Here’s Looking at You, Player

The Potential of Eye Tracking Analysis for Player-centered Learning Game Design

Susanne Friedemann, Katharina Meier and Klaus P. Jantke

Fraunhofer Institute for Digital Media Technology, Children’s Media Dept., Erich-K¨astner-Str. 1a, 99094 Erfurt, Germany

Keywords:

Game-based Learning, Serious Games, Evaluation, Eye Tracking.

Abstract:

There is no doubt that technology enhanced learning, in general, and game-based learning, in particular, needs

thoughtful preparation, a careful design and a reliable implementation to meet the expectations of impact.

But e-learning systems and serious games are digital media which usually are perceived very differently by

different human learners with individually varying background and experience and within varying contexts.

Perception and impact of media are a very delicate issue to be investigated carefully on a ﬁrm scientiﬁc basis.

Digital games designed and implemented for special purposes of learning are particularly difﬁcult to evaluate.

Eye tracking is a technology suitable for the analysis of essential features of media perception which are

relevant to learning. Eye tracking human learners’ interactions with a serious game allows for the detection of

substantial phenomena crucial for the design of game playing experiences likely to foster learning processes.

1 THE AUTHORS’ POSITION

“Here’s looking at you, kid.” is a citation from the

classic movie Casablanca by Michael Curtiz, 1942,

starring Ingrid Bergman and Humphrey Bogart. The

malapropism “Here’s looking at you, User.” is due to

Christian Wolff

in his 2009 lecture on eye tracking

in human-machine interaction. For the present paper,

this idea has been adopted and adapted to direct the

reader’s attention to eye tracking of human game play.

The authors’ particular focus is on game-based

learning (Prensky, 2001). Undoubtedly, play takes a

signiﬁcant place in the development of humans and

animals allowing for risk-free exploration and experi-

ment. But contemporary serious games largely fail in

meeting the high expectations of game-based learning

(see, e.g., (Jantke, 2006a), (Jantke, 2007)). The state

of affair is harshly, but felicitously summarized by

Simon Egenfeldt-Nielsen in his book on the potential

of serious games as follows: “Edutainment started as

a serious attempt to create computer games that taught

children different subjects. Arguably, it ended up as

a caricature of computer games and a reactionary use

of learning theory.”((Egenfeldt-Nielsen, 2007), p. 42)

There is a necessity to abandon this state of affair.

There is abundant evidence for the need of a wide

spectrum of methodologies and tools supporting for-

http://www.uni-regensburg.de/sprache-literatur-kultur/

medieninformatik/sekretariat-team/christian-wolff/vortraege

mative evaluation of the design and implementation

of serious games.

The authors’ position advocated throughout the

present paper is to employ eye tracking analysis for

the improvement of learning systems development, in

general, and of serious games, in particular.

According to the authors’ very best knowledge,

there is not yet much systematic usage of eye tracking

in the process of designing and implementing digital

games suitable for serious purposes such as training

and learning.

The authors’ opinion advocated by means of the

present conference contribution relies on a variety of

serious games developed and implemented by their

team (see, e.g., (Jantke, 2006b), (Gaudl et al., 2009),

(Jantke et al., 2009), (Arnold et al., 2013), and (Krebs,

2013) including the usage of eye tracking analysis.

The aim of the paper includes some exempliﬁed

application of eye tracking analysis applied to some

serious game project. There is a number of qualitative

questions to be answered.

In the authors’ opinion, the usefulness of eye

tracking analysis for player-centered learning game

design and implementation can be demonstrated.

This completes the positions to be advocated below.

532

Friedemann S., Meier K. and P. Jantke K..

Here’s Looking at You, Player - The Potential of Eye Tracking Analysis for Player-centered Learning Game Design.

DOI: 10.5220/0004959205320538

In Proceedings of the 6th International Conference on Computer Supported Education (CSEDU-2014), pages 532-538

ISBN: 978-989-758-020-8

 2014 SCITEPRESS (Science and Technology Publications, Lda.)

2 INTRODUCTORY EXAMPLE

In the following example, an eye tracking study on

the playing behavior in the serious game “1961” will

show how gaze data can provide insights in the very

individual as well as common procedures of the play-

ers when using a game-based learning application.

With the eye tracking system it is possible to follow

the subjects’ visual attention on the screen. This al-

lows for a better comprehension of the users’ game

play and bears the potential to indicate where and

when learning processes may be stimulated.

2.1 Learning by Game Playing

Digital game-based learning unites serious learning

and interactive entertainment. This is done by using

the fun of gaming for motivating players who get con-

cerned themselves with serious real world problems

(see (Prensky, 2001)). The focus is primarily on the

real action which is a basis for learning by doing.

In the point & click adventure game “1961” by

Anja Hawlitschek completed at Fraunhofer IDMT in

2011, e.g., the player ﬁnds himself on some virtual

journey back in time to the virtual year 1961 on the

virtual Sunday, August 13, in the virtual Berlin when

the Berlin Wall was built (Hawlitschek, 2010).

Figure 1: Experiencing life in Berlin in the game “1961”.

What is real in the virtual world of “1961” are

the arguments in the virtual characters’ utterances, the

problems addressed, and the positions advocated. The

virtual world of the game is wrapping the real learn-

ing contents (Jantke and Lengyel, 2012).

With the digital game “1961”, the virtual world

opens up the player’s access to this time. Virtual con-

versations of affected people confront players with

real problems, arguments, and positions of that time.

Different perspectives at the conﬂict are provided.

Accordingly, those conversations are key to learning.

2.2 Didactic Approach

The underlying didactic concept of the game “1961”

is based on the model of experiential learning and thus

includes self-experiencing and the reliving of events.

But for real learning, you need real aspects that are

presented in the virtual world. Thus, the players ﬁnd

in “1961”, e.g., virtual original documents such as a

newspaper or pictures of that time with real content

for a better authenticity complementing the content

of the people’s dialogs.

Utterances of virtual characters occurring in the

game world of “1961” are designed in such a way that

every utterance represents particular contents such

as, for instance, a certain social problem, particular

economic deﬁciencies and their impact on daily life,

or some political opinion or perspective of that time

((Hawlitschek and Niegemann, 2013), sec. 2, table 1).

The player is not given a speciﬁc learning goal, but

has a clear game goal, i.e., to ﬁnd a charged battery

to, once more, reach the present (see (Hawlitschek,

2010)). While playing the user gets to know different

arguments by the game characters. The true learn-

ing contents, so to speak, is presented implicitly and

learning is assumed to take place partially unnoticed.

2.3 General Assessment

In general, learning success can be achieved espe-

cially based on experimentation, exploration and self-

experience. The player gets implicit information

about the former “world view” through the virtual

communication built on the real arguments of the

characters.

First evaluations of the game “1961” investigated

usability matters, effectiveness in terms of pedagog-

ical aims as well as game experience and cognitive

processes. Positive correlations between motivation

and learning success could be reported (see (Hawl-

itschek and Niegemann, 2013)).

Subsequently, a ﬁrst exploratory evaluation which

was carried out by using an eye tracking system in

November 2013 now provides further information on

the game design. Adolescents aged from 14 to 16

years who belong to the target group of students

attending history classes in secondary school level

in Germany took part in the play testing. The ex-

ploratory study could beneﬁt from the eye tracking

method concerningresearch questions that considered

aspects like, e.g., the decision behaviorwith respect to

the chosen ways to interact with entities, how much

time the players dedicate to certain entities, or how

entities and dialogs are perceived. Thus, an approxi-

mation to understanding gameplay takes place.

Here'sLookingatYou,Player-ThePotentialofEyeTrackingAnalysisforPlayer-centeredLearningGameDesign

533

2.4 Observations by Eye Tracking

The overall interest of this research is to gather hints

where and how implicit learning in the game “1961”

might take place and how new re-arrangements could

help to improve the game-based learning setting.

Therefore, the goal of this ﬁrst and short exploratory

evaluation using the eye tracking system was to exam-

ine to what extent the test persons follow the factual

content and which persons and objects are focused

and with which intensity, e.g., by using instruments

of ﬁxation count, gaze duration, time to ﬁrst ﬁxation

and scan paths. The focus of the evaluation was on

how dialogs are perceived and under what circum-

stances they are read by the players. Therefore, the

gaze behaviors of persons who are provided with di-

alogs were studied. It was of special interest what the

players perceived (which objects were focused and

how often), which dialogs were requested and how

these retrieved dialogs were read (which, how often

and how long, which words were focused in which

order, e.g. looking jumps and repetitions). Different

characteristics could be found and will be described

in detail with reference to one exemplary scene.

In the chosen scene (see ﬁgure 1 in chapter 2.1)

the woman at the window gives information about

ways to ﬁnd the charged battery leading back to the

present on the one hand, and about the impact on peo-

ple’s daily lives through the construction of the Wall

on the other hand. Furthermore, there is a little girl

and two women speaking who are outraged by the

construction of the Wall. Some text information are

automatically displayed in this scene and thus auto-

matically appear to the player. Other dialogs however

have to be actively requested by the user through ex-

ploring the characters by clicking on them.

During the exploratoryanalysis of the eye tracking

recordings three different behaviors of handling the

dialogs in the game could be observed. The charac-

teristics of the detected ways of text perception are de-

scribed in the following paragraphs and roughly con-

ﬁrmed in the exemplary excerpt of the data.

Characteristic 1: No Text Perception. Regarding

the data, subject 4 did not request any dialog at all in

this scene (see ﬁgure 2). No interactions except walk-

ing around took place. But the eye tracking data con-

ﬁrm that the characters as well as automatically dis-

played text were ﬁxated several times by this player

(see ﬁgure 3).

Characteristic 2: Minimum Text Perception.

Subject 3 also spent some attention on the characters

and texts in this scene. In comparison to subject 4,

some further dialogs were chosen to be read addition-

ally, but perceived only in a limited number (see ﬁg-

ure 2).

Characteristic 3: High Text Perception. The

scene was explored in great detail by subject 1 and

2 what is shown in the high number of clicked dialogs

(see ﬁgure 2)–and more importantly–in the high num-

ber of ﬁxations (see ﬁgure 3). In the given example

high ﬁxation rates are an indicator of larger numbers

of words read by the player what was also conﬁrmed

by the scan path analyses.

Figure 2: Dialogs retrieved by the subjects (click data).

Figure 3: Some differences in attention are visible in the

ﬁxation counts (eye tracking data).

Altogether, these different ways of perceiving (ad-

ditional) textual information detected during the study

• no text perception,

• minimum text perception, and

• high text perception

gave an insight in how several items that are impor-

tant for achieving learning goals are dealt with. A

questionnaire and an interview with the test subjects

conﬁrmed the eye tracking ﬁndings and investigated

some reasons for the playing behavior such as, e.g., a

very goal-oriented strategy of subject 3. This led the

player to look only for promising information in order

to make fast progress in the game–but prevented him

from experiencing more interesting details which are

part of the learning goals.

In addition, it was possible to show what objects

(dialogs, words, objects or persons) had an impact

CSEDU2014-6thInternationalConferenceonComputerSupportedEducation

534

on the game behavior. For instance, in ﬁgure 4 it is

shown how single items like words within a dialog

can lead into the game world. The scan path in this

example demonstrates that stimulus words from the

woman’s monologue are searched in the game world

such as, e.g., “shopping” or “crisis”. The objects be-

longing to these keywords were strongly focused by

the subject’s eyes and thus gained the necessary atten-

tion to become part of a learning process at all. With

the eye tracking data it became possible to see, e.g.,

the order of words read, long lasting considerations or

keyword-oriented playing.

Furthermore, the data show that some characters

are addressed more often than others, e.g., the test

persons did not speak to characters like the girl in the

middle of the scene in ﬁgure 4, although it has been

ﬁxated before by all of the subjects (ﬁgure 2 and 3).

Changes in game design like animating the girl could

open up the contact to this character. Only con-

tent gathering attention can become part of implicit

learning processes. Eye Tracking supports analyzing

where learning becomes possible at all as it shows

learners’ real actions such as reading and watching

which build a fundamental precondition for learning.

Figure 4: The scan path shows how stimulus words lead

into the scene.

To sum up, the authors’ experiment reported

above has been demonstrating that eye tracking

works, at least, for purposes such as identifying sub-

stantially different behaviors of play which correlate

with success or failure, respectively, in perceiving in-

formation crucial to human learning to be fostered

by the serious game under consideration. Roughly

speaking, it turned out, e.g., that players who act too

stringently towards success of game play are likely to

miss substantial information needed for learning.

The implications for game design and revision are

manifold. Insights arrived at by means of such an eye

tracking analysis may lead to changes of the game

mechanics, to modiﬁcations of the visual experience,

and to game adaptation to varying styles of play.

3 POTENTIALS OF THE

TECHNOLOGY

Game-based learning applications have to fulﬁll some

elementary expectations concerning the user’s game-

play experience on the one hand, and didactic inten-

tions on the other hand that might strongly vary from

the context of use of a serious game as well as from

the user’s knowledge, skills and former experiences.

In general, eye tracking can help to inspect game play

experiences and can also be used, to some extent, to

regard and evaluate the impact of didactic arrange-

ments. In this way, it can give feedback to designers

and developers of learning games. The short exem-

plary eye tracking study on “1961” has shown just an

entry to reasonable employment of the eye tracking

method in the process of serious game development.

Basically, eye tracking is widely known as a

technology to uncover usability matters of software

applications–very often in terms of website design

as it is described by (Nielsen and Pernice, 2010).

But there have already been a variety of studies ap-

plying eye tracking to digital games as well. This

deﬁnitely took greater efforts according to the com-

plexity of games as dynamic stimulus, but clearly

showed the beneﬁts of eye tracking when evaluat-

ing selected game aspects (see, e.g., (Almeida et al.,

2011), (Buscher et al., 2010), (Johansen et al., 2008))

that might impact on the player experience. It became

possible, for example, to detect preferred watching re-

gions on a screen, unremarkable objects in a game and

even differing attentional progresses of inexperienced

and hardcore gamers. Furthermore, some attempts

were already made to retrace phenomena of game

play within exploratory studies like the feeling of im-

mersion or engagement (see, e.g., (Renshaw et al.,

2009), (Jennett et al., 2008)), however, still lacking

greater examinations, a clear validation of results and

systematization of the ﬁndings according to interven-

ing variables. Furthermore, it is a highly interesting

way to get insights into the interdependenciesof game

design and the mechanisms of game playing behavior

by different users that will enable us to implement di-

dactic concepts in serious games.

After more than one decade of intense eye track-

ing research some promising novel approaches came

up lately that show some ﬁrst creative enhancements:

Gaze Data Representation. Visualization ap-

proaches that allow new insights into the gaze data

Here'sLookingatYou,Player-ThePotentialofEyeTrackingAnalysisforPlayer-centeredLearningGameDesign

535

will lead to more speciﬁc results and also to further

questions. For example, illustrations like so-called

space-time-cubes (Kurzhals and Weiskopf, 2013)

directly uncover trends in viewing behavior and

sequences of attentional synchrony of several users.

Superimposed 3D scan paths, three-dimensional

attentional maps or models of interest timelines

(Stellmach et al., 2010) allow detailed multi-

perspective inspections of views at 3D environments.

At this moment, the exact potentials for examining

game play using these techniques are not explored,

yet. Nevertheless, even more creative approaches

might be needed to face the analysis of those highly

individual game play recordings with gaze tracking.

Systematizationof Gaze Patterns. Although looks

are always depending on the tasks users are perform-

ing (see (Nielsen and Pernice, 2010), p. 13 ff. and

p. 422 ff.), there are some helpful collections of of-

ten shown gaze behaviors, e.g., the so-called f-shape

when reading web content. Further existing registers

like Ehmke et al.’s summary of eye-movement metrics

related to usability problems and the derivation of eye

tracking pattern generalizations (Ehmke and Wilson,

2007) have to be very carefully interpreted because

of missing validation, but might give a hint on user’s

gaze behavior when using websites. Systematized eye

tracking in serious gaming contexts could open a wide

area of application and make game play experience

in manifold cases more comprehensible and, thus, in

terms of game-based learning, easier to anticipate.

Feedback on Gaze Patterns. Besides concepts di-

rectly using gaze control for steering digital games

there are some interesting game input mechanisms

like in the prototypically implemented horror-game

“Sophia”

(status 2013) that uses the recognition of

eye tracking patterns while playing as a parameter

to inﬂuence the story and–in this case–to terrify the

player by unexpected incidents. Seen from a didactic

point of view, concepts like this provide many ideas to

implement situations adequate for implicit learning.

Linking Gaze Patterns to Storyboards. In terms

of achieving player-centered learning game design,

eye tracking might become one key tool (among oth-

ers) to inspect crucial game sections and create ap-

propriate story alternatives upon the results. This

was basically shown in the introductory example on

“1961” (see chapter 2) when the progress of attention

on learning-relevantobjects has been analyzed.Future

research by the authors will examine, if and how such

ﬁndings can be implemented in a digital storyboard-

ing system that allows to anticipate different user-

http://www.uni-regensburg.de/pressearchiv/

pressemitteilung/302195.html

centered learning experiences by means of user mod-

eling and adaptive system behavior. Potentially, eye

tracking could contribute to game-based learning sce-

narios that provide customized didactic patterns un-

derlying a human user’s game play–just as he freely

explores a scene by playing.

4 LIMITATIONS OF THE

TECHNOLOGY

While eye tracking is already well-known by web-

site developers and is also partly becoming of inter-

est to the video games industry (see, e.g., (Almeida,

2012)) its potentials have not yet been systematically

explored and extended to the ﬁeld of serious gaming.

When starting to discuss possible potentials of eye

tracking to impact on serious games concepts several

limitations have to be considered that underlie several

reasons, such as:

• Technological Limitations

– the fact of games being a special case of dy-

namic 2D- or even 3D-stimuli being controlled

by indiviual users and causing individualized

data in every session that can hardly be com-

pared one to another,

– missing automation that helps processing the

huge amounts of individual data sets, e.g. in

recognizing dynamic areas of interest (AOI) or

ﬁnding object-dependent patterns in the gaze

recordings, as well as

– a “lack of suitable [analysis] techniques” that

help to overcome time-consuming frame-by-

frame-analyses like it is described by ((Stell-

mach et al., 2010)).

• Methodological Limitations

– the necessacity of combining eye tracking data

in carefully controlled experiments with addi-

tional data by further methods like retrospective

thinking aloud-protocols (see, e.g., (Eger et al.,

2007)) to extend the information value of the

stand-alone-investigations,

– still missing links in between eye tracking data

and further data gathered to describe game play

experience (Nacke et al., 2010) like using dif-

ferent methods like questionnaires, interviews,

game metrics, psychophysiological player test-

ing etc., and

– missing validation of gaze patterns described

in manifold exploratory case studies such as,

for instance,(Ehmke and Wilson, 2007) and

(Kivikangas et al., 2010).

CSEDU2014-6thInternationalConferenceonComputerSupportedEducation

536

To sum up, in order to spread eye tracking as a

supporting method in serious game concepting further

analysis tools need to be developed to face speciﬁc re-

search questions as well as technical challenges, e.g.

concerning the comparability of data sets, combina-

tion of eye tracking data with data of other sources

and creative ways of visualising the gaze data to en-

able new possibilities of analysis and novel ﬁndings.

5 CONCLUSIONS & OUTLOOK

The authors admit that a comprehensive evaluation of

“1961” based on eye tracking is still badly missing, to

some extent, due to several issues raised in section 4.

The game “1961” has been used to exemplify the

way in which eye tracking analysis may lead to some

key insights into problems of serious game design.

Missing utterances in “1961” is really a crucial issue.

Nowadays, eye tracking still is a costly, but in

many ﬁelds valuable method to evaluate and support

software development. Taking advantage from the

eye tracking ﬁndings in the area of serious gaming

in educational contexts, e.g., for offering individual

player experiences or speciﬁc didactic approaches,

is still open. Systematizing available ﬁndings and re-

deﬁning requirements of eye tracking analyses will

help fully exploring the potentials of the technology.

REFERENCES

Almeida, S. (2012). Video games and eye tracking: Does

the video game industry know eye tracking? In Play,

Game, and Society, Proc. of Videojogos 2012, 5th An-

nual Conf. in the Science and Art of Video Games,

pages 269–281.

Almeida, S., Veloso, A., Roque, L., and Mealha, O. (2011).

The eyes and games: A survey of visual attention and

eye tracking input in video games. In SBC, Proc. of

SBGames 2011.

Arnold, S., Fujima, J., Karsten, A., and Simeit, H. (2013).

Adaptive behavior with user modeling and story-

boarding in serious games. In 9th International Conf.

on Signal Image Technology & Internet-based Sys-

tems, Dec. 2-5, 2013, Kyoto, Japan.

Buscher, G., Biedert, R., Heinesch, D., and Dengel, A.

(2010). Eye tracking analysis of preferred reading re-

gions on the screen. In ACM Conf. on Human Factors

in Computing Systems (28th CHI conference 2010).

Egenfeldt-Nielsen, S. (2007). Educational Potential of

Computer Games. Continuum Studies in Education.

Continuum Intl. Publ. Group.

Eger, N., Ball, L. J., Stevens, R., and Dodd, J. (2007).

Cueing retrospective verbal reports in usability testing

through eye-movement replay. In Proc. of HCI 2007.

British Computer Society.

Ehmke, C. and Wilson, S. (2007). Identifying web usabil-

ity problems from eye-tracking data. In Proc. of HCI

2007. British Computer Society.

Gaudl, S., Jantke, K. P., and Woelfert, C. (2009). The good,

the bad and the ugly: Short stories in short game play.

In Proceedings of the 2nd International Conference

on Digital Storytelling, number 5915 in LNCS, pages

127–133. Springer-Verlag Berlin Heidelberg 2009.

Hawlitschek, A. (2010). Ein digitales Lernspiel f¨ur den

Geschichtsunterricht: Konzeption und Evaluation. In

eLearning Baltics, eLBa 2010, Proc. of the 3rd Intl.

Science Conference, pages 278–288.

Hawlitschek, A. and Niegemann, H. M. (2013). Geschichte

Lernen – Didaktisches Design digitaler Lernspiele.

Medienproduktion – Online-Zeitschrift f¨ur Wis-

senschaft und Praxis, (III):15–17.

Jantke, K. P. (2006a). Digital games that teach: A critical

analysis. Diskussionsbeitr¨age 22, TUI IfMK.

Jantke, K. P. (2006b). Knowledge evolution in game design

– just for fun. In CSIT 2006, Amman, Jordan, April

5-7, 2006.

Jantke, K. P. (2007). Serious Games – eine kritische Anal-

yse. In 11. Workshop Multimedia in Bildung und

Unternehmen “eLearning and Serious Games”, TU

Ilmenau, 20./21.09.2007, pages 7–14. TU Ilmenau,

ISSN 1436-4492.

Jantke, K. P., Gaudl, S., Hawlitschek, A., Hoppe, I., and

Lengyel, D. (2009). Knowledge games & knowl-

edge engineering. In Proc. 7th Conf. Computer Meth-

ods and Systems, Krak´ow, Poland, 26-27 November

2009, pages 129–134. Oprogramowanie Naukowo-

Techniczne.

Jantke, K. P. and Lengyel, D. (2012). Die Realit¨at in vir-

tuellen Welten. Zeitschrift f¨ur e-Learning, (1):7–22.

Jennett, C., Cox, A. L., Cairns, P., Dhoparee, S., Epps,

A., Tijs, T., and Walton, A. (2008). Measuring

and deﬁning the experience of immersion in games.

International Journal of Human-Computer Studies,

66(9):641–661. Academic Press, Inc. Duluth, MN,

USA.

Johansen, S. A., Nørgaard, M., and Rau, J. (2008). Can

eye tracking boost usability evaluation of computer

games? In CHI 2008: Evaluating User Experiences

in Games.

Kivikangas, J. M., Ekman, I., Chanel, G., J¨arvel¨a, S., Cow-

ley, B., Salminen, M., Henttonen, P., and Ravaja, N.

(2010). Review on psychophysiological methods in

game research. In Proceedings of 1st Nordic DiGRA,

DiGRA.

Krebs, J. (2013). Moral dilemmas in serious games. In

Proc. of the International Conference on Advanced

Information and Communication Technology for Ed-

ucation (ICAICTE 2013), Sept. 20-22, 2013, Hainan,

China, pages 232–236. Atlantis Press.

Kurzhals, K. and Weiskopf, D. (2013). Space-time vi-

sual analytics of eye-tracking data for dynamic stim-

uli. IEEE Transactions on Visualization and Com-

puter Graphics, 19(12):2129–2138.

Nacke, L., Drachen, A., and G¨obel, S. (2010). Methods for

evaluating gameplay experience in a serious gaming

context. International Journal of Computer Science

in Sport (Special Edition), Volume 9(2).

Here'sLookingatYou,Player-ThePotentialofEyeTrackingAnalysisforPlayer-centeredLearningGameDesign

537

Nielsen, J. and Pernice, K. (2010). Eye Tracking Web Us-

ability. New Riders, Berkeley.

Prensky, M. (2001). Digital Game-Based Learning.

McGraw-Hill.

Renshaw, T., Stevens, R., and Denton, P. D. (2009). To-

wards understanding engagement in games: An eye-

tracking study. On the Horizon, 17(4):408–420.

Emerald Group Publishing Limited.

Stellmach, S., Nacke, L., and Dachselt, R. (2010). Ad-

vanced gaze visualizations for three-dimensional vir-

tual environments. In Proc. of the 2010 Symposium on

Eye-Tracking Research & Applications, ETRA, pages

109–112.

CSEDU2014-6thInternationalConferenceonComputerSupportedEducation

538