

learning. Considering the mobility, autonomy of learners along together with current trend in open online

learning resources and contents, available embedded sensors are suitable for realising the concept of learning

companion for self-regulated learning. The paper reviews a self-regulated learning concept, a learning com-

panion pedagogy and proposes sensor technology for self-regulated learning and learning companion design

considerations.

1 INTRODUCTION

Successful learning does no longer depend on lack

of learning resources or contents nor types of tech-

nology, whereas it remains on availability of learning

support and learner’s ability to motivate and manage

learning. Learner’s determination and regulation on

what to know and how to learn becomes an especially

important skill for learners with constant distractions,

especially when students tackle their learning con-

tents in various places and time of their choice. Hase

and colleagues advocated the need of learner centric

learning environment approach (Hase and Kenyon,

2001) and various recent research (Fernandez-Rio

et al., 2017; Lee and Recker, 2017; Kizilcec et al.,

2017; Guy et al., 2017; Mega et al., 2014) discussed

the importance of self-regulated learning for success-

ful learning.

The aim of this position paper is to discuss the

possibility of promoting self-regulated learning us-

ing a learning companion pedagogy and the possibil-

ity of improvement of a learning companion for self-

regulated learning using current available sensor tech-

nology. The paper will ﬁrst present previous studies

on self-regulated learning and learning companion.

Following the literature review, sensors which can

provide information on the state of the self-regulated

learning are discussed to suggest the feasibility of

self-regulated learning support using current sensor

technology. Lastly, based on the literature on learning

companion and newly found limitation on sensors for

self-regulated learning state detection, the paper dis-

cuss the design consideration for a learning compan-

ion to facilitate self-regulated learning support using

sensors.

2 STATE OF THE ART

2.1 Self-regulated Learning

Self-regulated learning is a process where a learner

autonomously converts intellectual competence to

academic proﬁciency through control of thoughts,

feelings and behaviours based on a set goal (Zim-

merman, 2002). When faced with a learning task, a

self-regulated learner will actively follow four phases

explicitly or in a tacit way (Pintrich, 2004) : 1) fore-

thought, planning and activation, 2) monitoring, 3)

Yun H., Fortenbacher A. and Pinkwart N.

Improving a Mobile Learning Companion for Self-regulated Learning using Sensors.

DOI: 10.5220/0006375405310536

In Proceedings of the 9th International Conference on Computer Supported Education (CSEDU 2017), pages 531-536

ISBN: 978-989-758-239-4

control and 4) reaction and reﬂection. All phases

concern learner’s cognition, motivation, behaviour as-

pects along with context for external learning sur-

roundings. Self-regulated learning includes how to

attain knowledge better as well as how to control

learners’ emotion and motivation. Therefore, meta-

cognition which deals with knowing how to cogni-

tively learn is a component of a self-regulated learn-

ing framework as well as meta-motivation (how to

motivate oneself).

Although having self-regulated learning skills is

helpful, challenges remain. Zimmerman (2002) dis-

cussed few instructors engage learners to be equipped

with effective self-regulated learning skills and learn-

ers are not often provided with choices to choose their

own learning task. Trainings for self-regulated learn-

ing ability are available, yet trainings are mainly pro-

vided within a speciﬁc learning content or as a general

skill which are difﬁcult for learners to apply in other

tasks of their choice.

As a mean to facilitate self-regulated learning,

Azevedo and Cromley (Azevedo and Cromley, 2004)

advocated the possible integration of self-regulated

learning process in a learning companion which pro-

vide a feeling of non-authoritative, friendly pedagog-

ical support for learning enhancement.

2.2 Learning Companion

Learning companion is a non-authoritative educa-

tional agent that has human characteristics to facili-

tate social learning. The concept of learning compan-

ion derived from an intelligent tutoring system (ITS)

and the term, pedagogical agent is interchangeably

used in related research. The main role of ITS is pro-

viding an appropriate one-on-one feedback to learn-

ers by taking a speciﬁc role (domain export, student,

pedagogical or interface) (Wenger, 1987). Similar to

other ITS, the study on learning companion explains

the role of learning companion as a peer and available

collaborator that encourages students to learn cooper-

atively (Goodman et al., 2016). Additionally, another

study introduces three roles of learning companion as

competition, suggestion and collaboration. Competi-

tion refers to competing with peer in a form of com-

paring their work with each other upon independent

work (Chan and Baskin, 1988). Suggestion role is to

collaboratively work and observe each other’s work.

Collaboration role refers to collaborative work with

shared responsibility.

The difference between ITS and learning compan-

ion is not easily distinguishable as the common in-

teraction behaviours such as coherent dialogue with

a student and pedagogical decision making regard-

ing on what information and when to provide can

be found in studies in ITS, pedagogical agent and

learning companion (Johnson et al., 2000). 5 types

of learning companions (PHelpS, People Power, Dis-

tributed West, LC and SAM) were reviewed and they

reveal unique interaction characteristics between a

system and a learner:

• PHelpS (Peer Help System) : It offers assistance

to ﬁnd others to provide help in learning and it me-

diates communication without being directly in-

volved in learning activities (Greer et al., 1998).

• People Power : It questions students to seek an-

swers by themselves and interacts with a student

as a co-learner (Dillenbourg and Self, 1992).

• Distributed West : It acts as a non-adaptive coach,

allowing students to see each other in a system

and play as learning companions (Chan et al.,

1992) .

• LC : It initiates a dialogue and starts with a rap-

port building. During learning, it intervenes and a

learners is provided with learning status and learn-

ing strategy use (VanLehn et al., 2016).

• SAM : It has learner’s characteristics such as age

and communicate style and initiates dialogue with

a learner. It takes turns to communicate and

encourages a learner with compliments (Ryokai

et al., 2003).

Learning companion guides learners through

communication (PHelpS, People Power, LC, LuCy

and Sam) and it uses various pedagogical strategies

to motivate and provide opportunities for learners to

reﬂect (LC, LuCy and SAM). The most distinctive

difference between other ITS and learning compan-

ion is the non-authoritative, co-learner characteris-

tics. Unlike ITS, students perceive learning com-

panion as a fellow student or peer who has similar

knowledge level (Ryokai et al., 2003). It also cares

about a learner’s progress and shows social connec-

tion by working together and shows emotions to make

learners interact with them continuously (Woolf et al.,

2009). However, the current research on learning

companion is limited to a single subject (domain),

therefore the design of learning companion which can

integrate numerous domains continuously to stay as a

lifelong friend should be explored (Chou et al., 2003).

The following section explores mobile device plat-

form for learning companion.

2.3 Mobile Learning Companion

Usage of mobile device in daily life has been grown

and the device has been penetrated in various aspects

in life. The statistics shows that more than 70 %

of people in America own smartphone in 2016 and

smartphone dependency over time is gradually grow-

ing (PewResearchCenter, 2017). In Europe, the simi-

lar phenomena can be observed. The number of peo-

ple using internet on the move by using mobile de-

vices has grown to 57 % in 2015 compared to 36 % in

2012 (Eurostate, 2016).

For learning, mobile learning provides a person-

alised, unobtrusive, instant way for learners to access

learning materials and tools and extends the availabil-

ity of education to all people (Traxler and Kukulska-

Julme, 2005). Searching for information is possible

anytime using a mobile device and sharing informa-

tion became much more instant and intuitive. The

mobile device provides a rich learning contents in

inexpensive way to people and it serves a key role

to make information globally accessible (Wellman,

2007). People do not need to sit or even be in one

place to learn but they can learn whenever they want

and wherever they want.

The mobile device also proposes a distinctive op-

portunity to support learners continuously at the ap-

propriate time. Available technology in a mobile de-

vice allows learners to interact with others online,

share their work with others and with other devices in

learners’ household. Information can be found easily

and scheduling with alert function helps learners man-

age their time. The embedded sensors in a mobile de-

vice are also a beneﬁcial means as they provide infor-

mation regarding on learner’s location and state. Cur-

rent learning companion systems are designed mainly

on stable devices (computer), which fails to consider

mobility of learners.

Goodman and colleagues proposed the impor-

tance of designing a learning companion which can

interact with learners in more natural way (Goodman

et al., 2016) and considering learners’ mobility in

learning environment can be fulﬁlled by investigating

a mobile technology. With the progress of technology

advancement, more affordable, reﬁned sensors will be

embedded in a mobile device which will serve a good

tool to realise a learning companion.

3 SENSORS FOR A MOBILE

LEARNING COMPANION

Smartphone that people carry contains at least 15 sen-

sors which detect user’s behaviour and context which

can monitor learner state and context in unobtrusive

ways. This part proposes that sensors can support

self-regulated learning and how a learning companion

should be designed to serve as an effective interactive

co-learner.

3.1 Sensors for Self-regulated Learning

Based on the self-regulated learning framework (Pin-

trich, 2004), sensors that are used in health, learning

and everyday life have been reviewed and identiﬁed as

possible sensors to detect each self-regulated learning

state as shown in the Table 1.

Table 1: Sensors to detect self-regulated learning area and

phase (Cog. : Cognition, Motiv: Motivation, Prox.: Prox-

imity Sensor, HR : Heart Rate Sensor, Accel.: Accelerom-

eter, Mic.: Microphone, EDA: Electrodermal Activity Sen-

sor, Light: Ambient Light Sensor).

Cog. Motiv. Behaviour Context

1 Bluetooth

- - - Prox.

Light

2 Camera Camera Camera Camera

Prox. HR Prox. Mic.

Prox.

3 Mic. Mic. GPS Mic.

EDA Wiﬁ Prox.

Touch Gyro

Accel. Baro

Gyro GPS

Prox. Wiﬁ

Baro Accel.

4 HR Camera

EDA EDA - -

Sensors such as camera can detect learner’s con-

trol phase (phase 2) for all areas(cognition, motiva-

tion, behaviour and context) of self-regulated learn-

ing. For instance, camera can recognise a learn-

ers’ gaze, a presence of a user and facial expression

(Horvitz and Apacible, 2003). Detection of learner’s

attention to learning material can be done through

gaze detection (Sharma et al., 2016). A wearable

camera can recognise learner’s behaviour, speciﬁcally

the task that he or she is involved in (Fathi et al.,

2012). Camera on a smartphone could detect phys-

iological state such as fatigue (He et al., 2013). Emo-

tion detection is possible with a camera by detect-

ing head position and movement (Woolf et al., 2009).

Mobile application, inSightDemo (shown in Figure 1)

uses front end camera to detect facial expression and

shows 7 states (neutral, happiness, surprise, anger,

disgust, fear and sadness).

Figure 1: InSightDemo (AppStore).

Microphone is also widely used to detect ambient

noise, context and activity of users. In one study, mi-

crophone is used to deﬁne learners’ activity by relat-

ing speaking behaviour with being in a conversation

(Siewiorek et al., 2003). It also detects and validates

the use of self-regulated learning strategy (phase 3).

For instance, learners’ positive self talk to boost their

self-efﬁcacy (Bandura, 1997) and minimise negative

affect (Zeidner, 1998) to regulate their cognition and

motivation can be sensed by microphone.

Contextual information such as perceiving learning

environment can be detected through Bluetooth, prox-

imity sensor, camera and ambient light sensor. Prox-

imity and ambient light sensors detect the bright-

ness, artiﬁcial sources and the presence of other peo-

ple or object (Schmidt et al., 1999). Bluetooth used

in Miluzzo’s study detects the people who are using

the same mobile application and provides feedback to

learners to provide contextual information .

Motivation awareness and monitoring (phase 2) can

be detected using a heart rate sensor and galvanic

skin conductance (or electro dermal activity sensor so

called EDA) (Mandryk and Atkins, 2007). Especially

for cognition and motivation monitoring, heart rate

and EDA sensor can be used (Mandryk and Atkins,

2007). Even though current EDA sensor is restricted

in a wearable technology which needs to be resolved

for unobtrusive monitoring, the heart rate sensor has

been adopted in current smartphone using a rear end

camera (Samsung, 2016).

Location of a person and user’s activity inference

through detecting the location of the sensor device

can be realised through WiFi, GPS and touch sensor

(Kern and Schiele, 2003; Miluzzo et al., 2008; Ben-

Zeev et al., 2015; Hinckley et al., 2000). Learner’s

forfeiting behaviours can be detected using Wiﬁ, GPS

and touch sensor as giving up refers to leaving the

place where they set to study and enabling a mobile

device which can distract them from learning. Also

learner’s action to change or leave a context to con-

trol their learning environment can also be detected

using GPS, WiFi, accelerometer and even gyro and

barometer sensors.

Comparison of heart rate and EDA between the

beginning and the end of the learning can also pro-

vide learners with awareness of their learning. Fur-

thermore, camera used for gaze and facial expression

can provide additional information on learner’s atten-

tion and affective reaction during learning. Context

detection and location information that have been al-

tered during learning can be used to evaluate the task

and learner’s behaviour. Overall, reaction and reﬂec-

tion stage (phase 4) of self-regulated learning can be

fostered by providing various combination of sensor

data, mentioned previously. However, various areas,

especially phase 1 of self-regulated learning, are dif-

ﬁcult to correlate with sensors as they refer to percep-

tion of task, perception of context, task value, plan-

ning and goal orientation.

The sensor technology can provide non-invasive

information to support learning, especially for self-

regulated learning. Nonetheless, for a learning com-

panion to be served as a scaffold to facilitate learn-

ing, comprehensive self-regulated learning areas and

phases should be considered for design.

3.2 Design Consideration for a Sensor

based Learning Companion

The limitation from sensors integration for learning

support as discussed in the previous section and the

need for an adaptive feedback can be resolved by

introducing a learning companion pedagogy. For

instance, the learning companion system, Sam, en-

hanced learning experience by using sensors. Sam en-

gaged younger learners in a storytelling task by using

radio frequency tag embedded in a learning resource

and a microphone to take turns (Ryokai et al., 2003).

Sam integrated sensors technology to use a learning

resource cooperatively with a learner and also com-

municate with him or her in a way that a learner per-

ceives Sam as a peer and even a human.

Sensors adopted in a learning companion also de-

tect physiological state such as stress level and a com-

panion provides messages to relieve stress and con-

tinue engage in learning even learners are facing difﬁ-

cult tasks (Prendinger and Ishizuka, 2005). This em-

pathic companion co-exists with a learner and main-

tains a long relationship with a learner by providing

an emotional support.

For a learning companion to be successful in con-

necting with a learner and maintains a relationship,

design consideration should be made. Especially, the

design of learning companion should consider the

missing links to incorporate overall aspects of self-

regulated learning with emphasis on characteristics of

learning companion (non-authoritative, intuitive and

friendly). Based on the previous studies on learning

companion, desirable characteristics of learning com-

panion are as follows:

• A learning companion should correspond to each

learner’s characteristics (Kim, 2007).

• A learning companion should provide instruc-

tional advantages and provide encouragement to

learners (Goodman et al., 2016; Woolf et al.,

2009).

• A learning companion should initiate dialogue

and force students to engage in reﬂection (Good-

man et al., 2016).

• A learning companion should have a simple and

stylish visual with task and relation orientation

(Haake and Gulz, 2009).

• Learners should regard a learning companion as

a fellow learner and furthermore a real human

(Ryokai et al., 2003).

• Learners should enjoy interacting with it and have

positive perception of overall learning experience

(Woolf et al., 2009).

4 CONCLUSIONS

Learners, with numerous choices in educational con-

tents, can beneﬁt by having self-regulated learning

skills as they can support learners to tackle various

obstacles that learners face during learning. Train-

ing for self-regulated learning skills is possible yet

the transfer of this skill in another domain is ques-

tionable and not actively enforced. Mobile device

that has technical capabilities to detect valuable in-

formation related to learning environment, behaviour,

cognition and motivation can facilitate learners to be-

come more capable people and have better learning

experience. However, as Hase and colleagues em-

phasised that learning is only possible when learn-

ers perceive learning as process of one’s improvement

and it is guided without threats (Hase and Kenyon,

2001), learning companion pedagogy can be an ideal

method to meet current technology enhanced learning

environment. Sensor technology can support learners

by providing information on their cognitive and emo-

tional state and guide them to choose an optimal be-

haviour and a suitable ambience for learning.

In this respect, this position paper reviews the con-

cept of self-regulated learning and a learning compan-

ion pedagogy and discusses available mobile sensors

which can be utilised to detect self-regulated learn-

ing aspects. Limitations of sensor utilisation for self-

regulated learning were disclosed and the paper pro-

poses design considerations for a sensor based learn-

ing companion by considering a learning companion

pedagogy. The proposed design consideration for a

learning companion is broad however, the aims of this

paper are to propose that 1) sensors can detect some

areas and phases of self-regulated learning which can

support learners to be aware of their learning progress

and environment and 2) a learning companion peda-

gogy should be integrated when designing an effec-

tive pedagogical support for learning.

REFERENCES

Azevedo, R. and Cromley, J. G. (2004). Does training

on self-regulated learning facilitate students’ learning

with hypermedia? Journal of educational psychology,

96(3):523.

Bandura, A. (1997). Self-efﬁcacy: The exercise of control.

Macmillan.

Ben-Zeev, D., Scherer, E. A., Wang, R., Xie, H., and Camp-

bell, A. T. (2015). Next-generation psychiatric assess-

ment: Using smartphone sensors to monitor behavior

and mental health. Psychiatric rehabilitation journal,

38(3):218.

Chan, T.-W. and Baskin, A. B. (1988). Studying with the

prince: The computer as a learning companion. In

Proceedings of the International Conference on Intel-

ligent Tutoring Systems, volume 194200.

Chan, T.-W., Chung, I.-L., Ho, R.-G., Hou, W.-J., and Lin,

G.-L. (1992). Distributed learning companion system:

West revisited. In International Conference on Intel-

ligent Tutoring Systems, pages 643–650. Springer.

Chou, C.-Y., Chan, T.-W., and Lin, C.-J. (2003). Redeﬁn-

ing the learning companion: the past, present, and fu-

ture of educational agents. Computers & Education,

40(3):255–269.

Dillenbourg, P. and Self, J. A. (1992). People power: A

human-computer collaborative learning system. In

International Conference on Intelligent Tutoring Sys-

tems, pages 651–660. Springer.

Eurostate (2016). Digital economy and society statistics-

households and individuals.

Fathi, A., Li, Y., and Rehg, J. M. (2012). Learning to recog-

nize daily actions using gaze. In European Conference

on Computer Vision, pages 314–327. Springer.

Fernandez-Rio, J., Cecchini, J. A., M

endez-Gimenez, A.,

Mendez-Alonso, D., and Prieto, J. A. (2017). Self-

Regulation, Cooperative Learning, and Academic

Self-Efﬁcacy: Interactions to Prevent School Failure.

Frontiers in Psychology, 8.

Goodman, B., Linton, F., and Gaimari, R. (2016). Encour-

aging student reﬂection and articulation using a learn-

ing companion: A commentary. International Journal

of Artiﬁcial Intelligence in Education, 26(1):474–488.

Greer, J. E., Mccalla, G., Collins, J. A., Kumar, V. S.,

Meagher, P., and Vassileva, J. (1998). Supporting peer

help and collaboration in distributed workplace envi-

ronments. International Journal of Artiﬁcial Intelli-

gence in Education (IJAIED), 9:159–177.

Guy, R., Byrne, B., and Dobos, M. (2017). Stop Think:

a simple approach to encourage the self-assessment

of learning. Advances in Physiology Education,

41(1):130–136.

Haake, M. and Gulz, A. (2009). A look at the roles in em-

bodied pedagogical agents–a user preference perspec-

tive. International Journal of Artiﬁcial Intelligence in

Education, 19(1):39–71.

Hase, S. and Kenyon, C. (2001). From andragogy to heuta-

gogy. Ultibase articles, 5(3):1–10.

He, J., Roberson, S., Fields, B., Peng, J., Cielocha, S.,

and Coltea, J. (2013). Fatigue detection using smart-

phones. Journal of Ergonomics, 3(03):1–7.

Hinckley, K., Pierce, J., Sinclair, M., and Horvitz, E.

(2000). Sensing techniques for mobile interaction.

In Proceedings of the 13th annual ACM symposium

on User interface software and technology, pages 91–

100. ACM.

Horvitz, E. and Apacible, J. (2003). Learning and reasoning

about interruption. In Proceedings of the 5th inter-

national conference on Multimodal interfaces, pages

20–27. ACM.

Johnson, W. L., Rickel, J. W., Lester, J. C., et al. (2000). An-

imated pedagogical agents: Face-to-face interaction in

interactive learning environments. International Jour-

nal of Artiﬁcial intelligence in education, 11(1):47–

78.

Kern, N. and Schiele, B. (2003). Context-aware notiﬁcation

for wearable computing. In Proceedings of the 7th

IEEE International Symposium on Wearable Comput-

ers (ISWC’03), pages 223–230.

Kim, Y. (2007). Desirable characteristics of learning com-

panions. International Journal of Artiﬁcial Intelli-

gence in Education, 17(4):371–388.

Kizilcec, R. F., P

erez-Sanagust

ın, M., and Maldonado,

J. J. (2017). Self-regulated learning strategies predict

learner behavior and goal attainment in Massive Open

Online Courses. Computers & Education, 104:18–33.

Lee, J.-E. and Recker, M. (2017). Measuring Students’ Use

of Self-Regulated Learning Strategies from Learning

Management System Data: An Evidence-Centered

Design Approach About Analytics for Learning

(A4L).

Mandryk, R. L. and Atkins, M. S. (2007). A fuzzy physio-

logical approach for continuously modeling emotion

during interaction with play technologies. Interna-

tional journal of human-computer studies, 65(4):329–

347.

Mega, C., Ronconi, L., and De Beni, R. (2014). What

makes a good student? How emotions, self-regulated

learning, and motivation contribute to academic

achievement. Journal of Educational Psychology,

106(1):121–131.

Miluzzo, E., Lane, N. D., Fodor, K., Peterson, R., Lu, H.,

Musolesi, M., Eisenman, S. B., Zheng, X., and Camp-

bell, A. T. (2008). Sensing meets mobile social net-

works: the design, implementation and evaluation of

the cenceme application. In Proceedings of the 6th

ACM conference on Embedded network sensor sys-

tems, pages 337–350. ACM.

PewResearchCenter (2017). Mobile fact sheet. Accessed:

2017-03-15.

Pintrich, P. R. (2004). A conceptual framework for assess-

ing motivation and self-regulated learning in college

students. Educational psychlogy review, 16(4):385–

407.

Prendinger, H. and Ishizuka, M. (2005). The empathic com-

panion: A character-based interface that addresses

users’affective states. Applied Artiﬁcial Intelligence,

19(3-4):267–285.

Ryokai, K., Vaucelle, C., and Cassell, J. (2003). Virtual

peers as partners in storytelling and literacy learning.

Journal of computer assisted learning, 19(2):195–

208.

Samsung (2016). Galaxy s7 edge and galaxy s7. Accessed:

2017-01-31.

Schmidt, A., Aidoo, K. A., Takaluoma, A., Tuomela, U.,

Van Laerhoven, K., and Van de Velde, W. (1999). Ad-

vanced interaction in context. In International Sympo-

sium on Handheld and Ubiquitous Computing, pages

89–101. Springer.

Sharma, K., Alavi, H. S., Jermann, P., and Dillenbourg, P.

(2016). A gaze-based learning analytics model: in-

video visual feedback to improve learner’s attention in

moocs. In Proceedings of the Sixth International Con-

ference on Learning Analytics & Knowledge, pages

417–421. ACM.

Siewiorek, D. P., Smailagic, A., Furukawa, J., Krause, A.,

Moraveji, N., Reiger, K., Shaffer, J., and Wong, F. L.

(2003). Sensay: A context-aware mobile phone. In

ISWC, volume 3, page 248.

Traxler, J. and Kukulska-Julme, A. (2005). Mobile learning

in developing countries.

VanLehn, K., Zhang, L., Burlson, W., Girard, S., and

Hidago-Pontet, Y. (2016). Can an non-cognitive learn-

ing companion increase the effectiveness of a meta-

cognitive learning strategy? IEEE Transactions on

Learning Technologies.

Wellman, S. (2007). Google lays out its mobile user expe-

rience strategy. Information Week, April, 11.

Wenger, E. C. (1987). Artiﬁcial intelligence and tutoring

systems. Computational and Cognitive Approaches

to the Communication of Knowledge. Morgan Kauff-

mann, Los Altos, San Francisco, CA USA.

Woolf, B., Burleson, W., Arroyo, I., Dragon, T., Cooper,

D., and Picard, R. (2009). Affect-aware tutors: recog-

nising and responding to student affect. International

Journal of Learning Technology, 4(3-4):129–164.

Zeidner, M. (1998). Test anxiety: The state of the art.

Springer Science & Business Media.

Zimmerman, B. J. (2002). Becoming a self-regulated

learner: An overview. Theory into practice, 41(2):64–

70.