Content uploaded by Edward Melcer
Author content
All content in this area was uploaded by Edward Melcer on Oct 10, 2021
Content may be subject to copyright.
285
What is a Visual Novel?
JANELYNN CAMINGUE, University of California, Santa Cruz, USA
ELIN CARSTENSDOTTIR, University of California, Santa Cruz, USA
EDWARD F. MELCER, University of California, Santa Cruz, USA
Visual Novel (VN) is a widely recognizable genre of narrative-focused games that has grown in popularity over
the past decade. Surprisingly, despite being so widely recognizable, there is not a singular denition to help
guide the design and analysis of such games—with academic denitions and implementations ranging from
"interactive textbooks" to "adventure games with multi-ending stories". In this paper, we present a unied
denition of VNs drawn from an analysis of 30 prior academic denitions. We also examined 54 existing VNs
to further rene our denition and employ a deeper analysis of the interactivity within VNs. We highlight
key features of VNs that arise in our denition, and discuss resulting implications for the design of VNs. This
work is relevant for narrative game designers and researchers, aording a more unied structure and clearer
guidelines to identify, analyze, and design future VN games.
CCS Concepts: •Human-centered computing →Empirical studies in HCI.
Additional Key Words and Phrases: Visual Novel, interactive narrative, meta-analysis, video games
ACM Reference Format:
Janelynn Camingue, Elin Carstensdottir, and Edward F. Melcer. 2021. What is a Visual Novel?. Proc. ACM
Hum.-Comput. Interact. 5, CHI PLAY, Article 285 (September 2021), 18 pages. https://doi.org/10.1145/3474712
1 INTRODUCTION
The Visual Novel (VN) is a widely popular and recognizable narrative-focused game genre that
originated in Japan [
12
,
28
,
67
,
77
]. Their popularity has been steadily increasing outside of Japan
with 2,272 VNs currently available for purchase on Steam within the US
1
, and 27,140 VNs archived
on community run websites such as VNDB
2
that aim to document the production of VNs. This
popularity has been bolstered further still with the increase of free VN game engines such as
Ren’Py [
18
]. VN engines like Ren’Py provide a simple scripting language and easy to use graphical
user interface, which have made the genre accessible to designers and authors without technical
backgrounds. This in turn has made VNs simple to construct and beginner friendly for novice
designers, while simultaneously allowing expert designers to produce complex and engaging work.
Furthermore, the overall ease of creation, distribution, and accessibility for players at a variety
of skill levels with respect to VNs has also led to their frequent use in academic research—most
commonly in high-impact domains such as education [
4
,
13
,
26
,
29
,
31
,
40
,
49
,
51
,
68
,
87
] and health
1Number collected on April 18th, 2020
2The Visual Novel Database — https://vndb.org, number collected on April 18th, 2020
Authors’ addresses: Janelynn Camingue, University of California, Santa Cruz, USA, 1156 High St, Santa Cruz, CA, 95064,
jcamingu@ucsc.edu; Elin Carstensdottir, University of California, Santa Cruz, USA, 1156 High St, Santa Cruz, CA, 95064,
ecarsten@ucsc.edu; Edward F. Melcer, University of California, Santa Cruz, USA, 1156 High St, Santa Cruz, CA, 95064,
eddie.melcer@ucsc.edu.
Permission to make digital or hard copies of part or all of this work for personal or classroom use is granted without fee
provided that copies are not made or distributed for prot or commercial advantage and that copies bear this notice and
the full citation on the rst page. Copyrights for third-party components of this work must be honored. For all other uses,
contact the owner/author(s).
©2021 Copyright held by the owner/author(s).
2573-0142/2021/9-ART285
https://doi.org/10.1145/3474712
Proc. ACM Hum.-Comput. Interact., Vol. 5, No. CHI PLAY, Article 285. Publication date: September 2021.
This work is licensed under a Creative Commons Attribution International 4.0 License.
© 2021 Copyright held by the owner/author(s).
2573-0142/2021/9-ART285. https://doi.org/10.1145/3474712
285:2 Janelynn Camingue, Elin Carstensdoir, & Edward F. Melcer
[
77
,
93
]. This highlights the signicance of VNs as an area of study in their own right, and the
value that improving understanding of such designs can provide.
Unsurprisingly, as a result of the popularity of VNs both commercially and within research,
there have been a number of denitions created to provide a high-level description of key design
features underlying the genre. However, these denitions can vary wildly, which in turn has
substantial implications for the design of such games. For instance, denitions have included "VNs
are interactive textbooks" [
87
], which is in substantial contrast to “Adventure Games born in Japan,
that makes use of attractive characters, narrative engagement, puzzles and other interactive features
to maintain user interest while submerging players in complex stories" [
77
]. This wide range of
denitions can prove problematic for both designers and researchers of VNs, providing notably
dierent (under and over constrained) interpretations of the potential design space for such games.
Despite this lack of consensus on a clear denition of VNs as a genre, they are still easily
identiable as such by the publishers who produce them and the community of players who enjoy
them. Examples of this consensus can be seen through fan community documentation eorts
on VNDB
2
as well as the community genre labels on Steam. Furthermore, IndieDB
3
(a popular
database for indie games) has a specic genre section for Visual Novel, as well as Twitch
4
(the
largest platform for streaming games) which also has a tag for Visual Novel. This indicates that
there is a set of very specic criteria that can be used to identify VNs, which also implies that a
more unied denition and formalized understanding of VNs is achievable.
While there is evidence of an implicit understanding of VNs illustrated above, there is no clearly
dened census on the set of features that dene them—as exemplied by the 2,556 tags used to
describe features of VNs on VNDB
5
. Furthermore, these tags pertain to both the content and
structure of VNs. A lack of clear denition impacts the ability of designers to create VNs and realize
the full potential design space of the genre, and similarly makes classifying and analysing VN
games dicult for researchers. Ultimately, creating a unied denition of VNs as a genre as well as
understanding the key elements of a VN’s design allows us to better inform future work in the area
of VNs and interactive narrative.
In this paper, we propose a unied denition of VNs developed from analyzing existing denitions
of VNs as a genre and playing through a corpus of 54 VNs to rene the analysis. We then discuss
key features from our unied denition and their implications for the enhanced design of future
VNs.
2 RELATED WORKS
2.1 Game Genre Studies
Game genres are conceptual tools used to understand games and their contexts [
20
,
21
,
33
,
83
].
While not the only tool to classify games [
3
,
8
,
20
,
24
], genre is arguably the most widely used
manner to describe, group, and contrast games based on their dening characteristics. For games,
genre is commonly conceptualized, and sometimes named after, specic sets of mechanics, control
schemas, player goals, narrative structure and more, e.g.,Metroidvania, Walking Simulators, and
Role-Playing Games (RPGs). As an illustrative example, the well established rst-person shooter
(FPS) genre broadly consists of rst-person camera perspective, where the core gameplay consists
of the player pointing and clicking to a location where a projectile or eect is meant to make an
impact of some description, often requiring movement of the player’s camera perspective through
3https://www.indiedb.com/
4https://www.twitch.tv/
5The Visual Novel Database — https://vndb.org, number collected on April 18th, 2020
Proc. ACM Hum.-Comput. Interact., Vol. 5, No. CHI PLAY, Article 285. Publication date: September 2021.
What is a Visual Novel? 285:3
a virtual space. FPS games most often present this interaction via an entity carrying and using a
gun, hence the name.
Game genre not only allows us to classify and describe games, but provides a common ground
to contextualize and understand contributions to game design broadly as a research eld [
33
]. This
makes the study of genre and its construction of great importance for the game research community
as a whole.
Construction of genre has been widely discussed in academic literature [
3
,
8
,
20
,
21
,
33
,
38
,
55
],
as well as in commercial and consumer communities. Within the context of commercial game
genre denitions, there is a tendency to rely too heavily on the representational characteristics
(i.e., visual aesthetics) of a game [
8
]. In contrast, game researchers have argued that games and
their genres should also be i) described with respect to their interactivity and how a player non-
trivially progresses through a game [
8
]; ii) described through their game architecture in the form of
high-level design patterns [
54
]; or should be iii) described in terms of a simulation since knowledge
and experience—and by extension the story—is created through player actions [
2
]. With respect
to academic explorations, this more detailed analysis has been done in a number of past works
categorizing and dening games and their genres as a whole [
3
,
27
,
91
]. These game genres have
included idle games [
5
], educational games [
61
,
63
], serious games [
56
], games for rehabilitation [
74
],
games for dementia [
19
,
59
], exertion games [
65
], aective games [
52
], and games and simulations
[48].
There are a number of approaches to the construction of game genres. These include formula-
tion of guidelines [
21
,
33
], development of models [
38
,
83
], and using both quantitative [
55
] and
qualitative methods [
6
] to study existing games and meta-data associated with them. Building on
this related work, we employ a mixed methods approach, utilizing meta-data like Steam tags to
identify what the broader community considers to be a part of the VN genre as suggested by Li and
Zhang [
55
], and utilizing an extensive review of the literature,—similar to [
38
] to identify which
features and design elements existing work used to dene and refer to the genre,. Both of these
techniques are used in tandem to guide the analysis of existing and commercially available games.
2.2 The Visual Novel Genre
Visual Novels (VNs) are a highly popular genre of interactive narrative games. For instance, VNs
have been released on a large range of platforms such as PC [
76
], mobile [
88
], and PS Vita [
69
].
The VN player community has also created VNDB, a community managed database for tagging
and reviewing VNs with over 27,000 entries as of April 18th, 2020. Furthermore, commercial VNs
have also seen recent interest as a subject of study in academic papers. E.g., Carstensdottir et
al. [
15
], examined the structure and narrative progression of various interactive narrative games
including VNs such as Doki Doki Literature Club and Long Live the Queen. Similarly, other popular
commercial VNs such as Hatoful Boyfriend have been the subject of work analyzing how the VN
subverts tropes and conventions of Japanese popular culture [30].
Despite relatively simple core mechanics, VNs have incorporated a variety of themes ranging
from Horror Survival (e.g., Spirit Hunter: Death Mark [
46
]) and Murder Mystery (e.g., Danganronpa
V3: Killing Harmony [
82
]) to Historical Fantasy (e.g., Hakuoki: Edo Blossoms [
42
]) and Romance
(e.g., Amnesia: Memories [
41
]). The increasing popularity of VNs, coupled with the introduction
of free and accessible VN authoring tools, has raised both interest and awareness for the use of
VNs in other applications such as education [
13
]. However, the question of how to dene VNs as a
genre has remained elusive.
The genre’s heavy use of text and emphasis on reading as a central activity has prompted repeated
comparisons to books and seen VNs referred to as their interactive counterparts, such as in [
87
] and
[
10
]. Notably, some literary genres have made a successful transition between the two narrative
Proc. ACM Hum.-Comput. Interact., Vol. 5, No. CHI PLAY, Article 285. Publication date: September 2021.
285:4 Janelynn Camingue, Elin Carstensdoir, & Edward F. Melcer
mediums, with both straight and queer players recognizing romance VNs as “romance novels in
interactive visual form" [
60
]. However, while similar, VNs remain distinct from their non-digital and
non-interactive counterparts for several reasons. First, VNs only show one snippet of text at a time
as the player interacts with the game [
31
], the snippet being dependent on player action or choices
and even shaping future interaction. Further, while VNs share many traits with choose-your-own
adventure books [
10
]—a legacy shared with most modern interactive ction and storytelling design
in video games [
34
,
62
,
64
]—VNs often incorporate more involved mechanical and constraint based
gameplay to impact the progression of the story and the relevance/complexity of player choice
[15].
As a whole, the VN genre has ultimately not been extensively explored and researched [
12
].
While VNs have had a number of denitions in academic and commercial literature, most of these
denitions have focused primarily on the aesthetics of the games [
11
,
16
,
26
,
28
,
29
,
31
,
40
,
49
,
57
,
66
–
68
,
70
,
71
,
75
,
79
], structure and choice [
7
,
10
,
22
,
28
,
29
,
36
,
57
,
67
,
71
,
71
,
80
], and story quality
[
12
,
16
,
49
,
53
,
79
]. Surprisingly, there has been a limited focus on interactivity within existing
VN denitions, primarily referring to interactivity as being minimal or lacking [
4
,
26
,
36
,
40
,
67
].
Notably, there is some disagreement to what extent story quality and other elements like interaction
inuence how VNs should be dened. For example, it has been argued that VNs are "reliant on
the text and the actual content instead of interactivity" [
12
], while others describe certain types of
input performance and interactivity—such as the classic click-to-continue interaction where the
player clicks to receive the next part of story—as being a part of VNs [7,36,40,57].
More directly related to the work presented in this paper, Lynch et al. [
58
] focused on a broader
understanding of VNs where they examined six existing VNs and their feature sets—three of which
were in the top 100 most popular games on VNDB. Although it did provide a wider perspective of
VNs, this work was mainly focused on how a game engine could implement the features found in
their data set and did not fully dene features that were essential to all standard VNs. This work
also did not consider or compare existing denitions of VNs. Instead, the focus was on the problem
of how to make a VN and not what is a VN.
Ultimately, most denitions of VNs as a genre only contain a small number of elements, leaving
out concepts deemed as central to the genre in other denitions. A unied denition would allow
for a more comprehensive view and approach to the study/design of VNs, as well as to the use of
VNs in development of various high impact application areas, such as education and health, where
their accessibility and ease of use has the potential to be leveraged with great eect.
2.3 Serious Applications of Visual Novels
VNs have been used extensively as a tool to address numerous serious topics such as education,
health, and disabilities, to name a few. For instance, VNs have been employed to teach topics such as
second-language acquisition [
25
,
86
], chemistry [
89
,
92
], life management [
49
], research skills [
87
],
and cooking [
44
,
45
]. Recent work has even constructed a taxonomy categorizing how VNs can
deliver educational content [
13
]—describing the dierent teaching strategies currently employed
by educational VNs. With respect to health applications, past work has focused on teaching health
concepts such as nutrition [
77
], explored how an individual VN can improve patient perspectives
and support desired health outcomes [
93
], and addressed complex health related topics such as
disability [17,84].
Another frequent serious application of VNs is exploring narratives and issues faced by under-
represented communities/populations, such as gender inequality and queer narratives [
90
]. For
instance, recent work by Salter et al. [
79
] has examined the design of Buttery Soup and its portrayal
of queer identity and experience. Schaufert [
81
] has similarly analyzed the design of Dream Daddy,
highlighting that it normalizes the "queer daddy gure" yet can create negative feelings as well.
Proc. ACM Hum.-Comput. Interact., Vol. 5, No. CHI PLAY, Article 285. Publication date: September 2021.
What is a Visual Novel? 285:5
Notably, Dream Daddy also became an exceptionally popular and successful VN—where in 2017,
it replaced rst-person shooter Overwatch "as the most-discussed videogame on Tumblr for the
rst time in more than nine months," and it was at the "top of Steam’s global sales chart, unseating
battle-royal phenomenon PlayerUnknown’s Battlegrounds" [
39
]. The growing use of VNs as serious
contexts highlights the potential impact VNs could have beyond entertainment, and underscores
the importance of having a unied denition for VN in order to support research and design for
serious applications.
Fig. 1. An overview of the methodology used to construct a unified definition of VNs.
3 METHODOLOGY
Creating a unied denition of VNs requires identifying their essential core design features. To
do this, we performed a comprehensive search for VN papers and denitions—resulting in the
identication of 30 existing denitions from the literature. These denitions were then analyzed
to identify key theoretical design features underlying VNs. Simultaneously, we also collected a
VN corpus of 54 games in order to analyze playable instances of VNs in comparison with the
theoretical denitions. Each VN was played or at least 10 hours or until nished, whichever came
rst, and common design features that emerged during gameplay but weren’t identied in the
theoretical design features were recorded. The VN corpus was then played through and labeled a
second time utilizing all features from both the theoretical and emergent feature sets. The labeling
using the larger set of design features was then analyzed, resulting in a set of core design features
that occurred in 95% or more of the VN games from the corpus. See Figure 1for a visual overview
of the process. These core features were then used to construct a unied denition of VNs.
For feature labeling, we employed a single researcher approach. This was done to keep the data
consistent across the study by avoiding the creation data inconsistencies caused by individual
dierences between researchers. Importantly, this approach has been employed successfully before
in prior game research—specically for analyzing narrative games [
15
] and educational visual
novels [
13
]—and is therefore a reasonable method to employ for this kind of analytic research.
However, in order to mitigate individual researcher bias, the researcher that performed feature
labeling also met weekly with two additional senior researchers to present, discuss, rene, and
nalize all identied features and labeling.
Proc. ACM Hum.-Comput. Interact., Vol. 5, No. CHI PLAY, Article 285. Publication date: September 2021.
285:6 Janelynn Camingue, Elin Carstensdoir, & Edward F. Melcer
3.1 Academic Definitions
Collecting denitions
. First, the researchers compiled academic papers that addressed the topic
of VNs. These papers were identied through a systematic review of proceedings of recognized
venues for technical games research, game studies, and interactive narrative—specically FDG,
DiGRA, ICIDS, CHIPLAY, Game Studies, and Transactions on Games. This was then augmented
with a search through Google Scholar. The keyword "visual novel" was used across all searches,
and papers that did not provide an explicit denition of a VN were excluded. All papers selected
for the analysis were required to contain an explicit VN denition. This requirement was included
to prevent accidental researcher bias from interpreting and/or extracting denitions from papers
that didn’t explicitly provide one. This process resulted in 30 papers, each containing a denition
of a VN. Afterwards, a table was constructed for analysis. The rows consisted of the 30 denitions
collected, and the columns consisted of all VN features identied in the various denitions—these
features are explained in more detail in section 4.1. In one specic case, for the concept of level of
interactivity, the researcher recorded the varying degrees of interactivity that were detailed in each
denition—ranging from only pressing a button to advance the story (no interactivity) to providing
choices that impact the story (moderate interactivity) or minigames and puzzles to complete (high
interactivity). A rst pass was done to nd commonalities and dierences between the general
features that each denition covered. The result was then used to inform what specic features to
identify on the rst play through of the games.
To further reduce bias, the researcher performed a second pass, where they performed detailed
review of each denition with two additional researchers. Another table was constructed to describe
exact keywords for the identied VN features. One by one, the researchers reread each denition,
highlighted the key words within that denition, and if that keyword was not already in the table,
the keyword was added as a new column. Otherwise, if the keyword already existed within the
table or a very close synonym existed, the researcher marked down that this denition contained
this existing keyword. Notably, this approach helps to compare the explicit wordings of denitions
without interpretation by the researchers. These keywords were also ultimately used to construct
the nal unied denition.
3.2 VN Search Strategy
In order to identify which features from academic denitions were also commonly found in VNs
(and ultimately aid in constructing a more grounded unied denition), the researchers constructed
a corpus of games. Notably, this approach of collecting and analyzing of a large corpus of related
games has also previously been employed in other genre dening work, e.g., [
5
,
38
,
55
]. The
researchers began their search by formulating a set of three selection criteria. First, the game is
tagged as a VN on its respective distribution platform. These tags are usually assigned by publishers,
developers, and in the case of Steam by players. Games that fulll this criterion are representative
of the industry and players themselves identify as VNs. The second criterion was that the game’s
published description, or an academic paper, explicitly identies a particular game as a VN. This
criterion helps lter for games that developers or academics consider VNs. The third criterion was
that the game must be documented in VNDB, meaning they are considered VNs by the player
community. The researchers collected the initial set of VNs from previous academic literature that
explicitly identied VNs. These games fullled the second selection criterion. The researchers then
collected an additional sample based on critical acclaim and popularity within the VN community.
The researchers then looked for VNs that fullled one of the three selection criteria on three
dierent distribution platforms. These platforms were Steam, Itch.io, and the Google Play Store.
The researchers chose these three as they are currently the largest distribution platforms for PC
Proc. ACM Hum.-Comput. Interact., Vol. 5, No. CHI PLAY, Article 285. Publication date: September 2021.
What is a Visual Novel? 285:7
games (Steam), indie games (itch.io), and Android mobile games (Google Play Store). The researchers
collected 54 games in total. This provides a substantial sample and additional games would yield
diminishing returns for this study. 19 of the games within the nal corpus were educational.
3.3 Collecting VN Data
A single researcher played all 54 VNs. Each VN was played for at least 10 hours or until completion
if that occurred rst. The researcher, to the best of their ability, attempted to get as many endings as
possible if the game indicated that it had multiple endings. Additionally, cheats and walkthroughs
were used when available in order to explore and uncover as much of the content and design of
the VNs as possible during a play session. Furthermore, the researcher also created an illustrative
mapping of the story tree as they played the game in order to help organize understanding of
the structure of the story. The researchers chose to play the VNs instead of watching footage
since gameplay is vastly dierent than watching a playthrough, providing more authentic results
than video. This is also the typical approach in existing game research which generally relies on
gameplay rst and uses videos as secondary sources when necessary [6].
When playing the researcher identied and took note of the design features of each VN. A feature
in this context is an audio or visual element of a game (there were no kinematic elements—i.e.,
touch feedback and motion based interactions—for any of the games in our corpus). For analysis,
the researchers constructed a table of features identied from the earlier academic denitions in
order to record how they mapped to the design features of each VN. The features extracted into
the table were directly mention by each academic paper’s denition. Each column of this table
represented a specic feature. In this way, it was possible to identify how well the games t into
existing scholarly denitions, and a game and an academic denition could be compared visually
by seeing which columns were empty or lled out.
After playing each game, the researcher recorded a "1" for that feature if it was present and "0" if
it was not present. Additional features were also added to the table if an appropriate match did
not exist. Specically, four features were generated through playing games in the VN corpus, i.e.,
linear story structure,includes traditional text box,includes non-traditional text box, and on-click
progression for text box. On a second pass the researchers replayed games and grouped together
similar keywords/features to form nal unique design feature categories, i.e., axial coding [
78
]. In
total, there were 22 unique design features identied.
4 RESULTS
4.1 Comparing Definitions of Visual Novel
The data set of academic denitions of VNs consisted of 30 denitions, and collectively converged
on a set of four design categories: Art and Aesthetic, Narrative Structure, Interaction, and Story
Quality. Each VN denition incorporated one or more design feature from these categories, and
6.19 features on average out of the 22 unique features identied through analysis. See Figure 4for
each design feature and its frequency of occurrence in the academic denitions. Also see Figure 2
for the relationship of each design feature to the four design categories.
Art and Aesthetic
refers to elements like imagery, graphics, and art style. Aesthetic elements
are identied in 14 VN denitions and surrounding discussion, e.g.,[
4
,
22
,
26
,
28
,
29
,
40
,
47
,
49
,
57
,
68
,
71
,
75
,
77
]. The denitions varied in the way images were dened or described. For example,
some described static images or illustrations [
29
]. Other denitions specify the use of images as
backgrounds [
31
,
68
] and character art [
4
,
28
,
31
,
71
,
77
]. Several denitions specically identify the
Japanese inuence on both the art and general aesthetics of VNs, with 4 denitions stating that VNs
have anime-inspired aesthetics [
28
,
29
,
67
,
71
]. Specically, Faizal [
29
] describes VNs as containing
Proc. ACM Hum.-Comput. Interact., Vol. 5, No. CHI PLAY, Article 285. Publication date: September 2021.
285:8 Janelynn Camingue, Elin Carstensdoir, & Edward F. Melcer
Fig. 2. All 22 features and their relationship within the 4 larger design categories identified in academic
definitions.
mostly static images commonly represented in "anime cartoon" style, and Navarro-Remesal and
Longuillo-Lopex’s denition [
67
] similarly argues that VNs "derive from manga". Emi and Okuda
[
28
] describe VNs as featuring "static graphics and anime-styled characters". While Pratama et al.
[
71
] specify that VNs "have static backgrounds, dynamic characters that are generally manga or
anime-styled and sound eects". However, while VNs did originate in Japan, they are now widely
developed outside of it and convey stories outside of its cultural context, e.g. [
32
,
73
]. Notably,
a total of 9 denitions do not specify the art style of VNs [
4
,
26
,
40
,
47
,
49
,
57
,
68
,
75
,
77
], likely
reecting this shift.
Narrative Structure
refers to structure pattern types and structural elements, and are promi-
nently featured in 14 denitions. More specically, narrative structure is featured in the form
of specic structure patterns such as branching structures [
10
,
22
,
36
,
57
,
67
,
71
], and structural
elements like endings and choices [
7
,
28
,
29
,
71
,
80
]. Furthermore, some denitions specied that
VNs included multiple endings as a result of player choice [
28
,
29
,
57
,
80
], which can indicate the
presence of a branching structure to a player. This suggests that branching structures were included
implicitly in more denitions than explicitly stated.
Interaction
within VNs was incorporated in a variety of ways for almost all of the VN denitions
(23 out of 30). Notably, some denitions specied that VNs have limited interactivity [
4
,
36
,
40
,
67
]
or minimal gameplay [
26
]. However, denitions did not agree on how limited interaction was
dened or quantied, with descriptions of limited interaction varying from "limited" to "making
Proc. ACM Hum.-Comput. Interact., Vol. 5, No. CHI PLAY, Article 285. Publication date: September 2021.
What is a Visual Novel? 285:9
decisions that aect the progress of the game" [
4
,
12
,
40
,
47
,
49
,
57
,
67
]. In terms of design elements,
choices were frequently used to describe how the player might be able to mechanically interact with
and impact the narrative [
4
,
7
,
10
,
12
,
28
,
29
,
40
,
49
,
50
,
57
,
67
,
75
,
80
]. More specic descriptions
of interaction within the denitions identied the click-to-continue interaction, where the player
clicks to receive the next part of story [
7
,
36
,
40
,
57
]. At least one denition identied the dialog box
specically [
71
], which is where story content is often delivered via click-to-continue interaction.
Interestingly, more complex interaction elements such as mini-games and puzzle gameplay, as seen
in well-known and successful VN series like Danganronpa and Phoenix Wright, were not widely
incorporated—with only one denition including RPG elements like puzzles and minigames [68].
Story Quality
was identied as another signicant part of VN design, with 8 denitions specify-
ing the quality of the story itself and its elements as being an integral part of VNs. For instance, VNs
have been described as "reliant on the text and the actual content instead of interactivity" and that
characters play a large role in that content [
12
]. This is echoed by Salter et al. [
79
], stating that VNs
are inuenced by character-centric forms of storytelling and rely heavily on dialog. Moving beyond
discussions of importance, there were also several notable story quality elements identied by the
denitions. Cavallaro [
16
] highlights "plot depth and subtle characterization" as key elements of
VNs while Korhonen and Halonen [
49
] list immersive storytelling. Romance, a popular literary
genre of VNs 6, was also included in 2 denitions that focused on romance and dating simulation
elements [60,79].
4.2 Interactivity in Visual Novels
Although almost all academic denitions of VNs incorporated interactivity in some form, the wide
variety of ways it was employed warranted further exploration. For our study the denition of
interactivity was any aorded, provided, or forced action needed to progress within the game.
An interactive action, in the context of our investigation has to have some meaningful cognitive
load or decision making. We measured the level of interactivity based on perceived cognitive
load, frequency, and physical activity of an interaction. In particular, the focus on low amounts
of interactivity—and subsequent lack of more complex interaction elements such as puzzles or
minigames—could be examined by measuring the frequency of interaction for each game within
our game corpus (see Figure 3).
In order to determine the frequency of interaction, we set a timer for thirty minutes and replayed
each of the 54 VNs. We chose the thirty minute sampling as VNs in our corpus rarely had extensive
or elaborate tutorials and generally established their core loop and rhythm of gameplay early.
Before deciding on the sample method, each VN was evaluated in terms of whether the rst thirty
minutes were subjectively similar to their overall gameplay by the researcher who played each
VN extensively. On average, the rst thirty minutes were concluded to be representative of the
VNs overall gameplay and thus deemed to be sucient to indicate overall gameplay interaction
frequency.
As a VN was being replayed, we noted down how many interactions, such as menu choices, were
provided within the given time. Afterwards, the total number of interactions per game was divided
by 30 to calculate the frequency per minute. These values were then normalized to a range between
0 and 3 by dividing by the max frequency of interactions (i.e., Pastry Lovers at 2.36 interactions on
average per minute) and multiplying by 3.
Surprisingly, most VNs had less than one interaction per minute, with an average of only 0.37
interactions per minute. Furthermore, the only exceptions to this extremely limited interaction
6
Romance is tagged as the fth-most popular sub-genre of visual novels in VNStat [
1
], a community-ran site that identies
trends on the Visual Novel Data Base (VNDB).
Proc. ACM Hum.-Comput. Interact., Vol. 5, No. CHI PLAY, Article 285. Publication date: September 2021.
285:10 Janelynn Camingue, Elin Carstensdoir, & Edward F. Melcer
Fig. 3. An overview of the frequency of interactions per minute within the first 30 minutes of gameplay for
each of the VNs. Values were normalized to fit the range 0 to 3.
frequency still had less than three total interactions per minute. Overall, the ranges for number of
interactions per game varied quite substantially. A total of 9 games had 0 interactions, due to a
heavy focus on narrative, whereas stat-building VNs like Long Live the Queen and Pastry Lovers
had 51 and 71 interactions respectively.
Proc. ACM Hum.-Comput. Interact., Vol. 5, No. CHI PLAY, Article 285. Publication date: September 2021.
What is a Visual Novel? 285:11
4.3 Craing a Unified Definition
In order to craft a unied denition for VNs, we rst determined the set of essential core features
that would be incorporated into the nal denition. Breakdown per feature can be see in Figure
5. A core feature was identied as applying to at least 95% of the VNs included in the corpus. We
chose 95% as the cuto point since it gave us a high degree of condence that the corresponding
design feature was highly prevalent in VNs beyond just our corpus. We identied 10 core features
of VNs: Narrative-focused, Interactivity Required, Background Art, Static Characters, SFX, Music,
Includes Traditional Text Box, On-Click Progression, Action Menu Choices, and Dialogue Menu
Choices.
Using these core factors, and analysis of the VN corpus, we propose the following new denition
of a VN:
A Visual Novel (VN) is a digital narrative focused game that requires interactions where the
player must be able to impact the story world or the story’s progression. The story and interactions
are most commonly presented through a text box and often employ additional forms of interac-
tion including menu choices—which often contain sets of actions that the player character can
perform—or dialogue options representing the player character’s speech or thoughts. Crucially,
VNs have On-Click Progression, where the player clicks, taps or presses a button to see the next
part of the story. The aesthetics of VNs are most often conveyed through static images of characters,
background art, sound eects (SFX) feedback, and soundtracks.
Previously existing academic denitions of VNs included an average of 2.17 explicit mentions of
the 10 core features. The most comprehensive existing VN denition, by Lu [
57
], incorporated the
highest number of core features with 7 out of 10 features included. Overall, this demonstrates that
our unied denition of VNs is far more comprehensive than existing academic denitions, and
provides a number of clear, identiable features for the design of VNs. This denition also accounts
for visual novels in contexts beyond entertainment, such as education, and holds true for the 19
educational visual novels within our corpus.
5 DISCUSSION
5.1 Design Implications
Our ndings show that there is some mismatch in terms of emphasis between academic denitions
of VNs and existing work categorized as VNs. Some academic denitions in our sample specied
that VNs are interactive but overall focused more on the quantity of interaction rather than the
design features enabling it.
VNs have an inherent set of interaction aordances that manifest as a result of their use of
on-click progression of story content presented via text boxes and menu choice selection. While the
majority of academic denitions in our sample did not include features like text boxes and on-click
progression, these features were observed in all the games in our corpus. Likewise, 53 out of 54
games had menu choices presented with an onscreen menu, but these were not present in the vast
majority of denitions collected for this study. The notable exception here is Burn Your Fat With
Me!! [
43
] which did not have any choices presented to the player but did ask them to follow along
with exercises presented in the game in the real world. These interaction aordances make up core
design patterns or game architecture [
8
] of VNs. Use of these patterns allows players to "experience
Proc. ACM Hum.-Comput. Interact., Vol. 5, No. CHI PLAY, Article 285. Publication date: September 2021.
285:12 Janelynn Camingue, Elin Carstensdoir, & Edward F. Melcer
Fig. 4. All 22 features mapped by how many academic definitions they appeared in from the 30 collected
academic definitions.
Fig. 5. All 54 games mapped by feature, across all 22 features identified in the study. Core features were
identified as being present in 95% of the games in the corpus, represented by a cut-o line in the figure. The
core features are identified by a doed paern in the figure. All values were scaled by a factor so that all 54
games were visible.
variety while remaining within the scope of a familiar and appreciated game experience for which
they are already competent" [
8
]. In other words implementing the features players expect to be
present for the genre, such as the core features identied in our study, improves the usability of
the game and aligns with a player’s game literacy. Not including the core features might confuse
players and violate their expectations.
Proc. ACM Hum.-Comput. Interact., Vol. 5, No. CHI PLAY, Article 285. Publication date: September 2021.
What is a Visual Novel? 285:13
The aordances provided by VNs combined with limited interaction frequency supports the
genre’s ability to deliver educational content. On average VNs within the corpus had 0.37 inter-
actions per minute within the rst thirty minutes of game play. Recall that levels of interactivity
are measured based on perceived cognitive load. Common interactions in VNs, such as click to
continue progression, are relatively simple and require limited input from the player [
13
]. This
reduces the need for players to spend excessive cognitive load to play the game. Players can instead
cognitively engage in understanding material instead of engaging in learning new mechanics.
Story structures, like branching and linear structure patterns, are often used to describe the
nature of the inuence the player has when interacting with a story. Branching structures allow
for more variety in experience than linear structures, by allowing the player more opportunities
to inuence and shape the story and its outcomes through branching choices. Notably, 9 of 30
denitions in our sample included a requirement that VNs contain branching stories. However,
while branching structure was heavily utilized by VNs in our corpus, linear structures were also
notably prevalent, where 10 out of 54 or about 18% of the corpus did not include a branching
story structure. Most notably popular games such as Phoenix Wright: Ace Attorney Trilogy [
14
]
and Danganronpa V3: Killing Harmony [82] had a perceived linear story structure. Therefore the
denition of VN presented by this paper does not include the restriction of specifying a particular
structure type for VNs, instead leaving that design decision exible to the designer and goals of the
VN.
5.2 Improving the Design of Educational VNs
One benet of having a unied denition for VNs is that it more clearly outlines the potential
design space for such games. Notably, this also allows us to examine designs within important
subgenres such as educational VNs. Of the 54 VNs from the corpus, 19 were focused on educational
applications. In this subsection, we analyze these 19 educational VNs with respect to our unied
denition as a test—highlighting the ecacy of our denition for identifying underexplored design
features and potentially improving the design of future educational VNs.
These VNs shared many common design decisions with their entertainment focused counterparts
including personalization options for the player (e.g., choosing their gender or changing the player
character’s name) [
25
,
45
,
45
,
89
,
92
], having romantic love interests [
43
–
45
,
72
,
84
,
86
,
89
,
92
],
minigames [
25
,
37
,
43
,
72
], and branching storylines with multiple endings [
44
,
45
,
84
,
86
,
89
,
92
].
However, there were some notable dierences between the two where educational VNs often
utilized less of the design space within the VN genre—resulting in a number of features that could
potentially improve their designs and learning outcomes.
For instance, none of the educational VNs employed any story-based customization—i.e., choices
that alter the player’s character and aect the story. However, this is seen commonly in enter-
tainment VNs that use stats such as Dandelion - Wishes Brought To You [
85
] and Monster Prom
[
32
]. In both of these non-educational games, the player chooses activities for the protagonist
which increase or decrease certain stats. These stats then alter the route of the story and possible
choices presented to the player. In contrast, educational games relied more on menu choices as the
main form of interacting with the story. While menu choices are simple, clear, and a direct way
to communicate with players, story-based customizations have the potential to better represent
eects of long term decision making. For example, in Dandelion - Wishes Brought To You every
activity has a trade o. If the player chooses to do homework their character’s art skills increase
but their character’s stress also increases. This models the idea of a work-life balance, since the
player will reach a game over state if they don’t carefully manage all stats, and presents it in a more
intuitive form than possible with standard menu choices. This type of day planning could possibly
Proc. ACM Hum.-Comput. Interact., Vol. 5, No. CHI PLAY, Article 285. Publication date: September 2021.
285:14 Janelynn Camingue, Elin Carstensdoir, & Edward F. Melcer
be used to enhance educational games with stats focused topics such as nutrition, infection rates,
or ecosystems to name a few potential examples.
Similarly, none of the educational VNs featured player-centric interactions—i.e., interactions
that assist the player but ultimately do not aect the story—while a number of the entertainment
focused VNs did. For instance, Hakuoki: Edo Blossoms[
42
] and Valentyne Stories Necromancy [
9
],
had a glossary or journal that included important key words for the player to reference. It served as
a way to explain and give context pertaining to the story for the player. Educational visual novels
could benet from implementing similar features. Just as the glossary provides context in games
such as Hakuoki: Edo Blossoms, a glossary in an educational VN could remind players of previously
taught concepts. For example, the game After-Party Chemistry[
92
] assumes that the player already
knows basic chemistry and attempts to add on to that knowledge. A glossary in this case would
help the player understand and remember new terminology or recall terminology they may have
forgotten.
6 LIMITATIONS AND FUTURE WORK
When identifying features from the denitions from academic papers, three keywords were judged
either too broad or not suciently dened to be applicable during the analysis process, and thus were
excluded. These three keywords were "limited to reading," "complex stories," and "minimal gameplay".
While broad they represent constructs relating to both interactivity and content complexity. The
former was at least partially incorporated via analysis of frequency of interaction, while content
complexity of the story itself was deemed outside the scope of the work.
It is worth noting that while a signicantly sized corpus, 54 games might not be fully representa-
tive of the entire genre. Future work should aim to improve methods of constructing representative
sets of games for genre construction and games research more broadly, so that the eld can more
eectively study and contextualize its ndings.
In addition, certain features identied in the study warrant further exploration. One example is
story structure. Story structure, in the context of this work, was described in terms of how it was
perceived by a single researcher. Having more than one researcher play and document the games
would introduce variation in descriptions that could be of interest. Further, there are more structure
types and patterns that have been identied in the literature than strictly branching and linear
story structures, such as fold-back and hidden story structures [
15
], which might have further
impact on VN experiences.
7 CONCLUSION
The Visual Novel (VN) is a widely popular and recognizable narrative-focused game genre whose
popularity has been steadily increasing in recent years. Their popularity has been bolstered even
further with the increase of free VN game engines, such as Ren’Py [
18
], which have made VN
creation more accessible to novice designers while allowing expert designers to construct complex
and engaging work. In addition, the accessibility of VNs to a variety of players and skill levels has
led to their use in high-impact domains like education [
4
,
13
,
23
,
26
,
29
,
31
,
35
,
40
,
49
,
51
,
68
,
87
]
and health [
77
,
93
]. Accessibility for players is key in such domains to reach as many individuals as
possible, and its popularity and narrative focus make it a highly suitable choice of gameplay for a
variety of domains. Thus, the value of studying and improving VN design within HCI and Game
Studies is signicant. While existing academic work on VNs has provided a variety of denitions
of VNs, these denitions can vary wildly—which in turn has substantial implications for the design
of such games. Creating a unied denition of VNs as a genre and understanding the key elements
of a VN’s design allows us to better inform design and research of VNs, not just for entertainment
but also their application in high impact domains such as education and health. In this paper, we
Proc. ACM Hum.-Comput. Interact., Vol. 5, No. CHI PLAY, Article 285. Publication date: September 2021.
What is a Visual Novel? 285:15
propose a unied denition of VNs developed from analyzing existing denitions of VNs as a
genre and a large corpus of commercial and educational VNs. We discussed key features from our
unied denition and their implications on the design of VNs for both entertainment and serious
applications.
REFERENCES
[1] [n.d.]. Romance. https://vnstat.net/tag/96
[2] Espen Aarseth. 2004. Genre trouble. Electronic book review 3 (2004), 1–7.
[3]
Espen Aarseth, Solveig Smedstad, and Lise Sunnanå. 2003. A multidimensional typology of games.. In In Proceedings
of the 2003 DiGRA International Conference: Level Up. DiGRA, Utrecht, The Netherlands. http://www.digra.org/wp-
content/uploads/digital-library/05163.52481.pdf
[4]
Dr. Imam Dui Agusalim. 2012. Developing Interactive E-Learning Module of English Teaching to Support the
Distance Education Program at EEPIS. IOSR Journal of Humanities and Social Science 5, 1 (2012), 28–32. https:
//doi.org/10.9790/0837-0512832
[5]
Sultan Alharthi, Olaa Alsaedi, Zachary Toups, Theresa Tanenbaum, and Jessica Hammer. 2018. Playing to Wait: A
Taxonomy of Idle Games. 1–15. https://doi.org/10.1145/3173574.3174195
[6]
Sultan A. Alharthi, Olaa Alsaedi, Zachary O. Toups, Theresa Jean Tanenbaum, and Jessica Hammer. 2018. Playing to
Wait. Proceedings of the 2018 CHI Conference on Human Factors in Computing Systems (2018). https://doi.org/10.1145/
3173574.3174195
[7]
J Andrew, S Henry, A.N. Yudhisthira, Yulyani Arin, and Syarifah Diana Permai. 2019. Analyzing the Factors that
Inuence Learning Experience through Game Based Learning using Visual Novel Game for Learning Pancasila. Procedia
Computer Science 157 (2019), 353 – 359. https://doi.org/10.1016/j.procs.2019.08.177 The 4th International Conference
on Computer Science and Computational Intelligence (ICCSCI 2019) : Enabling Collaboration to Escalate Impact of
Research Results for Society.
[8]
Thomas Apperley. 2006. Genre and game studies: Toward a critical approach to video game genres. Simulation &
Gaming - Simulat Gaming 37 (03 2006), 6–23. https://doi.org/10.1177/1046878105282278
[9]
Mesange Arts. 2019. Valentyne Stories Necromancy. Game [Windows, macOS, Linux, Android]. Mesange Arts.
In-development. Distributed on Itch.io. Last played March 2020.
[10] Hiroki Azuma. 2009. Otaku: Japan’s database animals. U of Minnesota Press.
[11] Katerina Bashova and Veno Pachovski. [n.d.]. Poetry as a Visual Novel–a Multimedia Project. ([n. d.]).
[12]
Luca Bruno. 2017. Interpreting/subverting the database: Character-conveyed narrative in Japanese visual novel PC
games. Mutual Images Journal 3 (2017). https://doi.org/10.32926/2017.3.bru.inter
[13]
Janelynn Camingue, Edward F. Melcer, and Elin Carstensdottir. 2020. A (Visual) Novel Route to Learning: A Taxonomy
of Teaching Strategies in Visual Novels. In International Conference on the Foundations of Digital Games (Bugibba,
Malta) (FDG ’20). Association for Computing Machinery, New York, NY, USA, Article 77, 13 pages. https://doi.org/10.
1145/3402942.3403004
[14]
Capcom. 2019. Phoenix Wright: Ace Attorney Trilogy. Game [PC]. Capcom, Kyoto, Japan. Distributed on Steam. Last
played March 2020.
[15]
Elin Carstensdottir, Erica Kleinman, and Magy Seif El-Nasr. 2019. Player interaction in narrative games. Proceedings of
the 14th International Conference on the Foundations of Digital Games - FDG 19 (2019). https://doi.org/10.1145/3337722.
3337730
[16]
Dani Cavallaro. 2009. Anime and the visual novel: narrative structure, design and play at the crossroads of animation and
computer games. McFarland.
[17] Alexander Champlin. [n.d.]. Playing with Feelings. ([n. d.]).
[18] Robert Ciesla. 2019. Game Development with Ren’Py. Springer.
[19]
Dormann Claire. 2016. Toward Ludic Gerontechnology: a Review of Games for Dementia Care.. In 1st International
Joint Conference of DiGRA and FDG (Dundee, Scotland) (DiGRA/FDG ’16).https://doi.org/10.1145/3311350.3347173
[20]
Rachel Clarke, Jin Lee, and Neils Clark. 2015. Why Video Game Genres Fail: A Classicatory Analysis. Games and
Culture 12 (07 2015). https://doi.org/10.1177/1555412015591900
[21]
David A. Clearwater. 2011. What Denes Video Game Genre? Thinking about Genre Study after the Great Divide.
Loading... 5, 8.
[22]
Rebecca Crawford and Yuanyuan Chen. 2017. From hypertext to hyperdimension Neptunia: The future of VR visual
novels: The potentials of new technologies for branching-path narrative games. 2017 23rd International Conference on
Virtual System & Multimedia (VSMM) (2017). https://doi.org/10.1109/vsmm.2017.8346298
[23]
John Dominic S Diez and Edward F Melcer. 2020. Cookie Mania: A Serious Game for Teaching Internet Cookies to
High School and College Students. In Joint International Conference on Serious Games. Springer, 69–77.
Proc. ACM Hum.-Comput. Interact., Vol. 5, No. CHI PLAY, Article 285. Publication date: September 2021.
285:16 Janelynn Camingue, Elin Carstensdoir, & Edward F. Melcer
[24]
Damien Djaouti, Julian Alvarez, and Jean-Pierre Jessel. 2011. Classifying Serious Games: the G/P/S model. Handbook
of Research on Improving Learning and Motivation through Educational Games: Multidisciplinary Approaches (01 2011).
https://doi.org/10.4018/978-1- 60960-495-0.ch006
[25]
Sleepy Duck. 2018. Learn Japanese To Survive! Kanji Combat. Game [PC]. RIVER CROW ST UDIO. Distributed on
Steam. Last played March 2020.
[26]
Nour El Mawas, Michael Bradford, Josephine Andrews, Pramod Pathak, and Cristina Hava Muntean. 2018. A Case
study on 21st century skills development through a computer based maths game. In EdMedia+ Innovate Learning.
Association for the Advancement of Computing in Education (AACE), 1160–1169.
[27]
Christian Elverdam and Espen Aarseth. 2007. Game Classication and Game Design: Construction Through Critical
Analysis. Games and Culture 2, 1 (2007). https://doi.org/10.1177/1555412006286892
[28]
Keiji Emi and Shigeto Okuda. 2016. EPUB-type visual novel gives us happy math: A prototype of edutainment software
on math. 2016 IEEE 5th Global Conference on Consumer Electronics (2016). https://doi.org/10.1109/gcce.2016.7800538
[29]
M. Aliv Faizal. 2016. The eects of conversation-gambits visual-novel game on students english achievement and
motivation. 2016 International Electronics Symposium (IES) (2016). https://doi.org/10.1109/elecsym.2016.7861054
[30] Emily Flynn-Jones. 2015. Well Played–Vol. 3, No. 2 (Seriously Weird Special Issue). (2015).
[31]
Pedro Gabriel, Tsukasa Hirashima, and Hayashi Yusuke. 2018. A Serious Game for Improving Inferencing in the
Presence of Foreign Language Unknown Words. International Journal of Advanced Computer Science and Applications
9, 2 (2018). https://doi.org/10.14569/ijacsa.2018.090202
[32]
Beautiful Glitch. 2018. Monster Prom. Game [PC]. Those Awesome Guys. Distributed on Steam. Last played March
2020.
[33] W. Goddard and A. Muscat. 2017. Towards Genre as a Game Design Research Approach. In DiGRA Conference.
[34]
Katelyn M. Grasse, Edward F. Melcer, Max Kreminski, Nick Junius, James Ryan, and Noah Wardrip-Fruin. 2021.
Academical: A Choice-Based Interactive Storytelling Game for Enhancing Moral Reasoning, Knowledge, and Attitudes
in Responsible Conduct of Research. In Games and Narrative: Theory and Practice. Springer International Publishing.
[35]
Katelyn M. Grasse, Edward F. Melcer, Max Kreminski, Nick Junius, and Noah Wardrip-Fruin. 2021. Improving
Undergraduate Attitudes Towards Responsible Conduct of Research Through an Interactive Storytelling Game. In
Extended Abstracts of the 2021 CHI Conference on Human Factors in Computing Systems. Association for Computing
Machinery, New York, NY, USA, Article 265, 8 pages. https://doi.org/10.1145/3411763.3451722
[36]
Rizky Yuniar Hakkun, Dio Al Sabah A Z, Kholid Fathoni, Nana Ramadijanti, and Achmad Basuki. 2018. Online Visual
Novel Game Framework. 2018 International Conference on Information and Communication Technology Convergence
(ICTC) (2018). https://doi.org/10.1109/ictc.2018.8539388
[37] Angela He. 2017. Grey Plague. Game [PC]. Angela He, Distributed on itch.io. Last played March 2020.
[38]
Stephanie Heintz and Ee Lai-Chong Law. 2015. The Game Genre Map: A Revised Game Classication (CHI PLAY
’15). 175–184. https://doi.org/10.1145/2793107.2793123
[39]
Laura Hudson. [n.d.]. Gay Dating Simulator ’Dream Daddy’ Might Just Be the Gaming Miracle of the Year. https:
//www.wired.com/story/dream-daddy-game-review/
[40]
Duy Huynh, Phuc Luong, Hiroyuki Iida, and Razvan Beuran. 2017. Design and Evaluation of a Cybersecurity Awareness
Training Game. In Entertainment Computing – ICEC 2017, Nagisa Munekata, Itsuki Kunita, and Junichi Hoshino (Eds.).
Springer International Publishing, Cham, 183–188.
[41]
Design Factory Idea Factory. 2015. Amnesia: Memories. Game [PC, PS Vita]. Idea Factory International. Distributed on
Steam and PlayStation Store. Last played March 2020.
[42]
Design Factory Idea Factory. 2018. Hakuoki: Edo Blossoms. Game [PC, PS Vita]. Idea Factory International. Distributed
on Steam and PlayStation Store. Last played March 2020.
[43] Creative Freaks Inc. 2016. Burn Your Fat With Me!! Game [Android]. Creative Freaks Inc. Last played March 2020.
[44]
Creative Freaks Inc. 2016. Gochi-Show! for Girls -How To Learn Japanese Cooking Game. Game [PC]. Creative Freaks
Inc. Distributed on Steam. Last played March 2020.
[45]
Creative Freaks Inc. 2016. Gochi-Show! How to Learn Japanese Cooking. Game [PC]. Creative Freaks Inc. Distributed
on Steam. Last played March 2020.
[46]
Experience Inc. 2019. Spirit Hunter: Death Mark. Game [PC]. Aksys Games. Distributed on Steam. Last played March
2020.
[47]
Jaana Kemppainen, Tanja Korhonen, and Teija Ravelin. 2014. Developing Health Games requires multidisciplinary
expertise.
[48]
Jan Klabbers. 2003. The gaming landscape: A taxonomy for classifying games and simulations. In Digital Games
Research Conference 2003.
[49]
Tanja Korhonen and Raija Halonen. 2017. On the Development of Serious Games in the Health Sector - A Case Study of
a Serious Game Tool to Improve Life Management Skills in the Young. Proceedings of the 19th International Conference
on Enterprise Information Systems (2017). https://doi.org/10.5220/0006331001350142
Proc. ACM Hum.-Comput. Interact., Vol. 5, No. CHI PLAY, Article 285. Publication date: September 2021.
What is a Visual Novel? 285:17
[50]
Tanja Korhonen, Raija Halonen, Teija Ravelin, Jaana Kemppainen, and Kyösti Koskela. 2017. A Multidisciplinary
Approach To Serious Game Development In The Health Sector.. In MCIS. 30.
[51]
Tanja Korhonen, Teija Ravelin, and Raija Halonen. 2019. Development of a serious game as a method to support youth
work: A case study. In GamiFIN.
[52]
Raul Lara-Cabrera and David Camacho. 2018. A taxonomy and state of the art revision on Aective Games. Future
Generation Computer Systems (01 2018). https://doi.org/10.1016/j.future.2017.12.056
[53]
Richard Lee and Gabriel Chia. 2015. Surfacing Intuitions Through Visual Novels. In Disciplinary Intuitions and the
Design of Learning Environments. Springer, 145–153.
[54]
Jonathan Lessard. 2014. Game genres and high-level design pattern formations. In Proceedings of the 2014 Foundations
of Digital Games Conference. Florida.
[55]
Xiaozhou Li and Boyang Zhang. 2020. A Preliminary Network Analysis on Steam Game Tags: Another Way of
Understanding Game Genres. In Proceedings of the 23rd International Conference on Academic Mindtrek. 65–73. https:
//doi.org/10.1145/3377290.3377300
[56]
Rafael P. De Lope and Nuria Medina-Medina. 2016. A comprehensive taxonomy for serious games. Journal of
Educational Computing Research 55, 5 (2016), 629–672. https://doi.org/10.1177/0735633116681301
[57]
Brian Lu. 2014. Hikikomori: The Need to Belong and the Activation of Narrative Collective-Assimilation through
Visual Novels. Journal of Interpersonal Relations, Intergroup Relations and Identity 7 (2014), 50–61.
[58]
Samuel Lynch, Karsten Pedersen, Fred Charles, and Charlie Hargood. 2019. M22 - A Modern Visual Novel Framework.
In Proceedings of the 8th International Workshop on Narrative and Hypertext (Hof, Germany) (NHT ’19). Association for
Computing Machinery, New York, NY, USA, 9–13. https://doi.org/10.1145/3345511.3349284
[59]
Simon Mccallum and Costas Boletsis. 2013. Games for Health. Chapter A Taxonomy of Serious Games for Dementia,
219–232. https://doi.org/10.1007/978-3- 658-02897-8_17
[60]
Heidi McDonald. 2015. Romance in Games: What It Is, How It Is, and How Developers Can Improve It. QED: A Journal
in GLBTQ Worldmaking 2, 2 (2015), 32–63. http://www.jstor.org/stable/10.14321/qed.2.2.0032
[61]
Edward Melcer and Katherine Isbister. 2016. Bridging the Physical Divide: A Design Framework for Embodied Learning
Games and Simulations. In Proceedings of the 2016 CHI Conference Extended Abstracts on Human Factors in Computing
Systems (CHI ’16).https://doi.org/10.1145/2851581.2892455
[62]
Edward F. Melcer, Katelyn M. Grasse, James Ryan, Nick Junius, Max Kreminski, Dietrich Squinkifer, Brent Hill, and Noah
Wardrip-Fruin. 2020. Getting Academical: A Choice-Based Interactive Storytelling Game for Teaching Responsible
Conduct of Research. In International Conference on the Foundations of Digital Games (Bugibba, Malta) (FDG ’20).
Association for Computing Machinery, New York, NY, USA, Article 78, 12 pages. https://doi.org/10.1145/3402942.
3403005
[63]
Edward F Melcer and Katherine Isbister. 2021. Learning with the Body: A Design Framework for Embodied Learning
Games and Simulations. In Software Engineering Perspectives in Computer Game Development. Chapman and Hall/CRC,
161–195.
[64]
Edward F. Melcer, James Ryan, Nick Junius, Max Kreminski, Dietrich Squinkifer, Brent Hill, and Noah Wardrip-Fruin.
2020. Teaching Responsible Conduct of Research Through an Interactive Storytelling Game. In Extended Abstracts of
the 2020 CHI Conference on Human Factors in Computing Systems (Honolulu, HI, USA) (CHI EA ’20). Association for
Computing Machinery, New York, NY, USA, 1–10. https://doi.org/10.1145/3334480.3382973
[65]
Florian Mueller, Martin Gibbs, and Frank Vetere. 2008. Taxonomy of exertion games. In Proceedings of the 20th
Australasian Computer-Human Interaction Conference (OZCHI ’08). 263–266. https://doi.org/10.1145/1517744.1517772
[66]
Cristina Hava Muntean, Nour El Mawas, Michael Bradford, and Pramod Pathak. 2018. Investigating the impact of an
immersive computer-based math game on the learning process of undergraduate students. In 2018 IEEE Frontiers in
Education Conference (FIE). IEEE, 1–8.
[67]
Victor Navarro-Remesal and Antonio Loriguillo-López. 2015. What Makes Gêmu Dierent? A Look at the Distinctive
Design Traits of Japanese Video Games and Their Place in the Japanese Media Mix. Journal of Games Criticism 2 (01
2015), 1–17.
[68]
S. M. S. Nugroho, A. S. Utama, M. Hariadi, U. L. Yuhana, and M. H. Purnomo. 2018. HEIRDOM: Multiple Ending
Scenario Game For Mathematics Learning Using Rule-Based System. In 2018 International Conference on Computer
Engineering, Network and Intelligent Multimedia (CENIM). 192–197.
[69]
Otomate. 2013. Norn9 Var Commons. Game [PlayStation Portable, iOS, PlayStation Vita]. Idea Factory. Last played
March 2020.
[70]
Dendi Pratama, Winny Gunarti, and Tauq Akbar. 2017. Understanding visual novel as artwork of visual communication
design. Mudra Jurnal Seni Budaya 32, 3 (2017).
[71]
Dendi Pratama, Winny Gunarti widya Wardani, and Tauq Akbar. 2018. Building Contextual Perception Through
Background Illustrations in Visual Novel with Historical Theme.
Proc. ACM Hum.-Comput. Interact., Vol. 5, No. CHI PLAY, Article 285. Publication date: September 2021.
285:18 Janelynn Camingue, Elin Carstensdoir, & Edward F. Melcer
[72]
prettysmart games. 2016. Code Romantic. Game [PC]. prettysmart games. In-development. Distributed on Steam and
itch.io. Last played March 2020.
[73] Toge Productions. 2020. Coee Talk. Game [PC]. Toge Productions. Distributed on Steam. Last played March 2020.
[74]
P.A. Rego, Pedro Moreira, and Luís Reis. 2010. Serious Games for Rehabilitation: A survey and a classication towards
a taxonomy. In 5th Iberian Conference on Information Systems and Technologies (CISTI). 1–6.
[75]
Nadia Rowena, C. Leetian, Regina E. Estuar, and Michael B. Syson. 2013. Designing Mobile Educational Games on
Voter"s Education: A Tale of Three Engines. International Journal of e-Education, e-Business, e-Management and
e-Learning (2013).
[76] sakevisual. 2016. Backstage Pass. Game [PC]. sakevisual. Distributed on Steam. Last played March 2020.
[77]
Francisco Lepe Salazar, Tatsuo Nakajima, and Todorka Alexandrova. 2013. Visual Novels: An Methodology Guideline
for Pervasive Educational Games that Favors Discernment. Grid and Pervasive Computing Lecture Notes in Computer
Science (2013), 234–243. https://doi.org/10.1007/978-3-642-38027-3_25
[78] Johnny Saldaña. 2015. The coding manual for qualitative researchers. Sage.
[79]
Anastasia Salter, Bridget Blodgett, and Anne Sullivan. 2018. "Just because its gay?". Proceedings of the 13th International
Conference on the Foundations of Digital Games - FDG 18 (2018). https://doi.org/10.1145/3235765.3235778
[80]
Moechammad Sarosa, Zamah Sari, Suhari Suhari, and Hilmy Bahy Hakim. 2017. Multiplatform Visual Novel for
Culture Education in a Part of Using ICT for Developing a Lifelong Learning.
[81]
Braidon Schaufert. 2018. Daddy’s Play. Subversion and Normativity in Dream Daddy’s Queer World. Game Studies 18,
3 (2018).
[82]
Ltd. Spike Chunsoft Co. 2017. Danganronpa V3: Killing Harmony. Game [PC]. Spike Chunsoft Co., Ltd. Distributed on
Steam. Last played March 2020.
[83]
Heintz Stephanie and Lai-Chong Law Ee. 2015. Game Elements-Attributes Model: a First Step towards a Structured
Comparison of Educational Games. In DiGRA ’15 - Proceedings of the 2015 DiGRA International Conference. Digital
Games Research Association.
[84] Four Leaf Studios. 2012. Katawa Shoujo. Game [PC]. Four Leaf Studios. Last played March 2020.
[85]
Seona Lee Somi Lee Sohyeon Jeon Sujin Ri, Marcos Arroyo. 2012. Dandelion - Wishes brought to you -. Game [PC].
Cheritz. Distributed on Steam. Last played March 2020.
[86]
SukeraSparo. 2019. The Expression Amrilato. Game [PC]. MangaGamer, mirai works. Distributed on Steam. Last
played March 2020.
[87]
Dean Sullivan and Jessica Critten. 2014. Adventures in Research: Creating a video game textbook for an information
literacy course. College & Research Libraries News 75, 10 (Jan 2014), 570–573. https://doi.org/10.5860/crln.75.10.9215
[88] Ubisoft. 2017. City of Love: Paris. Game [Android, iOS]. Ubisoft, Ubisoft Entertainment. Last played March 2020.
[89]
VanadiumValor. 2019. Periodicity: Your Chemical Romance. Game [PC]. VanadiumValor, Distributed on itch.io. Last
played March 2020.
[90]
Mauro Vanetti. 2016. Two Interviewees. Game [Windows, macOS, Linux, Android]. Mauro Vanetti.Distributed on
Itch.io. Last played March 2020.
[91] D.P. Vossen. 2004. The nature and classication of games. Avante 10 (01 2004), 53–68.
[92]
xxmissarichanxx. 2017. After-Party Chemistry. Game [Windows, macOS, Linux]. xxmissarichanxx, Distributed on
Itchi,io. Last played March 2020.
[93]
Langxuan Yin, Lazlo Ring, and Timothy Bickmore. 2012. Using an interactive visual novel to promote patient
empowerment through engagement. In Proceedings of the International Conference on the Foundations of Digital Games.
41–48.
Received February 2021; revised June 2021; accepted July 2021
Proc. ACM Hum.-Comput. Interact., Vol. 5, No. CHI PLAY, Article 285. Publication date: September 2021.