Teaching Me Softly: Experiences and Reflections On Informal Educational Game Design
Teaching Me Softly: Experiences and Reflections On Informal Educational Game Design
Introduction
Game-based learning has become an important issue for economy, society and
research. The demand of lifelong learning increases the need for eective, motivating learning tools. Meanwhile, computer graphics and game technology made
a quantum leap. Moreover, computer games have become common and popular
among young people. As a result, scientists, educators and pedagogues consider
computer games as an educational tool. However, this subject still has issues.
This article addresses one of them inferior integration of the learning content
into the game and reects some theoretical thoughts and practical experiences
regarding a more elegant and subtle educational game design.
At rst, this article distinguishes play, games and computer games, as well as several uses of game-based learning. Part two discusses the potential of game-based
learning from a pedagogical perspective, stressing on motivational and didactical
benets of digital games. The following sections describe a way to combine educational content with video games and its practical application within the Eduventure
Z. Pan et al. (Eds.): Transactions on Edutainment II, LNCS 5660, pp. 90104, 2009.
c Springer-Verlag Berlin Heidelberg 2009
91
II, a learning adventure targeting historical knowledge. Last, but not least, some
conclusions regarding the used approach and further research are drawn.
This article will be prefaced with a short paragraph on terminology. Play takes
place within a frame often regarded as the magic circle [1, p. 94]. Within
this frame, the player does not encounter negative consequences or functional
pressures, which thereby supports lusory behaviour [2]. The world within the
magic circle is not only risk-free, but also imaginary. Players overcome the borders of reality and apply new meanings to objects: During play, a matchbox can
become a train, a car, a house or whatever the player wants it to be. Fabricatore
refers to this circumstance as the concept of as if [2].
Salen and Zimmerman dene a game as a system in which players engage
in an articial conict, dened by rules, that results in a quantiable outcome
[1, p. 80]. By mentioning the articial character of game worlds, this denition
adapts the concept of as if . To simplify matters, this paper uses the terms
digital game, video game and computer game synonymously, following Juuls
denition: A video or computer game is, generally speaking, a game played
using computer power and a video display [3, p. 36].
The terminology on game-based learning, however, still seems to be somewhat fuzzy and vague. Nevertheless, one could distinguish two forms of learning
games. Serious games focus on the technical potential of video games for training and simulation, stressing authenticity and realism of the educational content.
In contrast, educational games try to preserve a balance of fun and education,
imparting the educational content in an entertaining and playful manner [4].
Both serious games and educational games (and even regular video games when
applied in an educational context) can be used for several purposes, which I suggest to break down into four dierent ideal types. First, learning games can take
over the role of a teacher, imparting knowledge and information. Secondly, they
can be designed in order to train specic capabilities and skills. They can also be
used for overlearning and recapitulating (as in the Triple A Game Show [5]).
Last, but not least, pedagogues can use them as some sort of bait , which makes
it much easier for them to get in touch with their clients and build a relationship.
However, this article does not cover all possible forms and purposes of learning
games. It focuses (a) on educational games that are (b) used for teaching.
Computer games can be powerful educational tools mainly for two reasons: First,
they incorporate well-established learning principles and provide resources for
authentic simulation. Secondly, computer games are fun to play and therefore
engaging. However, this does not automatically mean that educational games
benet from these factors as well. The following section provides background
information and research ndings concerning these issues.
92
3.1
U. Wechselberger
Playing is often connected to learning processes. Fabricatore for instance connects the process of playing video games with several learning occasions (cf.
gure 1), such as training of analytic, strategic and psychomotor skills as well
as acquisition of game-relevant knowledge [2]. Moreover, scientists have indicated several learning principles embodied in computer games. For instance,
Gentile and Gentile found seven factors benecial for learning, including adaptation to player skills, interactivity, practice and overlearning [6]. Van Eck states
that learning in computer games takes place within a meaningful (to the game)
context [7, p. 18] and emphasises the role of a relevant context for learning
principles (such as situated cognition and viability). Gee even elaborated 36 educational values of computer games, including motivation to start and keep on
learning, customisation, possibility to test hypotheses and others [8]. Also, due
to the increasing performance of state of the art hardware, video games can
simulate graphics, physics and even articial intelligence (e.g. in the game facade [9]). This enables the player to interact with systems that (for dierent
reasons) can not be examined in real life (e.g. historical societies, hazardous environments, biological microsystems). Thereupon, one could consider computer
games as eective learning environments.
On the other hand, research ndings about learning in video games are unsatisfying. On the one hand, there is evidence of computer games very eectively
supporting a narrow scope of psychomotor skills. They also train cognitive capabilities such as visual attention skills, spatial visualisation, problem solving,
Fig. 1. Learning occasions within computer game play regarding to Fabricatore [2]
93
creative and critical thought [10]. However, research ndings about knowledge
acquisition through video games are rare and contradictous. This also applies to
the internalisation of norms and moral values. Both values and information seem
to be bound within the magic circle, which may in many cases block learning
transfer into real life. This may be due to certain theoretical problems which
arise when dealing with complex issues such as knowledge construction. Section
4.2 will deal with this issue in detail.
3.2
Playing games is fun. Play and games have been used to enhance the interest
of adolescents in useful, yet somewhat unexciting matters (for example curricular contents) for decades. Thus, the most important hidden agenda regarding
game-based learning is to utilise the motivational power of play and games for
educational purposes. But where does this motivational potential come from?
Many scientists believe that the power a player can wield over his environment
while still challenged is a very important source of intrinsic motivation [11]. Computer games provide instant feedback and thereby deliver instant gratication.
They also come up with other motivating features such as aesthetic graphical representation, a wide variation of options and scenarios, and others [12,
pp. 128129]. Accordingly, computer games appear to have great motivational
potential.
However, although video games are very entertaining, learning games are not
necessarily fun as well. Two main problems may come up when trying to benet from the motivational potential by using learning games in an educational
context: As mentioned earlier in this article, play is free of functional pressures
and does not have negative consequences. Thus, a teacher cannot impose playing a game on a pupil without running the risk of disrupting the nature of play
(and thereby corrupting part of its motivational source). Video games may become teacherised in an educational context [4], leading to the loss of their
engaging potential. This is a constitutive problem with play and games used for
teaching: By being exploited for purposes outside of the magic circle, play and
games become serious, which is somehow paradox. This constitutive issue may
not be xable by game design. It is, however, more of a hypothetical nature. The
second problem is more of a practical nature and concerns game design. Curricular objectives and game goals are often incongruent [11,13], and, as a result,
educational games are often of inferior quality. Seymour Papert criticised the
low-grade design of most edutainment products and educational games: Shavian reversals ospring that keep the bad features of each parent and lose the
good ones are visible in most software products that claim to come from a
mating of education and entertainment [14, p. 88]. Indeed, in most educational
games there seems to be little cohesion between educational content and game
structure [2]. Integration of the curricular content is stilted and articial, thus
gameplay, fun and entertainment are corrupted.
The here presented article concentrates on the second issue.
94
U. Wechselberger
In order to overcome the obstacle of inferior educational game design, one has
to nd subtle, elegant ways to connect video game structure and educational
content. Fabricatore proposed that educational game designers should embed
knowledge naturally and with contextual relevance into gameplay [2], and
curricular tasks should become elements of the gameplay. He exemplies his
thoughts by describing a Super Mario game. In this game, collectable coins
are replaced by educational content (e.g. numbers). This approach can be taken
even one step further. First, the connection between learning tasks and gameplay
could be strengthened by creating a functional, semantic relation between them
(this thought will be elaborated in the game player section later in this article
(cf. section 4.1)). Secondly, it might not be possible to connect any kind of
information to gameplay. For these contents one has to identify and elaborate
further information carriers within digital games. In order to achieve this, one
has to understand the semiotic structure of video games.
4.1
A useful semiotic framework, suitable both for game analysis and design, was
introduced by Lindley [15]. He refers to video games as ludic systems involving
three layers of encoding, each having dierent tradition, language and methodology: game play, narrative and simulation. Various game genres may focus on
dierent layers (cf. gure 2).
95
Game Layer. The most obvious semiotic level of video games is the game
layer. Many abstract games like Pong [16] take place primarily on this layer.
Lindley describes it as a framework of agreed rules [15]. These rules dene
the possibilities and boundaries of what a player can legally do while playing
a game. They also specify the consequences of player actions. According to the
more practical denition by Adams and Rollings [17, pp. 1214], gameplay is
constituted by the challenges a player faces during the game as well as the
actions he addresses them with. Adams and Rollings describe numerous common
challenges, for example:
96
U. Wechselberger
These parts may provide many information carriers for curricular content. For
instance, designers of an educational game could bring back historical celebrities to a new virtual life in the form of non-player-characters. Information on
the background of these celebrities could be connected to character masks and
behaviour. Common plot structures and story techniques may be used to add
connotations which may also encode curricular information (just imagine the
semantic dierence it makes if you connect a renowned, historical president not
to the role of a hero, but to the role of a villain instead).
Simulation Level. Simulation games like Civilization [19] focus on the simulation level. According to Lindley, the simulation level addresses basic features and functions of the game world: It is the level at which the authored
logic and parameters of a game system together with the specic interactive
choices of the player determine an (implied) diegetic (i.e. represented) world
[15]. Logic and parameters include, but are not limited to animation, physics,
non-player character behaviour (and others). Moreover, this layer of semiotic
encoding is based on an underlying set of rules, functions and constraints,
developed by the game designers. Modern computers and game engines are
capable of simulating these sets quite realistically and convincing. Interdependencies within complex systems (for instance ecosystems) can be rendered in
realtime, thus creating authentic interactive simulation environments. In addition, modern computer technology simulate physics and photorealistic graphics.
For educational games, these simulations may be directly connected to curricular
content.
Conclusion. Lindleys framework provides three useful semiotic levels and
thereby reveals adequate information carriers. Embedding knowledge deeply into
these semiotic layers could lead to a harmonic and subtle educational game design, preserving game structure and thereby fun and entertainment. However,
there may be a downside of a suchlike approach, which I would like to touch in
the following section.
4.2
97
Although all bags contained identical foods, the test subjects stated that the
ones in the branded wrapper tasted better than the others. Robinson and his
colleagues traced this result back to the impact of branding on the childrens
perception. This research may be connected to cognitive frames (branded foods
are generally better than unbranded foods) which are activated by cues and context (branded wrappers) and produce (subjective) meaning and information (I
like this food better than that one).
The concept of schemata and frames can also be applied to the reception of
educational games. Imagine adolescents, playing a leisure game (or an educational game, subtly combining curricular content and gameplay) they believe to
have no didactical purposes. Regarding the denition of play and games, the
players stay within the magic circle, an articial world mentally separated from
the real word, following the concept of as if (cf. section 2). In principle, this
magic circle is nothing but cognitive frame, separating meanings of the game
world from the ones in the real world and organising the players perceptions.
Thus, if there is no wrapping branding the game elements as educational content, the players may only activate their play and games schemata. Because
without any curricular cues, how should they know that they are dealing with
authentic pieces of information that are supposed to be used in the real world?
As a result, when it comes to knowledge acquisition through game-based learning, educational content may theoretically be overlooked. This hypothesis might
apply mostly to (a) games that teach information (whose content is especially
context-specic) and (b) educational games and regular video games (because,
contrary to serious games, they usually miss the this is educational content
cues). This hypothesis might explain the research ndings mentioned above,
which primarily document the training of skills and capabilities rather than the
teaching knowledge in computer games.
Both this issue and the approach of an informal educational game design as
described earlier have been researched within a project that was carried out
by the Institute For Computational Visualistics at the University of KoblenzLandau in Germany. This project will be introduced in the following section.
5
5.1
The Eduventure-II-Project
Research Questions and Project Procedure
98
U. Wechselberger
curricular content was dened. In the meantime, video game semiotics as well as
the combination of learning content and game structure was investigated. Based
on these two steps an educational game was designed. After that, the game
prototype was implemented, which was meant to be followed by the evaluation
of the design approach and prototype. However, due to a cut of funding, the
project was discontinued during the implementation phase. Nevertheless, our
project team was able to design a game concept and implement a working, yet
reduced prototype. The next section deals with the details of the concept.
5.2
Educational Content
For many pupils, history is a dry subject, and they are not particularly engaged by history lessons. The potential motivational and didactical benets of
video games may be of great value when addressing this issue. For this reason,
the project focused on German cultural heritage and history. The curriculum
consisted primarily of the following learning contents:
After years of democracy under occupying forces from France, the city of
Koblenz got under Prussian rule due to the Congress of Vienna in 1814/1815
and was subjected to the nationalistic authority of Prussia.
After obtaining the city of Koblenz, the Prussian occupiers built a huge
fortication system, designed in conformance with the most recent ndings
in fortication technology, protecting the city.
This fortication system was said to be impregnable by force. It could only
be compromised by either treason or espionage.
There were plenty of disagreements between the citizens of Koblenz and the
Prussian forces, based on ideological and organisational issues.
Life was hard for the Prussian soldiers, for the fortress was not very comfortable and daily routines were exhausting.
In 1848, the civilian population in the German states rebelled against the
authorities, demanding freedom, democracy and national unity.
99
Game Description
Game Concept. The Eduventure-II-prototype is a classic role playing / adventure game. We chose this game type because its structure is evenly distributed
among the three semiotic layers mentioned above (cf. gure 2).
Within the game, the player assumes the role of a thief. He is hired by an
eminence grise to inltrate the Prussian fortress during the Revolutions of 1848
in the German states. His objectives are to gather as much information about the
Prussian forces as possible, to capture the contents of a mysterious box brought
to the headquarters of the fortress and to nd vulnerabilities in the Prussian
system of defence. To accomplish this, the player character joins a group of
workers that are arming the fortress. Next, he waits for the night in a hiding
place. After sunset, he sneaks into the fortress, searching for an adequate uniform
he could wear during the day. Having found a uniform, he is able to enter the
core of the fortress and nally full his remaining objective.
This gameplay experience was connected to the educational content that was
described above. On the simulation layer, we encoded information about uniforms, equipment, architectural structures and tasks regarding the arming. For
example, virtual uniforms and models are authentic. Also, we paid much attention to a high recognition value of the architecture. Additionally, we resorted
to the game engines capability of importing heightmaps for terrain rendering.
This way we were able to use real geoinformation (acquired by the local land
surveying oce) for the simulation of the fortresses environment (cf. gure 4).
Furthermore, we encoded information on the narrative level. Story and virtual
characters were connected to information about the background of the revolutions, the conict between civilians and Prussian forces, and what life was like
for a soldier working at the fortress. For instance, dialogues with labourers and
soldiers reveal details about the dierences between the civilians and military
forces. Also, some archetypical virtual characters demonstrate the rough mentality of the Prussian occupying forces. In addition, the recruitment of the player
character as a secret agent (and therefore the initial point of the story) is linked
directly to the political background in 1848. In order to accentuate the tremendous obstacle the fortresses walls posed for potential attackers (cf. gure 5), we
connected the task of conquering it to the common plot element crossing the
threshold (cf. [18, pp. 61-62].
Last but not least, we connected some educational elements to the gameplay
layer. Knowledge encoded on this layer includes Prussian fortication strategy,
the tasks arming consisted of, as well as Prussian uniforms and equipment. For
example, the player has to disguise himself as a Prussian soldier. Addressing factual knowledge challenge (cf. section 4.1), this task would result in getting caught
100
U. Wechselberger
Fig. 5. The fortresses main wall (on the left) is connected to the common narrative
element crossing the threshold
if the player chose the wrong uniform. Furthermore, the player has to perform
authentic tasks coming up as the Prussians arm the fortress. Last, but not least
he has to investigate and exploit authentic weaknesses within the fortication
system in order to advance to the fortresses headquarters.
Game Technology. The Eduventure-II-prototype is a modication of the roleplaying-game The Elder Scrolls IV: Oblivion [22]. Oblivion seemed like a
fair choice because of its freely available construction set (used for creating
101
Player Reactions
Due to cut of funding, our ndings on player reactions rest upon occasional, informal observations with only a handful of individuals instead of the originally
intended, methodologically elaborated evaluation. Although this may have led to
a lack of validity (considering statistical standards), our rst experiences seem to
conrm both assumptions regarding perception and reduced knowledge acquisition and increased motivation due to coherent game design (cf. section 4).
Regarding motivational eects, the game engines aesthetics seemed to attract
young players. An early version of the Eduventures prototype was presented
at the computer expo Cebit in 2007 and seemed to be quite an eye-catcher.
Young visitors enjoyed the aesthetics of the virtual fortress and characters. These
players did not know about the educational purpose of the Eduventure. They
just moved around the virtual territory and interacted with the game for the
sheer fun of it. Unfortunately, it is an open question if they recognised any of
the curricular contents of the game. However, they clearly had fun playing the
game and were not scared away by stilted and articial game design. Thus,
the approach to educational game design as described above seems to preserve
game structure and entertainment.
Older players provided some insights on the perception of the educational content. They had been told about the didactical intentions of the game. As a result,
they expected a distinct presentation of educational content, pointed out by cues
(cf. section 4.2), which they could not nd due to the subtle design approach.
102
U. Wechselberger
One player even played several parts of the prototype and nally asked: Okay,
but where is the educational content? Concerning our goal to overcome the gap
between educational content and gameplay in order to preserve entertainment
and motivation, this could be considered as a positive result. Subtle educational
game design prevents the game from becoming teacherised , thereby supporting
motivation and entertainment. However, when it comes to instructional potential, the informal game design and absence of educational framing clues (cf.
section 4.2) may cause players to overlook educational content. It is still unclear
if this leads to reduced knowledge acquisition.
Lessons Learnt
In conclusion, subtle educational game design, which connects curricular elements to gameplay, story and simulation, preserves game structure. This has a
positive eect on the players engagement and motivation. However, compared to
conventional educational game design, this approach apparently leads to framing
processes which make the players mentally remain within the magic circle (cf.
section 2). As a consequence, one might suggest to add more curricular cues to
the gameplay and its social context in order to conduct the players away from the
magic circle back to reality. But this could possibly amplify the constitutional
conict of play vs. seriousness mentioned above (cf. section 3.2). This issue is
currently treated by a new project of our working group (cf. section 7).
Within the scope of the Eduventure-II-project, we experienced that in practice theoretical thoughts may lose some of their discriminatory power. Lindleys
framework of semiotic levels within video games is a great tool for analysing games.
However, in practice, borders between semiotic layers may become blurred. For instance, within a role-playing game, gameplay and narrative usually interdepend:
The plot produces game objectives, which, in return, aect the story. Therefore,
the distinction between two layers of semiotic encoding is sometimes more of a
theoretical nature, as the practical game experience seems to take place on several
blended semiotic layers. This does not only aect game design, but also its evaluation (for instance when one tries to nd out if it makes any dierence if curricular
information is connected to the gameplay layer instead to the narrative). Therefore, for evaluation purposes it is important to keep these layers apart from each
other in practical gameplay. Otherwise, one would produce measurement artefacts,
thereby reducing the signicance of his research results.
Apart from that, there are other factors inuencing the impact of educational
games. Scientists point out that not only the content, but also the amount of
video game play, as well as interaction mechanics, may signicantly aect learning outcomes [24]. Furthermore, some curricular contents may receive more attention than others (especially when connected to dierent semiotic layers or
even dierent game tasks). These eects may be hard to foresee in advance.
Investigating the instructional and motivational potential of a game design
approach as described above requires a functional, well designed game with
high discriminatory power. Complex games and game engines (like the
103
Eduventure-II-prototype) may not be ideal for this purpose, for they can sustain
exceptionally many technical pitfalls. Therefore, one might consider to concentrate on less complex game types like casual games, implemented within a less
complex framework (e.g. Adobe Flash). Moreover we would encourage interdisciplinary projects, bundling resources and competencies of academics (from both
educational and computer science) as well as professional game designers. This
way, scientists could concentrate on their research and leave the practical design
tasks to the experience and creativity of game designers.
Future Work
Framing processes and knowledge acquisition within computer games are highly
complex, theoretical concepts, which have not yet been entirely investigated.
For example, not only the game design and method of knowledge encoding may
function as clues used during the framing process. The social and situational
context in which the player encounters the game may play an important role
as well. As a result, the issue of player perception within subtle educational
games needs further research.
This also applies to the issue of motivational aspects. Connecting educational
content to the semiotic structure of the video game might solve the problem of
inferior and therefore unexciting educational game design. However, it is still
unclear if the hypothetical constitutional conict of play vs. seriousness (cf.
section 3.2) has any impact on the players engagement.
Both didactical and motivational issues are addressed by a research project
currently carried out by the Institute For Computational Visualistics at the
University of Koblenz-Landau, Germany. The objectives of this project are
to develop and evaluate a theoretical framework specifying the impact of
context and clues on cognitive framing
and to measure potential eects of framing processes on both motivation
and information.
Primarily, the computer games required for these tasks will be small and uncomplicated casual games based on Adobe Flash. In the meantime, our team
is establishing contact to prospective game designers that might support us in
designing the games.
References
1. Salen, K., Zimmerman, E.: Rules of Play. In: Game Design Fundamentals. MIT
Press, Cambridge (2004)
2. Fabricatore, C.: Learning and videogames: An unexploited synergy. In: 2000 AECT
National Convention, Long Beach (2000),
http://www.learndev.org/dl/FabricatoreAECT2000.PDF (retrieved March 29,
2009)
3. Juul, J.: Half-real. In: Video Games between Real Rules and Fictional Worlds. The
MIT Press, Cambridge (2005)
104
U. Wechselberger