A Thesis
entitled
Usage of Emotes and Emoticons in a Massively Multiplayer Online Role-Playing Game
by
Leanna Marie Diaz
Submitted to the Graduate Faculty as partial fulfillment of the requirements for the
Master of Arts Degree in English with Concentration in ESL
______Dr. Douglas W. Coleman, Committee Chair
______Dr. Anthony Edgington, Committee Member
______Dr. Kasumi Yamazaki, Committee Member
______Dr. Amanda Bryant-Friedrich, Dean College of Graduate Studies
The University of Toledo
August 2018
Copyright 2018, Leanna Marie Diaz
This document is copyrighted material. Under copyright law, no parts of this document may be reproduced without the expressed permission of the author. An Abstract of
Usage of Emotes and Emoticons in a Massively Multiplayer Online Role-Playing Game
by
Leanna Diaz
Submitted to the Graduate Faculty as partial fulfillment of the requirements for the Master of Arts Degree in English with Concentration in ESL
The University of Toledo August 2018
When playing a Massively Multiplayer Online Role-Playing Game (MMORPG), methods of communication are more restricted in comparison to real life. Although methods of communicating through voice chat are commonly provided or are readily available through third party software, players of MMORPGs are often faced with the inability to express real-life body language, facial expressions and other non-verbal cues.
To compensate for this, players are offered a variety of in-game emotes which allows them to be able to make their in-game character perform predetermined actions such as dancing or hugging. A string of text may also narrate the action. In other words, it is possible to have a player’s character actually hug another character within the game. In addition, players may communicate through text, either privately or publicly, and are free to use text-based emoticons. This study examines the usage of these emotes and text- based emoticons in the MMORPG Final Fantasy® XIV, how players apply these in-game communication methods to express themselves and what factors determine how and why a player chooses to communicate with other players.
iii
I would not be where I am today if not for the passion shown to me by the educators at the University of Toledo. To Kristi Pawlak, who instilled within me a confidence to pursue my love for the Japanese language and culture; to Dr. Hara and Dr.
Yamazaki, for always supporting me in my endeavors and giving me so many wonderful opportunities; to Dr. Reichelt who taught me the foundations of linguistics and treated the
TAs like her children; to Dr. Coleman and Dr. Edgington, who always went above and beyond their expected duties: from the bottom of my heart, thank you. To my friends, who listened to me and supported me outside of my educational endeavors, who always had my back: thank you. I hope you all know how much you mean to me. LeeAnn,
Kristen, Lindsay and Stephen, you helped me get through life and I will always be grateful for that. There are no words that could express my gratitude adequately.
Table of Contents
Abstract iii
Table of Contents v
List of Tables vii
List of Figures ix
List of Abbreviations x
I. Literature Review 1
A. Introduction 1
B. Structural Framework of Communication 3
C. Impact of Roles and Discourse Communities 9
D. Perceived Prestige 14
E. Neonateness 15
F. Task-based vs Social-based 16
II. Methodology 19
A. Methodology Roadmap 19
B. Participants 19
C. Apparatus and Materials 25
a. Online questionnaire 25
b. Text-based interview 24
c. Participant observation 25
D. Procedure 28
III. Results and Discussion 31
A. Results and Discussion 31
v B. Results and Discussion Roadmap 31
C. Online Questionnaire Results 31
a. Geography 31
b. Player characters 32
c. Sex and gender 36
d. Roles 39
e. In-game investment 42
D. In-game Text-based Interview Results 44
E. Participant Observation Results 47
F. Limitations and Future Research 52
G. Conclusion 52
References 55
Appendices
A. Online Questionnaire Questions 56
B. Semi-Structured Interview Questions 63
C. Measurement of Neonateness 64
D. Final Fantasy XIV Experience 70
a. Character Creation 70
b. Roles 77
c. Communities 78
d. Terms and Mechanics 79
vi List of Tables
Table 1 Comparisons between HSL terminology and examples of their counterparts
within the context of Final Fantasy® XIV...... 4
Table 2 List of participants based on their main data center...... 19
Table 3 List of participants, broken down by where they live...... 20
Table 4 Attributes of the players organized by their real life sex and their real life
gender...... 21
Table 5 List of participants, broken down by age...... 21
Table 6 List of communication channels, their purpose and who can see them...... 27
Table 7 Participant data separated by race and clan...... 32
Table 8 Ranking of races based on neonateness using criteria from Baudouin, and
Tiberghien (2004)...... 34
Table 9 List of races and their associated perceived frequency of emotes and
emoticons...... 35
Table 10 List of participant’s in-game characters, broken down by “gender” as
described by Final Fantasy® XIV...... 36
Table 11 List of key questions and responses sorted by player’s real life sex...... 38
Table 12 List of roles and the number of participants who play that role as a main
role...... 40
Table 13 In-game race and in-game gender of interview participants...... 45
Table 14 List of the rate of recorded shorthands, emotes and emoticons.
stands
as a substitute for the player’s name...... 50
Table 15 Breakdown of occurrences of shorthands, emotes and emoticons by race. ....51
vii Table 16 Ranking of races based on the rate of shorthands, emote and emoticon
usage...... 51
viii List of Figures
Figure 1 The “social” menu expanded in-game...... 7
Figure 2 The “emotes” menu expanded in-game...... 7
Figure 3 The result of either clicking the “blush” emote icon or typing “/blush” into
the chat box with no target selected...... 9
Figure 4 Example of a symmetrical female face from Baudouin and Tiberghien
(2004)...... 17
Figure 5 The default female appearances of the six available races a player may
choose from when creating a character in Final Fantasy® XIV...... 18
Figure 6 A sample from a log file...... 28
Figure 7 Distribution of the player’s main character race, distributed by sex and
gender...... 37
Figure 8 The distribution of roles by sex and gender...... 42
Figure 9 Sample from a log of greeting an in-game interview participant...... 45
Figure 10 Sample of a participant not responding with o/ ...... 46
Figure 11 Interaction between party members inside of a dungeon...... 48
.
ix List of Abbreviations
DPS ...... Damage Per Second
FFXIV ...... Final Fantasy® XIV
MMO...... Massively Multiplayer Online (Game) MMORPG ...... Massively Multiplayer Online Role-Playing Game
RPG ...... Role-Playing Game
x Chapter One
Literature Review
Introduction
Two individuals stand side-by-side after completing an arduous task. The one on the left wears mismatched plated armor and has brightly colored hair while the one on the right wears a matching set of armor and sports a more realistic hairstyle. In celebration of completing their task, the mismatched individual dances in place. It’s a dance they learned as a member of their clan. The other individual also dances, choosing instead to perform a dance they learned in their travels, ending the dance by jumping in circles.
Then, moments later, the mismatched individual snatches the item those two and six others had journeyed for, laughs, and disappears out of sight. The matching individual says nothing and disappears shortly after.
In reality, the above scenario is what might happen after a group of eight players from various parts of the world join together to fell an enemy in a dungeon of a video game. The player whose character is wearing brightly colored clothes represents the player’s newness to the game; they do not have all the pieces of the top tier gear and so they must put together the best gear they are able to access. The dance that player makes their character perform is a dance all characters from a specific in-game race begin the game with because the player hasn’t done optional quests to learn more dances.
In contrast, the other player’s dance is learned because they have gone out of their way to complete those optional quests to learn several dances. They are an experienced player who has noted that players often have their characters jump around to celebrate the defeat of a powerful enemy. In truth, they have actually defeated the enemy several times
1 that day. Their gear matches because they the entire armor set except for one piece—the chest piece that the enemy could drop upon defeat. Obtaining the body piece would complete their armor set and is their entire reason for coming on the journey in the first place.
Upon defeat, the enemy drops the body piece just as the second player hopes. All eight players cast their die in hopes of rolling high enough to claim the item. However, the item is bound to a specific role which the above two players happen to fill. They both claim “need” on the item, rendering everyone else’s claim to it as worthless. In the end, the newer player rolls higher than the more experienced player and obtains the chest piece. The new player, who is friends with six of the party members, are experienced with the dungeon and so even if their inexperienced friend did not win the chest piece, that player and their friends could have simply re-queued for the dungeon.
The more experienced player does not have that luxury. Like every other party the player has joined, once someone else gets the item they want they disband the party. The experienced player who has played the game by relying on joining groups of strangers to complete content will once again have to spend the time searching for a group that is skilled enough to complete the dungeon and hope that their chest piece is what drops instead of one of the multitude of other armor pieces that could drop.
With the advancement of technology, individuals are able to play video games together, even if they are miles apart or have never once met in person. However, the increased distance oftentimes results in restricted methods to communicate with each other. For example, when playing Final Fantasy® XIV, an MMORPG, a player will typically not use video communication because it can be assumed both players will be
2 looking at their gameplay, rather than the reaction of the person with whom they are playing. Moreover, both video and audio communication rely on third-party software which can be frustrating or confusing for some players. Some MMORPGs like Final
Fantasy® XIV offer in-game methods of communication such as the ability to chat with text and emotes—programmed animations usually associated with a line of descripted text—which allow a player’s character to perform predetermined actions. Some players may manipulate the text to modify it as they see fit, including adding individual emote actions or a series of emote actions to suit their needs. This study examines how players interact with each other using text communication and emotes, specifically the type of
Final Fantasy® XIV specific text that is used, the frequency of such text or usage of emotes, and how differences in communication may arise due to the location of the player or due to a certain task the player is trying to achieve. For those unfamiliar with
MMORPGs or Final Fantasy® XIV, more information is available in Appendix D.
Structural Framework of Communication
When theoretical linguistic objects are treated as though they are real world objects, domain confusion occurs (Sypniewski, 2007). An example of this is attempting to study grammar scientifically, treating the study of grammar on the same level as the study of two or more people communicating with one another. Because of how easily domain confusion can occur, it is important to lay the groundwork to root this research in the real world. In order to remain in the physical domain with real world examples rather than the logical domain with supposition alone, the idea of Discourse communities, prestige and neonateness will be examined in order to tie everything back to the physical domain.
3 To this end, it helps to compare Yngve’s (1996) Hard-Science Linguistic terms and definitions as applied to everyday scenarios such as shopping with their application to virtual scenarios such as playing a video game with others.
Table 1.1: Comparisons between HSL terminology and examples of their counterparts within the context of Final Fantasy® XIV.
Channel Text communication, emote communication, other methods of communication (third party software). Participant Player of Final Fantasy® XIV.
Props In-game items such as ore, potions and interactable objects within the environment Role Parts A healing, tank or damage dealing job. In addition, the player’s in-game character may be considered a clothing prop whose role part they can change at will (ie switching from the above jobs to a gatherer/crafting job). Setting Outposts, towns, dungeons, other locations within Final Fantasy® XIV.
In HSL, a linkage models when two or more individuals communicate which includes both the individuals and their surroundings. Consider the following example:
Mary is a customer at a grocery store while George works at the register. Short on time and in need of some quick energy, Mary purchases a candy bar and George handles the transaction. During the transaction, George smiles asks Mary how her day is going so far; she answers positively and asks the same of George. When the transaction is complete,
4 Mary leaves the store with her newly purchased candy bar. For this scenario, HSL would model Mary and George as participants with the role parts of customer and cashier respectively. The grocery store, on the other hand, would be modeled as the setting, and the candy bar that Mary purchased would be modeled as a prop. In addition, channels would model several things, including the smile George gave Mary (light waves making his facial expression visible) and the sound waves allowing Mary and George to hear one another and converse about each other’s day.
The above scenario extends beyond HSL modeling Mary and George communicating as a linkage: it also serves as an example of structural vs functional aspects. Imagine Mary has blonde hair and wears vintage dresses while someone else, perhaps Violet, is a blonde who wears blue jeans instead. Although they are structurally distinct individuals, they can both serve the same function as customers. In other words, the role of customer is not bound to Mary because she is a blonde who wears vintage dresses but rather extends to all individuals who take on the role part of “customer.” This means that Mary, Violet or even Frank or Paul can take on the role of “customer.” The same can be said for the candy bar that Mary purchased; she (or someone else) could have purchased another brand of candy bar instead. Although the two brands of candy bars are structurally distinct, they are functionally the same. Due to their identical function, both brands of candy bars in this example are props in the same functional role, or prop parts.
Massively Multiplayer Online Role-Playing Games (MMORPG) are no different.
For example, in a party of three individual players who are on a quest to defeat a monster, there are three participants who may have three different role parts: a tank, whose job is
5 to hold the attention of the enemy; a damage dealer, whose job is to cause as much damage to enemies as possible; and a healer, whose job it is to restore the health of members of the party to have been injured on their quest. Just as a customer is expected to pay for their goods, tanks, damage dealers and healers are expected to perform their respective jobs when grouped together to complete a task. In this case, the three individuals are modeled in HSL as participants; the dungeon where the quest took place would be modeled as the setting while the quest itself would be modeled as a prop. The players’ characters may be structurally distinct from one another, but the role parts of tank, damage dealer and healer are not specifically bound to those three individuals. That is to say that someone else could replace one of the three individuals if they shared the same role part, for example a healer replacing another healer.
The difference between these two scenarios (purchasing a candy bar versus slaying a monster in an MMORPG) is that the channels of communication (words, gestures) are more restricted in the second scenario. While a customer and a cashier can use a variety of gestures and sounds at their disposal, players of MMORPGs are more limited. Many times, players do not have access to voice chat software or do not wish to use it. This leaves the players restricted to communicating through text and in-game character gestures, otherwise known as “emotes.”
Emotes are buttons in the game which, when activated, allow a player’s in-game character (henceforth known simply as “character”) to perform certain, predetermined actions such as dancing or hugging. Unless disabled, pressing an emote button also provides a string of text describing the action.
6
Figure 1-1: The “social” menu expanded in-game. FINAL FANTASY XIV © 2010 - 2018 SQUARE ENIX CO., LTD. All Rights Reserved.
There are several ways for a player to have their in-game character perform an emote. One example is to have the player open up the “social” menu. From there, they are able to select the type of social activity they wish to examine or initiate. For players wishing to have their characters perform predetermined actions, they would select
“Emotes” from the menu, which would open up a separate menu.
Figure 1-2: The “emotes” menu expanded in-game. FINAL FANTASY XIV © 2010 - 2018 SQUARE ENIX CO., LTD. All Rights Reserved.
7 Within the “Emotes” menu, the player can find several pages of possible emotes for them to choose from. They are separated by general use emoticons, persistent emoticons which will allow the player to have their character perform the same remote repeatedly until interrupted through movement, and expressions which simply change the expression on the character’s face.
For players more comfortable with shortcut commands, they may choose to directly input the emote commands into their chat textbox. This is done by prefacing the command with a forward slash followed by the associated emote command. For example, if a player were to type /blush then the resulting text would be “You blush with embarrassment.” This string of text is paired with a predetermined action if the command already exists. Obviously, typing /dothethrillerdance will not have your character perform any of the dance moves from “Thriller,” although the player could make the text imply as much. By typing /em performs moves from Thriller, the character will stand still but text will appear to indicate that Name> performs moves from Thriller” appears in the chat box: /em performs moves from Thriller /dance The game takes into consideration whether the player’s own character performed the action versus other players, and will change the “you” in the string of text to the name 8 of the player who performed the action if appropriate. The exception is if the player uses a custom emote phrase, in which case the line will always begin with the name of the character whose player input the text regardless of whether it is your character or another player’s character. In addition, the game does not automatically add punctuation to custom emotes, so it is possible to create a custom emote without punctuation or with excessive punctuation. Figure 1-3: The result of either clicking the “blush” emote icon or typing “/blush” into the chat box with no target selected. FINAL FANTASY XIV © 2010 - 2018 SQUARE ENIX CO., LTD. All Rights Reserved. In addition, players may communicate through text either privately or publicly, and are free to use text-based emoticons. Impact of Roles and Discourse Communities Players are more limited in their channels of communication when playing a MMORPG. Players’ gestures (channels) are restricted to the animations the player’s character is programmed to be able to perform; unless both players agree to use voice chat software, they are barred access from oral communication (another channel); text is available as a means to communicate with other players, but similarly to voice chat, some players simply choose to not use it or do not respond to text in the chat box, making the discourse in Final Fantasy® XIV restricted. Without the luxury of seeing other players’ body language, players must, according to Gee (1989), rely on their knowledge of the discourse of other Final Fantasy® XIV players. Gee (1989) makes a distinction between discourse and Discourse. 9 By his definition, the former is the language individuals use to communicate. Discourse with a capital “D” refers not just to the language, but how the communication occurs, taking into consideration gestures and internal jargon. Gee summarizes a Discourse community with the following ideas: “saying (writing)-doing-being-valuing-believing combinations” (p. 6). By extension, a group of individuals who share the same beliefs, values and discourse may be referred to as a Discourse community. Examples of Discourse communities include book clubs, sports teams and communities of gamers. According to this definition, the community of players of Final Fantasy® XIV also fit the description of a Discourse community due to their shared knowledge and jargon of the game. The newer player with the sprout icon is most likely unfamiliar with expected behavior within the game. To explain further, when joining a party through the Party Finder, it is important to note the purpose of the duty. If the purpose is to “farm” for an item, there is the unspoken expectation that if you join the party, you will stick around for multiple runs of the dungeon or trial. An even greater unspoken expectation is that if you win the item you are after that you stick around for at least one more run of the dungeon or trial. Winning your item and then immediately leaving the party is considered rude, and yet this is exactly what the new player with the sprout did in the introduction’s scenario. These kinds of behaviors do not necessarily translate to other online games and so newer players must take care to pay attention to how other players react in order to be a true member of the Discourse community, as Gee would define it. In addition, Gee (2003) indicates that players can become deeply invested in the progress of their characters. Gee called this investment a “projective identification” in which the character becomes what the player wants it to become. This means that the 10 goals and aspirations of the player are projected onto the character and that this projected identity serves as an interface between the various real world identities an individual has and the virtual world identity that an individual assumes when they play a video game. It is important to note that projective identity can only occur when the player imposes their own agency onto their character in a video game. Final Fantasy® XIV allows this right from the beginning of the game with the player’s choice in race, with each race having a different backstory and class, which determines the initial role part the player will have by association with their character. From the very beginning, players are encouraged to think about what kind of goals they want their character to have and why their character is travelling to one of the three city-states. The various tiers of gear are also an inventive for the player to make their character “the best,” at least in terms of equipment. While gear cannot always replace skill, it can serve as a buffer. After all, if you are unable to move out of the range of an attack, more health from better gear will increase your chances of surviving; likewise, an experienced player will simply get out of the range of the attack. More importantly, the gear serves as a status marker, so much so that players often opt to not change the appearance of high-tier gear. In keeping the appearance of their high-tier gear intact, it allows others to know that the player completed a difficult trial in the game and was lucky enough to win the loot. Moreover, Yngve’s (1996) theoretical framework closely examines the way people communicate in the real world and the ease with which people may become domain confused. He defines a communicative event as a linkage involving participants, a setting, props and channels. For example, for the purposes of this study, an individual who plays Final Fantasy® XIV would be a participant; the setting would be the location 11 they play the game in addition to the in-game location; the props would be the things or ideas which are communicated; and the channels would be the energy flow. By using Yngve’s theoretical framework, this study will avoid domain confusion, ensuring that the communication documented takes place in the real world. Jones (1982) also helps root the issue in reality. According to Jones, although what takes place in a video game may be a simulation, that does not diminish the fact that the environment in which the simulation takes place has structure. It is this structure that allows the reality of the simulation to be coherent. Although Final Fantasy® XIV does contain many voiced lines, the majority of the game is in text form without the lines being vocalized. If a player does not understand text from a quest or the text on one of their character’s ability tooltips (a dialogue box that explains what will happen if the ability button is pressed), the player may come off as being unfamiliar with the mechanics of the game. This is especially true if they are indeed a new player, as they will be marked with a sprout icon above their heads, indicating that they are either low level or that they are not very far into the game’s main quest. Because the game counts play time separately for each of a player’s characters, a player who has reached end game content may create another character and may be mistaken for a new player because of the sprout icon. Here, we can see an opportunity for new players to “mushfake” their way into the Final Fantasy® XIV community. As Gee (1989) describes it, mushfake is an attempt to adapt the form of their communication to that of a Discourse community. A player noticing that other players have their characters jump around after the defeat of a difficult enemy may adapt similar behavior in order to get their foot in the door of the Final Fantasy® XIV’s Discourse community. 12 It is important to note that data centers in Final Fantasy® XIV are separated by region, although there are no restriction on which data center in which an individual chooses to play. As the data centers are further separated into servers—or worlds, used interchangeably—this creates an environment in which the Discourse communities may be distinct from one another. There are many minor differences between the Discourse communities between data centers, but the most major one is how players are expected to complete certain duties in game such as a dungeon, raid or trial. There are often times different strategies utilized by the Japan data center and the two data centers in North America. A player’s inability to do the mechanics of a duty in the expected way can often either label that player as an outsider or, as is more likely, an inexperienced player. The location a player is in within the game itself also has an impact on how a player is expected to act, but also what players expect to see. The three starting city states are areas that can be accessed by any player. This means that bots, characters created for the sole purpose of automated behavior such as trading real life money for the in-game currency, are prevalent in the starting areas. However, this also means that the further away a player ventures from the starting areas, the less likely they are to encounter these bots. There are less intrusive bots, such as those used teleport to specific areas players are not normally allowed to teleport in order to autonomously finish quests or gather materials. If a player is in an instance such a trial or a raid, they are expected to perform the duties of their role parts effectively. Players who refuse to or are unable to pull their weight are handled in a variety of ways such as other players ridiculing them in the chat box or the party leader simply kicking them from the party. In contrast, players who are 13 idling in one of the main cities such as late-game Kugane do not have any real responsibilities and are free to simply stand by the market board for hours if they so choose. Players in town are often ignored by others unless a friend wanders by or there is something interesting about their appearance, such as their gear. Perceived Prestige In order to examine if there are notable differences between individuals who use emotes frequently versus individuals who forgo using emotes, players’ role parts (tank/damage dealer/healer), player’s sex and gender, and the player’s character’s sex and race must be examined. Labov (1972) famously conducted a department store study in which he used the variables of casual “fourth floor” and emphatic “fourth floor” to investigate the social stratification of (r) in New York, similarly to how the weight of role parts in Final Fantasy® XIV may affect perceived prestige. According to Labov’s study, there was a concrete stratification of (r) between the three department stores of Saks Fifth Avenue, Macy’s and S. Klein. Labov ranked these stores through key factors such as location, number of pages in advertising in the New York Times and Daily News, and the prices of dresses, girls’ coats, stockings and men’s suits (pp. 170-171). Saks was located in the center of a high-end fashion district and was surrounded by several prestigious stores; Macy’s was located nearby middle-ranged stores while S. Klein, due to its location in the Lower East Side, was ranked the lowest in terms of prestigious locations. Newspaper advertisement was included due to it being understood that the New York Times had a middle class readership while the Daily News had a working class readership. Lastly, the prices of clothing were most expensive at Saks, with Macy’s pricing ranging in-between Saks’ and S. Klein’s. All of these combined resulted in Labov 14 ranking Saks Fifth Avenue as the most prestigious store, with Macy’s ranking in the middle and S. Klein ranking the lowest (p. 170). Keeping this in mind, replicating parts of this study will be beneficial in investigating the reasons behind emote and emoticon usage in an MMORPG. As Labov’s experiment separated parts of his data based on sex, occupation and race, these ideas could be loosely translated to in-game sex, occupation (role parts) and race. In addition, because players may freely create both male and female characters, their sex and gender outside of the virtual reality may be linked to their in- game character’s sex to determine the type and frequency of emotes and emoticons used. Neonateness Another quality that may determine a player’s likelihood of using emotes and emoticons is how neonate the player has decided to make their character. Baudouin and Tiberghien (2004) determined that facial characteristics considered to be average were reported to be more attractive than characteristics that were not average. There were several exceptions to this, such as women who were more neonatal in appearance than other women. Women who demonstrated slightly bigger eyes or a smaller nose, neonatal characteristics, were found to be more attractive than women with explicitly average characteristics. This research can be used to scientifically determine whether there is a link between characters whose players either use or do not use emotes and emoticons and how cute or attractive the player perceives their character. For example, the Miqo’te and Lalafell races within Final Fantasy® XIV would be considered either cute or attractive because of their average appearance with only slightly exaggerated features, especially in contrast to the other available races. Final Fantasy® XIV features a character creator, which can be used to change the size and shape of facial features of the default 15 appearance of all six races. In order to determine whether neonateness plays a role in players choosing to use emotes or emoticons, their characters can be compared to the default versions of their race and clan (the “average” appearance) within the character creator. Furthermore, the neonateness of in-game races will be used in various analyses, particularly within Chapter 3. Please refer to Appendix C for the full details of each in- game race’s neonatal characteristics. Task-based vs Social-based In addition, Derks et al. (2007b) concluded that one of the primary differences between whether an individual chooses to use emoticons is based on the context. In particular, individuals who participate in socio-emotional chats are more likely to use emoticons than individuals who participate in task-based chats. These decisions and the evaluation of the context by an individual are dictated by society’s norms for appropriateness. In this case, Derks et al.’s (2007a, 2007b) research will help determine the significance of the venues within Final Fantasy® XIV, and whether there is a significant difference in the amount of emotes or emoticons used in task-oriented versus social-oriented goals. However, it is important to know that the idea of context is limited. As individuals, we are all structurally distinct. We look different from one another, come from different backgrounds and have had different experiences throughout our lives. Because of this, no two people understand a situation or scenario in exactly the same way. However, Final Fantasy® XIV allows individuals to be functionally similar. This is especially true as the game allows a player’s character to fulfill any of the three combative roles in addition to the two non-combative roles of gathering and crafting. 16 Figure 1-4: Example of a symmetrical female face from Baudouin and Tiberghien (2004). 17 Figure 1-5: The default female appearances of the six available races a player may choose from when creating a character in Final Fantasy® XIV. FINAL FANTASY XIV © 2010 - 2018 SQUARE ENIX CO., LTD. All Rights Reserved. Unlike real life, roles within video games are typically more clearly defined than roles in real life. This helps makes roles and the role parts players take on through their characters more distinct and easier to define. With the above in mind, this study examines the usage of these emotes and text- based emoticons in the MMORPG Final Fantasy® XIV and how players apply these in- game communication methods to express themselves. In particular, this study examines how neonateness, race, venue and virtual world immersion affect the frequency of emote and emoticon usage within the game. 18 Chapter Two Methodology Methodology Roadmap The objectives of this study were to analyze the perceptions and attitudes of the Final Fantasy® XIV playerbase toward emotes and emoticons; to examine whether self- reported attitudes matched with their in-game behavior; and to establish a reference of neonateness to connect a player’s in-game race choice with the likelihood of that player using emotes or emoticons. This chapter identifies and defines the participants of this study, apparatus and materials used, and data collection procedures. Participants Table 2.1: List of participants based on their main data center. Data Center Percentage Total Participants Aether (NA) 42.77% 429 Primal (NA) 29.61% 297 Chaos (EU) 22.63% 227 Elemental (JP) 4.29% 43 Gaia (JP) 0.30% 3 Mana (JP) 0.40% 4 NA = North America; EU = Europe; and JP = Japan. 19 Table 2.2: List of participants, broken down by where they live. Country Total Participants Australia 18 Austria 3 Belgium 6 Brazil 8 Canada 62 Chile 3 Denmark 4 Finland 6 France 17 Germany 26 Ireland 6 Japan 8 Malaysia 4 Netherlands 11 New Zealand 3 Norway 6 Philippines 7 Portugal 3 Russia 3 Singapore 10 Spain 3 Sweden 11 United Arab Emirates 3 United Kingdom 50 United States of America 414 Countries with 2 or less participants: 25 20 Due to Final Fantasy® XIV only allowing for male and female gendered characters to be played, the online questionnaire was structured in a manner to only allow participants to answer as male or female, or to choose to not respond. This was done in order to both protect the identity of players should they be transgendered and to ensure data was more consistent as Final Fantasy® XIV only offers two binary genders and does not separate sex from gender in its character creation screen. Table 2.3: Attributes of the players organized by their real life sex and their real life gender. Player Attributes Total Participants Male (sex): 429 Female (sex): 255 Male (gender): 431 Female (gender): 243 Total (sex): 684 Total (gender): 674 Table 2.4: List of participants, broken down by age. Participant Age Total Participants 18-20 120 21-25 268 26-30 206 31-40 122 41-50 4 51-60 1 61+ 3 Preferred not to say: 9 Total: 733 21 Apparatus and Materials Three tools were used in order to collect and triangulate the data. These tools were an online questionnaire hosted on SurveyMonkey, in-game text only interviews and participant observation. The online questionnaire was used in order to collect a large volume of player characteristics as well as a variety of players’ perceptions and attitudes toward emotes, emoticons and other social interactions. In-game interviews were used in order to give participants an opportunity to expand and explain view points as well as to cross-reference players’ interview answers with their online questionnaire answers to monitor for consistency. Finally, participant observation was used in order to see if the general behavior of players within a particular server was consistent with the data logged from the online questionnaire and in-game interviews. Online questionnaire. A link to an online questionnaire hosted on SurveyMonkey was linked to a popular Final Fantasy® XIV forum on a website called Reddit. Each forum within Reddit is called a subreddit in which the topic can vary between something broad like video games or literature to more specific topics like Final Fantasy® XIV or linguistics. At the time of this study, there were over 176,000 individuals who were a member of the Final Fantasy® XIV subreddit. Due to Reddit’s format, its users are encouraged to upvote topics they find interesting or contributive to the community and downvote topics that do not spark good discussion or contribute very little to the community. The online questionnaire’s level of popularity, and therefore public visibility on the subreddit, was dictated by the subreddit of Final Fantasy® XIV. While the study itself received more responses than expected, the responses totaled less than 1% of the entire community. However, just because an 22 individual is a member of a subreddit community does not mean they are an active reader or even that they visit the subreddit daily or every other day. Due to the nature of online questionnaires, it is still possible for a 1% sample to be representative of a community (Cook, Heath, and Thompson, 2000). Members of the Final Fantasy® XIV subreddit were informed that it was okay to pass the link along to their friends and so not all data comes from the Final Fantasy® XIV subreddit. The link was passed around to other social media sites, particularly Plurk, which is used as an alternative to Twitter. Many Plurk users frequent Reddit, although some users have ties to other websites such as Dreamwidth and LiveJournal, which are popular platforms to roleplay fictional characters. The link to the online questionnaire was also shared on Discord, a popular voice and text chat communication application. Discord is geared specifically toward gamers. In general, all participants were expected to have some experience playing Final Fantasy® XIV and none of the responses indicated otherwise. Although the online questionnaire was available for five weeks, virtually all responses came within the first two days of the online questionnaire being publically available. This was most likely due to the fact that the Final Fantasy® XIV subreddit is quite large and dozens of posts can be made in a day. Therefore, once the initial popularity of the link to the online questionnaire died down, it was no longer on the first page of the subreddit and so it was no longer easily visible. No follow-up reminders were posted on the subreddit, nor were any incentives offered. In other words, the post was just like any other post on the subreddit which allowed for the average user to notice and complete the online questionnaire rather than the contrast of emailing every user of the subreddit and sending 23 follow-up reminders. This explains the less than 1% of user response, but also helped ensure no users in particular were singled out. The online questionnaire itself asked various questions about the participant and the character they created. Questions about geography were asked to determine if there were trends within countries, as well as to determine if there were trends within specific data centers. Participants were asked their thoughts and opinions on certain in-game races and roles (tank, healer, DPS) so that it could be determined whether there were links between character race, character role and frequency of emote or emoticon usage. Participants were also asked about their thoughts and opinions on emoticons and were asked to self-report how frequently they used emotes and emoticons so that it could be compared with data collected in participant observation. Other questions such as ownership of housing and membership of a Free Company, which can be likened to a guild, were asked in order to ascertain the level of a player’s investment and whether the player was potentially focused on battle content, social content or a mixture of both. See Appendix A for the complete online questionnaire. Text-based interview. At the end of the online questionnaire, individuals were asked if they would like the opportunity to participate in an in-game text only interview. Every single individual who was interviewed had also taken the online questionnaire and was aware of what the general topic for the online questionnaire and the interview were. Interviews took place for five weeks, around the same length of time the online questionnaire was open to responses. The in-game interview was a list of ten questions that exclusively concerned emotes and emoticons. Participants were asked to explain in more detail how they felt 24 about emotes and emoticons so that their responses could be compared with data collected from the online questionnaire. During the interview, players were monitored for consistency and their responses were cross-referenced with the answers to their online questionnaire in order to ascertain whether or not participants were presenting themselves differently than how they presented themselves within the online questionnaire. This was done in an effort to verify self-reporting. See Appendix B for the complete list of questions. Participants were asked to expand upon some answers from their online questionnaire. In particular, they were asked to explain in detail whether there was anything that influenced their decision to use or not use emotes or emoticons. They were also asked to explain their perceptions of other players and why other players may or may not choose to use emotes or emoticons. The responses from these questions were then compared not only to the other interview participants’ answers, but also to the answers from the online questionnaire in order to identify any existing trends. When selecting the participants for the interview, a random number generator was used to help randomize participants. However, this study wanted at least one representative of each in-game race and in-game interview participants were randomly selected through a random number generator with excess potential participants being discarded until each race was represented by at least one individual. Participant observation. Data was also collected through participant- observation. This was restricted to just a single server (Siren) within a data center (Aether, located in North America). However, it is important to note that Final Fantasy® XIV allows for cross-server interaction in many cases, particularly the completion of 25 quests or objectives that require more than one individual. This means that, in the case of task-based objectives, it was possible to collect data from all thirteen servers that are tied to the North American Aether data center. Approximately 50 hours of in-game text was collected, copied and logged in external, offline text files. However, note that players can only see most public text in their current vicinity, although there are some exceptions to this rule. For example, a player in New Gridania, one of the three starting areas, will not be able to see the local chat (Say Chat, Shout Chat, Yell Chat) in Limsa Lominsa, another starting area. Some text is semi-public and is visible no matter where the player’s location is. These semi- public chats are Alliance Chat, Free Company Chat and Linkshell Chat. In order to see these three chats, a player must be in a party large enough to be considered an Alliance (minimum of three parties with one member each) or a member of a specific group (Linkshell, Free Company). There are many Linkshells and Free Companies in-game, but players are restricted to membership of one Free Company per character, though may be a member of multiple Linkshells. In contrast, some chats, such as Party Chat and Tell Chat, are private and can never be seen by anyone other than the sender and recipient of a direct message. Therefore, only chats that were publicly visible were logged. Public chats that were logged were Say Chat, Shout Chat and Yell Chat. Chats that require some sort of group membership such as Alliance Chat, Free Company Chat, Linkshell Chat and Party Chats were also logged, but were restricted in that I needed to be in a particular group and therefore could not log all existing group chats. See Table 2.5 for further explanation of the different chats and their purpose. 26 Table 2.5: List of communication channels, their purpose and who can see them. Public Chats Say: Typing in this channel allows players to publicly address other players. The range is quite small, especially in comparison to Yell chat and Shout chat. Shout: When a player types in this channel, everyone in the zone can see it. Zones are quite large and can take anywhere from 5-10 minutes to traverse from one side to the other, depending on the mode of transportation. Yell: This channel is a compromise between Say chat and Shout chat as its ranger is bigger than Say chat but smaller than Shout chat. Semi-Public Chats/Group Chats Alliance: Alliances contain three parties with each party consisting of eight players for a total of twenty-four players. It is a broader “party chat” (see below). The chat is visible when a player is in an alliance which is typically when completing tasks involving combat. Free Its visibility is restricted to members of a particular Free Company. Company: Free Company members cannot see the chat of another Free Company as players are restricted to one Free Company. The chat is visible no matter the player’s location as long as they are a member of a Free Company. Linkshell: Linkshells can contain up to 128 members. They are often used to organize parties, either for content such as dungeons and raids, or to organize for hunts which are either rarely spawned or require multiple players to defeat. Private Chats Party: Parties may have 2-8 individuals. Only members of a particular party may see that particular party’s chat. The chat is visible no matter the player’s location is as long as they are in a party. Tell: These are direct messages from one player to another. These messages are hidden from everyone except the sender and the recipient. Tell chat is unavailable in some instances such as content requiring Alliances. Echo: It is most often used as reminder text, such as when using a crafting macro. For example, a player might choose to include an echo text at the end of each macro when crafting, especially when more than one macro is necessary. Only the player can see this text and Player 2 cannot see the echo text of Player 1. 27 When collecting data, all dialogue was logged in text files. This data included time, player name, player server and then the content of the dialogue itself. All of this was provided by the chat system within the game itself and so no extra work was necessary to add times, names or servers. However, extra notes to differentiate location were added to the logs. In Figure 2-1, Kugane is the location and the time is based on a 24-hour clock. After the time are character names with their dialogue separated by a colon. Tranquil is explaining how they left another server (Balmung) for Siren. Figure 2-1: A sample from a log file. Procedure Those who participated in the online questionnaire were asked approximately 30 questions. The questions changed depending on what was applicable. For example, if a participant answered that their data center was Aether then they would not be asked if their main server was a server outside of Aether. Most questions were multiple choice, but questions regarding opinions allowed participants to write as much or as little as they desired. Participants were allowed to skip any question in an effort to encourage participants to not only finish the online questionnaire but to also give solid, meaningful answers. Feedback from participants from the subreddit indicated that this was appreciated. However, the result was that often times only around 70% of participants responded to any given question. General questions were placed at the beginning of the 28 online questionnaire, such as data center location and character race, while questions asking for participants to explain their thoughts in detail were placed toward the end of the online questionnaire. In addition to disclosing their preferred data center and server, participants were also asked to give their opinions on matters involving roles (tank, healer, damage dealer) and in-game character races. Some questions asked if the participant felt certain roles or races encouraged players to behave or speak in a certain manner. Other questions were related to how invested a player was in the game, such as having membership in a Free Company or owning housing. Lastly, some questions asked about the roleplay community on a participant’s server to help determine if the participant frequently communicated or witnessed communication of this style, which could affect how a player communicates with others outside of roleplay. At the conclusion of the online questionnaire, participants were asked if they would like to participate in an optional follow-up interview. Participants who were interested were asked to leave details of their main server and character name so they could be contacted. Interview participants were asked ten questions about their opinions and perceptions of emote and emoticon usage in the game. Due to the semi-structured nature of the interview, some participants were asked follow-up questions in the event of their answers not being clear enough. Once the online questionnaire was closed to new respondents, the data was analyzed to look for trends. Some trends, such as traits related to “cuteness” and physical appearance were established in advance. Other trends were discovered during the data collection. Discourse analysis was used to determine which traits participants mentioned 29 the most frequently such as “cuteness” and “height” so that connections could be made with other factors such as character race. Notes were taken not just about the trait mentioned, but also about the details participants provided such as wanting their character to be “as tall as possible” versus wanting their character to be short because “short characters are cute.” Characteristics about participants who answered these questions were also logged and compared with other players with similar characteristics, such as players who played the same in-game race. The data collected was entered into a spreadsheet and tests for statistical significance were run through R-Commander. Participant observation was similarly coded. Log files were thoroughly examined and the number of total dialogue lines was counted. The number of times players used emotes or emoticons within their dialogue was then tallied. Both raw frequency and relative frequency were documented and the data was tallied and formatted into a table. Again, only the data that was publicly available to me as a player was logged. Finally, the semi-structured interviews were used to validate response answers and to triangulate the data. Answers were compared to their online questionnaire responses to ensure that their responses were consistent. 30 Chapter Three Results and Discussion Results and Discussion Roadmap This chapter discusses and breaks down the data for online questionnaire results by geography, player character selection, participant sex and gender, in-game race, roles and investment in the game itself. Results of the semi-structured interview and participant observation are also addressed. Lastly, the effect of neonateness as perceived by players and as calculated through measurement of in-game facial structures is detailed. Online Questionnaire Results The online questionnaire had 1,019 participants in total. As participants were free to answer only questions they felt comfortable answering, not all participants answered each question. The number of responses for each question hovered around 700-800 responses while the first two questions (data center and server location) had over 1,000 responses. Geography. As noted previously, the location of a data center does not dictate where players actually live. Please refer to the Table 2.1 and Table 2.2 for exact details. There were 726 players who indicated they played on a North American data center, while only 414 were from the USA and 62 from Canada for a total of 476 North American users. This indicates that at least 250 of the participants do not play on their own country’s server. Likewise, there were at least 275 players who lived in various parts of Europe while only 227 actually play on a data center located in Europe. Finally, 50 players indicated they play on a Japanese server while only 8 participants listed Japan as 31 the country they live in. However, including other Asian countries such as Malaysia (4) and Singapore (10), this boosts the number from eight to 22. Player characters. In Table 3.1, data consists only of main characters, meaning that players may have additional characters but that their response is the race they play as the most. Here, the most popular race choice is the Miqo’te with a slight favor toward Seekers of the Sun clan. Table 3.1: Participant data separated by race and clan. Race and Clan Participants Totals Raen Au Ra 77 171 Xaela Au Ra 94 Duskwight Elezen 20 63 Wildwood Elezen 43 Midlander Hyur 92 135 Highlander Hyur 43 Plainsfolk Lalafell 40 119 Dunesfolk Lalafell 79 Seekers of the Sun Miqo’te 158 277 Keepers of the Moon Miqo’te 119 Sea Wolves Roegadyn 15 42 Hellsguard Roegadyn 27 Total: 807 807 When explaining why they selected their particular race, 69 out of 733 participants indicated that at least part of their reasoning was based on how cute the race was in comparison to others. A total of 54 out of 733 participant responses indicated that part of their selection was based on the similarity of a particular race to humans. There 32 were seven participants who specified that they took the proportions of their characters into consideration. Of these seven participants, three participants noted in particular how disproportional Elezen were, especially their long necks. Out of those seven participants, one indicated they liked the proportions of Roegadyn in comparison to the other races while the remaining three noted that only Hyur, Miqo’te and Au Ra were proportionally correct. On several occasions, male Elezen and male Miqo’te were noted to either appear too stereotypically gay for their liking or that they did not want to be mistaken as gay by playing either of those races. In general, the consensus was that participants’ selection of race was mostly a matter of aesthetics to them and that they picked as closely as they could to match real life (Hyur), or picked closely to real life with slight variations (Au Ra, Miqo’te). Others chose their race based on the type of character design they were used to in other games or liked the exaggerated proportions (Elezen, Roegadyn). When describing why players modeled their characters the way they did, the most common response was that it was based on cuteness as perceived by the participants (131 responses), hair (77 responses), character height (54 responses) animal-like features such as tails (29 responses), scales (31 responses) and length or style of ears (54 responses). This data agrees with Baudouin, and Tiberghien (2004), in which players were more frequently drawn toward either a more normal appearance (Hyur) or an appearance with only slightly exaggerated features (Miqo’te, Au Ra). Lalafells were the fourth most popular race choice. However, the gap between the number of players who selected Elezen (63) or Roegadyn (42) is significant in comparison to those who chose Lalafell (119). Even so, the proportions of a Lalafell are mathematically closest to that of a child, 33 which may encourage players to select their race for their cuteness or for their childlike appearance. However, neonateness alone was not the primary factor in choosing an in- game race. Table 3.2: Ranking of races based on neonateness using criteria from Baudouin, and Tiberghien (2004). Race Neonatal Ranking Lalafell 1 Au Ra 2 Hyur 3 Elezen 4 Miqo’te 5 Roegadyn 6 Elezen, which rank closely to Hyur, were consistently mentioned to have disproportionate limbs. Responses indicated that Elezen were older or mature looking, which played into their reasoning for selecting or not selecting the race. Despite being nearly as neonatal as Hyur, participants consistently indicated that they chose Miqo’te over Elezen because they were the most human looking but with the perk of additional features such as cat ears and a tail. See Table A-2 and Table A-3 for further explanation of the method used to rank the races. There were 683 players who responded to which of the in-game races were most likely to use emotes or emoticons, Lalafells (223) ranked the highest and Miqo’te (119) ranked as the second most likely to use emotes and emoticons. Elezen were ranked as the 34 least likely, with only twelve responses mentioning them at all. Participants indicated that Elezen are played by more serious players or that Elezen were more “human” in appearance and so were less likely to use emotes or emoticons. Hyur and Roegadyn were both noted to be masculine by participants, several pointing out that Hyur were well- known for the /flex emote, while Roegadyn were typically not as communicative with emotes or emoticons. In both cases, these reports were beyond the scope of the study. Table 3.3: List of races and their associated perceived frequency of emotes and emoticons. Race Perceived Frequency Ranking Lalafell 223 1 Miqo’te 119 2 Au Ra 59 3 Roegadyn 26 4 Hyur 16 5 Elezen 12 6 The top three ranks were most often accused of macro-spamming, which is either copying and pasting the same emote in chat multiple times or using a macro to post the macro automatically. Overall, players were more likely to report that certain races, in particular Roegadyns, did not use emotes or emoticons or that they did not see one race using emotes or emoticons more than other races. The most common reasons reported for not using emotes were because the players were “too serious” or because some players 35 equated emotes with spam. The data collected confirms the general consensus of the participants. Sex and gender. As noted before, the online questionnaire did not allow for non- binary answers in order to protect the identity of transgendered participants, and due to the fact the game itself does not distinguish between sex and gender, only providing “male” and “female” genders. While not a significant number, there are some players who identified as a gender that does not match their sex. Regardless, there were more males, both biologically and as a gender identity, who responded to the online questionnaire. Table 3.4: List of participant’s in-game characters, broken down by “gender” as described by Final Fantasy® XIV. RL Sex RL Sex x Character Sex Total M F M 207 221 428 In-game Sex F 57 197 254 Total 264 418 682 When explaining why they chose the gender of their character, nearly all responses noted that it was to match their own gender. Other common reasons were because the gender they picked was the most common player character gender in video games they played that did not allow for players to select their gender. 36 Distribution of in-game race, sex and gender 140 131 132 120 97 100 89 91 91 80 76 80 66 66 65 65 60 34 34 35 31 40 26 32 31 23 23 23 Number Number Participants of 20 7 5 0 Lalafell Au Ra Hyur Elezen Miqo'te Roegadyn Races Males (sex) Male (gender) Female (sex) Female (gender) Figure 3-1: Distribution of the player’s main character race, distributed by sex and gender. Out of the total participants who answered both about their character’s gender and their real life gender, 254 female characters were played by 197 female participants and 57 were played by male participants. Likewise, 428 male characters were played by 221 female participants and 207 male participants. This data is unsurprising, especially when compared to data from Baudouin, and Tiberghien (2004) in which individuals were more drawn toward an average appearance with only slightly exaggerated features. Just as is typical in real life, the character models in Final Fantasy® XIV have female models that are more neonate than their male counterparts. 37 Table 3.5: List of key questions and responses sorted by player’s real life sex. Questions Male Female Ratio Do you ever explain mechanics Yes (382) Yes (226) Male (0.893) of dungeons, trials or raids to No (38) No (22) Female (0.893) other players? Unsure (8) Unsure (5) Do you ever use acronyms Yes (349) Yes (221) Male (0.814) such as "lol," "lmao" or other No (73) No (27) Female (0.867) similar acronyms when Unsure (7) Unsure (7) communicating with other players? Do you ever use emotes to Yes (278) Yes (242) Male (0.850) communicate with other No (47) No (12) Female (0.949) players? Examples include Unsure (2) Unsure (1) /hug, /dance, /bow? Do you ever use emoticons to Yes (305) Yes (209) Male (0.713) communicate with other No (115) No (42) Female (0.823) players? Examples include XD Unsure (8) Unsure (3) :D ;n;? Overall, participants revealed that they were more likely to communicate with other players than not. This is a logical response, considering that it is impossible to complete much of the game’s content alone, either through the sheer difficulty of some content or because the game simply will not allow a player to complete content alone. Most participants admitted to having a sense of out of character-ness when interacting with other players as the game’s non-playable characters (NPCs) do not use immersion- breaking dialogue, such as “lol.” In all cases, female participants were more likely to engage in communication than male participants. The largest difference between male 38 and female responses was were the participants who used emotes to communicate (0.099 difference) and the participants who used emoticons to communicate (0.110 difference). Roles. There are three roles within the game: tanks, which maintain the attention, or aggro, of enemies; damage dealers, also known as DPS (damage per second), which have access to a plethora of damage dealing spells, actions or abilities; and healers, who heal the party. Players are able to switch between these roles, but only when not inside of a dungeon, trial or raid. It is important to note that roles are simply expectations. For example, DPS are capable of “stealing” aggro of enemies from the tank. Likewise, healers and tanks are both able to deal damage to enemies. Many roles can self-heal or heal other party members who have taken damage. In general, performing actions outside of your expected role is not as effective. In other words, most healers will not deal more damage than a DPS, either because their attacks are not as potent or because they are expected to split their attention between healing and attacking. Tanks were noted to be the most confident players who were the most likely to lead the group and explain mechanics. Responses also indicated that tanks were considered one of the least talkative roles and that the majority of their conversation was to their co-tank if applicable, usually asking which of them wanted to be the main tank while the other off-tanks, when to provoke an enemy and when shirk (an ability that manages enemy aggro/attention) was necessary. Given that tanks are expected to physically lead the way in dungeons, raids and trials, the information here makes sense. Their button combinations for maintaining aggro on enemies or for maintaining higher damage is relatively simple in comparison to specialized damage dealers, which affords them more opportunity to talk in chat. Despite this, they were reported to be mostly 39 silent. Several participants reported that tanks were the most likely to have an “assertive” or “high and mighty” personality. Tanks who were more talkative were labeled as more likely to have a negative attitude. Table 3.6: List of roles and the number of participants who play that role as a main role. Role Participants Tank 301 Melee DPS 247 Ranged DPS 172 Magical DPS 328 Healer 410 Total 1,458 Responses about damage dealers were incredibly varied, just as the types of DPS (melee, magical and ranged) are also varied. Melee DPS were reported to be silent just as frequently as they were reported to be chatty. They were said to be the ones most likely to call out positions other players needed to take for certain bosses, meaning they were the most likely to take on a leadership role. In addition, they were reported as the most likely to have a spam-like emote when performing the melee Limit Break, a group’s ultimate attack that is typically only able to be used once per dungeon, raid or trial. Melee DPS were also reported to be able to show fondness or trust of another player through their use 40 of player buffs available to them. Player buffs boost the defense, attack or other stat of either themselves or other party members. One reported example is a Dragoon’s buff which increases their damage, as well as a player they tether themselves to. When tethered, players must stay within a certain range of each other, otherwise the tether, and thus the buff, will disappear. Due to the range limit, sometimes other players may be too far away to give a buff and so melee DPS (or any other buff-giving DPS) may choose someone closer. Therefore, if Player A gives Player B a buff, it does not necessarily mean Player B is playing well; it could simply be that Player B is the only player that is consistently nearby to receive a buff. Magical DPS were often said to probably main a healer and so were most likely to communicate like a healer while ranged DPS were either reported to be the quietest of the DPS or compared to healers, similarly to magical DPS. Overall, the consensus was that there is no singular “attitude” that can be used to describe DPS. The most reported method of communication for healers were macros which indicated when they were reviving another player. These macros are typically paired with a line of text, the player’s name who is being revived and sometimes a sound effect. This is because in the case of healers, they are afforded one instant-raise every 60 seconds and it is preferable to use an instant raise over a hard cast which can take around seven seconds. Because one healer’s instant revive can override another healer’s revive if they are cast around the same time, macros like this were reported to be generally appreciated. Aside from macros, healers were reported to be the meekest type of players who rarely took charge of the group. They were also reported to be the most likely to either take the blame or be accused of failing mechanics in a dungeon, raid or trial regardless of whether 41 or not it was their fault. In general, the consensus was that healers tended to be nicer players, but that there were also some healers who were arrogant. In terms of self-reporting on the online questionnaire, participants were mixed in their opinion of how closely their own communication methods matched other players who played the same role. Out of 737 responses, 287 indicated that they communicated similarly to those they shared a role with; 161 indicated that they communicated differently than those they shared a role with; and 289 were unsure. In addition, an overwhelming 90% of participants indicated they picked their main role because it was a role they enjoyed or because it was a role they performed best. Distribution of roles by sex and gender 200 189 186 179 184 180 167 158 159 160 160 138 139 140 119 120 114 100 84 83 80 67 64 61 63 58 60 52 40 Number Number Participants of 20 0 Tank Melee DPS Ranged DPS Magical DPS Healer Races Male (sex) Male (gender) Female (sex) Female (gender) Figure 3-2: The distribution of roles by sex and gender. In-game investment. Finally, players were asked several questions relating to content that was not task-specific. The majority of responses (65.8%) indicated that they 42 chose their server in order to play with friends. Despite this, 66.6% of responses indicated they did not take the starting area into consideration when they first began playing the game. The likely reasoning for this is because players have eventual access to all roles and jobs. The majority of responses (92.2%) indicated that they were currently a member of a Free Company at the time of the online questionnaire. Out of 800 responses, 620 players noted that they had housing of some kind, whether it was shared with a friend or housing in which they had ownership. Houses in Final Fantasy® XIV allow players to express themselves in various ways. Out of 741 responses, 612 indicated that they frequently use shorthands like “lol” or “lmfao” when communicating with other players. When reviewing the responses to the online questionnaire, there were only eight individuals who responded in this way to the online questionnaire itself. Seven individuals used “lol” at some point in their online questionnaire and all but one answered “yes” to this question; the one who did not chose did not answer the question. There was a single instance of “\o/” in the online questionnaire, which is an emoticon for throwing one’s hands up in the air, usually in greeting or in celebration. Out of 739 responses, 553 indicated that they used emoticons when communicating with other players. When reviewing the responses to the online questionnaire, there were several instances of “XD” and “^^” which are a laughing and happy face respectively. However, these were only used as an example of what other players might use and were not the participant using these emoticons in their own answers. 43 Out of 739 responses, 671 indicated they used emotes when communicating with other players. When reviewing the responses to the online questionnaire, there were no responses which used a string of text that would be used for an emote if entered in-game. While it is possible this was due to the fact it would simply exist as a string of text rather than as an emote due to the nature of the online questionnaire, examples of participant observation indicated that players do still use strings of emote text in chat, sometimes going so far as to place a space before the text string so that the emote shows up as text rather than as the preformatted text and action. Overall, players were consistent with their answers in that players who answered “no” to using emotes, emoticons or shorthands did not use them in their responses to the online questionnaire. However, there was no noticeable difference in how players who answered “no” typed versus how players who answered “yes” typed. Most responses were written as a narrative response and the most common characteristic in the answers was the lack of capitalization, something common in text speak and within the game itself. There was also no noticeable difference in the typing styles of males versus females or in the way male participants responded versus female participants. In-game Text-based Interview Results Participants were asked exclusively about their thoughts on others’ usage of emotes and emoticons. Ten interviews were conducted on various North American and European data centers. In interviewing the participants, I selected the default male Hyur appearance to be consistent. See Figure 3-3 for an example. For additional consistency, each interview began the same way with an introduction and an interest check to see if each participant was still willing to do the in- 44 game interview. This was done to eliminate variables such as responses to different introductions. Please refer to Table 3.7 for the list of in-game interview participants. Figure 3-3: Sample from a log of greeting an in-game interview participant. Table 3.7: In-game race and in-game gender of interview participants. Participant Character Race Gender Lalafell-1 Female Lalafell-2 Female Au Ra-1 Male Miqo’te-1 Female Miqo’te-2 Female Miqo’te-3 Female Miqo’te-4 Male Hyur-1 Female Elezen-1 Female Roegadyn-1 Male 45 All but one of the interview participants responded in return with a o/ emoticon of their own with one exception. Figure 3-4: Sample of a participant not responding with o/. Miqo’te-3F was the only one who acknowledged that they used emotes and emoticons more than other players. Eight others stated that they used them a normal amount while Roegadyn-1M stated they did not use emotes or emoticons at all. None of the participants found a “reasonable” amount of emotes or emoticons to be bothersome. “Reasonable” was defined by Miqo’te-2F to be infrequent enough that it would not clutter up someone’s chat window. The Roegadyn-1M and the Elezen-1F both reported that they disliked Limit Break macros which were several lines long and often times congested the chat window. When asked which of the in-game races were most likely to use emotes or emoticons, all ten participants indicated either Miqo’te or Lalafell would be the most likely to use them; Hyur-1F and the Au Ra-1M both indicated that Au Ra were also likely candidates. Six of the participants (Miqo’te-2F, Miqo’te-3F and Miqo’te-4M; Hyur-1F; Roegadyn-1M and the Elezen-1F) also indicated that bots located in the three starting city-states tended to be male Hyurs and that they frequently used bothersome macros. However, Miqo’te-4M and Elezen-1F both indicated they felt that those types of bots were fewer in number than they used to be a year or two ago. Miqo’te-3F and Au Ra-1M were the only participants to indicate that they purchased emotes from the Mog Station, the official website for Final Fantasy® XIV’s 46 real money transactions. It is likely the number is so low because many of the emotes are emotes that used to be in the game but are no longer able to be acquired. Therefore, the longer a player actively plays the game, the less likely they will have to spend money on purchasing emotes from the Mog Station. Finally, all ten participants indicated that they either did not play another MMORPG or that the culture of emotes and emoticons was not significantly different from other MMORPGs. Miqo’te-2F and Roegadyn-1M did note that the culture was “nicer” and that there were more friendly players in Final Fantasy® XIV than in other MMORPGs. When asked why they thought this, Miqo’te-2F indicated that Final Fantasy® XIV used emotes and emoticons a little more than other MMORPG players which made them seem a bit friendlier; Roegadyn-1M cited the atmosphere in general rather than a use or lack of use of emotes or emoticons. Participant Observation Results Approximately 50 hours of text was copied from the game on the Siren server and into a text document for comparison. These logs took place daily for approximately five hours each session. At the end of the 50 hours, 956 lines of text had been logged. During this participant observation, I remained quiet unless I was directly addressed as a player. 812 lines were recorded from various different chats. Out of these 812 lines of text, 229 contained either an emote, an emoticon or shorthand. In the culture of Final Fantasy® XIV, shorthands like “lmfao” and other abbreviations such as “bio” for “bathroom” are usually considered acronyms. Emotes are both a string of text such as “/hug” and the automatic text that displays by default with 47 that string of text. Emoticons consist strictly of keyboard characters such as :) or 8D and are used to express emotions rather than an action. Figure 3-5 is an example from the participant observation. It took place in Doma Castle, a level 67-68 dungeon, meaning 2-3 levels away from the end game dungeons at level 70. The dungeon took about 33 minutes to complete and in that time, only twelve lines of dialogue were exchanged. Figure 3-5: Interaction between party members inside of a dungeon. To further expand upon the dialogue, the Lalafell and Elezen were the only ones to explicitly greet their party members and they chose to do so in different ways. Three minutes into the dungeon, the Au Ra informed the group that their friend, the Miqo’te, needed to leave the computer for a bit. The Lalafell, who was the tank at the time, waited for the Miqo’te to return, but the Au Ra assured them that it was okay to progress through the dungeon. During one of the boss encounters, the Elezen failed a mechanic which caused her to be thrown around the room. She compared being thrown around the room 48 to a ping pong ball being hit. She pointed out their mistake with a “whoops” while laughing about it and the Miqo’te joined in as well. During the final boss encounter, the Lalafell informed the DPS to use the limit break, which would deplete anywhere from 5- 10% of the boss’ health. After the boss was defeated, the Lalafell told the group they had done a good job while the Elezen thanked them for the party and the Lalafell thanked the group in general. The most likely place emotes took place were in the city-states or in dungeons, trials or raids as the party prepared to engage a boss. This data indicates similarity to the data from Derks et al.’s (2007a, 2007b) research on task-based versus social-based emoticon usage. Prior to engaging the boss, players will often wait at the entrance to the boss room. It is during this time that interactions become social based, as there are no real responsibilities, similarly to how players have their characters idle in a city-state. Once the boss is engaged, the situation becomes task-based. In city-states or questing areas, players are free to speak in however manner they want because there is generally nothing they need to communicate in order to complete a task. The same can be said for the emotes and emoticons that occur in dungeons, trials and raids before the boss. Before the boss is engaged, players have little responsibility. There is some expectation to appropriately buff and protect the group, but this preparation typically only takes a few seconds. Likewise, if it is a larger group such as a 24-man raid, players who are familiar with the fight have little to do if another player is explaining the mechanics of the fight. It was during these lulls that emotes were recorded the most in a spam-like fashion. 49 Table 3.8: List of the rate of recorded shorthands, emotes and emoticons. stands as a substitute for the player’s name. Type Expression Emote Text City Rate Dungeon Rate (%) (%) Shorthand bio N/A 0 (0%) 3 (1.25%) Shorthand lmao N/A 2 (0.83%) 8 (3.33%) Shorthand lmfao N/A 1 (0.42%) 11 (4.58%) Shorthand lol N/A 28 (11.67%) 8 (3.33%) Emote /vpose celebrates victory. 0 (0%) 9 (3.75%) Emote /bstance readies for battle. 1 (0.42%) 27 (11.25%) Emote /cheer lets out a cheer. 4 (1.67%) 8 (3.33%) Emote /songbird struts like a songbird 2 (0.83%) 7 (2.92%) does the Black Emote /brpb 0 (0%) 2 (0.83%) Ranger pose. Emote /dance dances happily. 3 (1.25%) 2 (0.83%) gives makes a straight Emote /straight 0 (0%) 3 (1.25%) face. Emote /simper smiles weakly. 2 (0.83%) 5 (2.08%) Emoticon o/ N/A 22 (9.17%) 20 (8.33%) Emoticon :) N/A 0 (0%) 2 (0.83%) Emoticon XD N/A 5 (2.08%) 0 (0%) Emoticon ^^ N/A 2 (0.83%) 4 (1.67%) Emoticon :3 N/A 15 (6.25%) 4 (1.67%) Total Total Total Shorthands Emotes Emoticons 61 105 74 City Rate % Total Dungeon Rate % Total % of lines 11.51% 23.60% 25.10% 50 Table 3.9: Breakdown of occurrences of shorthands, emotes and emoticons by race. Total Shorthand Emote Emoticon Total Race Lines of Occurrence Occurrence Occurrence (%) Dialogue (%) (%) (%) Lalafell 184 6 (3.26%) 37 (20.11%) 3 (1.63%) 46 (25.00%) Au Ra 104 18 (17.31%) 16 (15.38%) 2 (1.92%) 36 (34.62%) Hyur 180 2 (1.11%) 22 (12.22%) 4 (2.22%) 28 (15.56%) Elezen 165 8 (4.85%) 5 (3.03%) 3 (1.82%) 16 (9.70%) Miqo’te 237 7 (2.95%) 76 (32.07%) 23 (9.70%) 106 (44.73%) Roegadyn 86 1 (1.16%) 6 (6.98%) 1 (1.16%) 8 (9.30%) Total 956 42 162 36 240 Table 3.10: Ranking of races based on the rate of shorthands, emote and emoticon usage. Shorthand Emote Rate Emoticon Rate Combined Race Rate Ranking Ranking Ranking Total Ranking Lalafell 3 2 6 3 Au Ra 1 3 3 2 Hyur 6 4 2 4 Elezen 2 6 4 5 Miqo’te 4 1 1 1 Roegadyn 5 5 5 6 51 Overall, emotes, emoticons and text speak were more common during participant observation than during the interviews or the online questionnaire. While it is possible that those who used emotes, emoticons and shorthands would have talked more formally in an interview, it is just as likely that those who use these typing patterns do not feel the need to “save face” because they do not have a negative perception or attitude toward the typing pattern. Limitations and Future Research One of the major flaws of this study was that the data was concentrated on one particular group of players. In the future, it would be beneficial to study other MMORPGs in order to determine the differences between each Discourse community. Another fact outside the control of this research was the fact that not all races (and in-game jobs) were available at the launch of the game. For example, the Au Ra race was added as a playable race in the Heavensward expansion which released in 2015, approximately two years after the launch of the base game, FINAL FANTASY XIV: A Realm Reborn. Conclusion This study set out to examine how neonateness, race, venue and virtual world immersion affects the frequency of emote and emoticon usage within the game. Data from the online questionnaire, semi-structured in-game interview and participant observation show the following information. Although Hyur are advertised as the “average” race, there is only a 0.016 difference between Hyur and Elezen in terms of neonateness. Au Ra have slightly exaggerated features, providing a more neonatal appearance but Miqo’te, despite their 52 large eyes, were ranked fifth out of the six races in terms of neonateness. Lalafells are the most neonatal race while Roegadyn with their broader features are the least neonatal. Players of Miqo’te were the most likely to use emotes and emoticons while they ranked fourth in shorthand rate. Au Ra were ranked second overall, but ranked third in both emote and emoticon rate while ranking first in shorthand rate. Lalafell, the most neonatal race, ranked third overall while ranking sixth in emoticon rates but ranking second in emote rates. Hyur, the base average race, ranked fourth overall, with its second ranking of emotes and sixth ranking of shorthands standing out. Elezen were ranked fifth overall while ranking second in the rate of shorthand. Lastly, Roegadyns were ranked sixth overall, ranking fifth across the board for rates of shorthands, emotes and emoticons. There was not a statistical significance between character race and shorthands, emotes and emoticons used when a Kruskal-Wallis test was applied (χ2 = 4.7598, df = 5, p-value = 0.4459). However, there was a statistical significance between character gender and the frequency of shorthands, emotes or emoticons used (χ2 = 4.5016, df = 1, p-value = 0.03386). Venues itself did not have as much as an impact as the responsibility a player had. Although players were more likely to use emotes and emoticons in combative areas than in non-combative areas such as city states, a Kruskal-Wallis test indicated that rate of emotes, emoticons and shorthand in cities (χ2 = 5.9911, df = 5, p-value = 0.3071) and the rate of emotes, emoticons and shorthand in dungeons (chi-squared = 7.8317, df = 5, p- value = 0.1658) were statistically insignificant. Lastly, virtual world immersion did not seem to have as much of an effect as a player’s character race or current responsibilities. Players who indicated an active 53 participation in the social aspect of Final Fantasy® XIV were no less likely to use emotes or emoticons than those who did not participate in any social-based groups or activities. If anything, it was a player being drawn in by the cute or neonatal appearance of a particular race that made them a more likely candidate to use emotes or emoticons. With the data provided in this study, the conclusion of Derks, Bos and Grumbkow (2007a, 2007b) that people are more likely to use emotes and emoticons for social-based activities rather than task-based activities is not strictly true, as there is evidence to support the opposite as seen in Table 3.8. However, one must take into consideration the overlap between social-based activities and task-based activities. In this case, more research should be done on whether the players perceived completing a dungeon to be a social-based or task-based activity. Although the completion of a dungeon is clearly a goal, some players may have the separate goal of simply playing for fun or interacting with others, rather than completing story or quest objectives. 54 References Baudouin, J., & Tiberghien, G. (2004). Symmetry, averageness, and feature size in the facial attractiveness of women. Acta Psychologica, 117(3), 313-332. doi:https://doi.org/10.1016/j.actpsy.2004.07.002 Cook, C., Heath, F., & Thompson, R. L. (2000). A Meta-Analysis of Response Rates in Web- or Internet-Based Surveys. Educational and Psychological Measurement, 60(6), 821-836. doi:10.1177/00131640021970934 Derks, D., Bos, A. E., & Grumbkow, J. V. (2007a). Emoticons and online message interpretation. Social Science Computer Review, 26(3), 379-388. doi:10.1177/0894439307311611 Derks, D., Bos, A. E., & Grumbkow, J. V. (2007b). Emoticons and social interaction on the Internet: the importance of social context. Computers in Human Behavior, 23(1), 842-849. doi:10.1016/j.chb.2004.11.013 Gee, J. P. (1989). Literacy, discourse, and linguistics: Introduction. Journal of Education, 171(1). Gee, J. P. (2003). What video games have to teach us about learning and literacy. New York: Palgrave Macmillan. Jones, K. (1982). Simulations in language learning. New York, NY: Cambridge University Press. Labov, W. (1972). The social stratification of (r) in New York City department stores. Sociolinguistic Patterns, 43-54. Philadelphia: University of Pennsylvania Press. Sypniewski, B. (2007). Domain confusion. LACUS Forum XXXIII. 399-404. Yngve, V. H. (1996). From Grammar to Science: New Foundations for General Linguistics. Philadelphia: John Benjamins Publishing Co. 55 Appendix A Online Questionnaire Questions 1. Which data center is your main (or only) character located? a. (List of data centers were available to select from a dropdown menu.) 2. Please select the server on which your main (or only) character is located. a. (List of servers were available to select from a dropdown menu.) 3. Select your main (or only) character’s race and clan from the list below. a. (List of available races and clans were available to select from a dropdown menu.) 4. Please select the gender of your main (or only) character. a. Male b. Female 5. Please check any roles you consider to be your main roles. a. Tank (Warrior, Paladin, Dark Knight) b. Healer (White Mage, Scholar, Astrologian) c. Melee DPS (Monk, Dragoon, Ninja, Samurai) d. Ranged DPS (Bard, Machinist) e. Magical DPS (Black Mage, Summoner, Red Mage) f. I don’t have a job unlocked 56 6. What qualities do you like about the physical appearance of your character? a. (Text input box) 7. What reasons, if any, were there for selecting your race and clan? a. (Text input box) 8. Did the choice of starting area affect the class you selected at the beginning of the game? a. Yes b. No c. I don’t remember 9. What is the main reason you play your favorite roles (tank, healer, dps)? a. (Text input box) 10. Did you choose your main server so you could play with friends? a. Yes b. No c. Unsure 57 11. Do you feel your main server is highly populated? a. Yes b. No c. Unsure 12. Do you consider your main server to be a RP server? a. Yes b. No c. Unsure 13. Are you a member of a Free Company on your main server? a. Yes b. No c. I don’t know what a Free Company is 14. Do you own a house on your main server? a. Yes b. Yes, and I have shared housing with another player c. No d. No, but I have shared housing with another player 58 15. Do you ever explain mechanics of dungeons, trials or raids to other players? a. Yes b. No c. Unsure 16. Do you ever use acronyms such as "lol," "lmao" or other similar acronyms when communicating with other players? a. Yes b. No c. Unsure 17. Do you ever use emotes to communicate with other players? Examples include /hug, /dance, /bow? a. Yes b. No c. Unsure 18. Do you ever use emoticons to communicate with other players? Examples include XD :D ;n; a. Yes b. No c. Unsure 59 19. Do you believe certain races are more likely to sue emotes and emoticons? If no, why not? If yes, which races? a. (Text input box) 20. Do tanks have a common way of communicating with others? For example, do people who play tanks tend to use certain phrases or emotes, or have a similar attitude? a. (Text input box) 21. Do healers have a common way of communicating with others? For example, do people who play healers tend to use certain phrases or emotes, or have a similar attitude? a. (Text input box) 22. Do DPS have a common way of communicating with others? For example, do people who play DPS tend to use certain phrases or emotes, or have a similar attitude? a. (Text input box) 23. Do you feel that your methods of communicating are different or similar to those who mainly play the same role as you? a. Yes, we communicate similarly. b. No, we communicate differently. 60 24. Please tell us about yourself: Sex a. Female b. Male c. Prefer not to say* 25. Please tell us about yourself: Gender a. Female b. Male c. Prefer not to say* 26. Please tell us about yourself: Age a. 18-20 b. 21-25 c. 26-30 d. 31-40 e. 41-50 f. 51-60 g. 61+ h. Prefer not to say 27. Please tell us about yourself: Where do you live? a. (List of countries were available to select from a dropdown menu.) 61 28. Follow-up interview: What is your main character’s name and server? a. (Text input box) 29. Please provide an email address, Discord contact information or other contact method so we may get in touch with you to schedule an interview if necessary. a. (Text input box) 30. What time would be best to reach you for an in-game interview? a. (Text input box) * This option was listed as “prefer not to say” so that transgendered players were protected as they may have provided their in-game name and server. In addition, it was also due to how character gender is presented in the character creation menu, by using “gender” to mean physical appearance. 62 Appendix B Semi-Structured Interview Questions 1. How do you feel about people who use emotes or emoticons in-game? 2. How frequently would you say you use emotes or emoticons? 3. Do you think there is anything that influences whether you use emotes or emoticons? 4. What kinds of people do you feel are most likely to use emotes or emoticons excessively? 5. What kinds of people do you feel are the least likely to use emotes or emoticons? 6. Have you ever found the usage of emotes or emoticons disruptive to your gameplay? 7. There are several emotes that must be purchased before they can be used. Examples include /playdead from the Mog Station and /pushups from the Grand Company vendors. Have you ever purchased one of these emotes? 8. In general, do you feel that your friends within the game use emotes or emoticons around the same frequency that you use them? 9. Outside of the game, such as on forums or other online communities unrelated to Final Fantasy® XIV, do you use emotes or emoticons more or less frequently? 10. Are there situations when using emotes or emoticons may be inappropriate? 63 Appendix C Measurement of Neonateness These calculations for symmetry and neonatal appearance are simplified from Baudouin and Tiberghien (2004) due to the nature of video game polygons versus the appearance of real human beings. Table A-1 is a list of points around the face and their ratios. Ratios are relative to the height and width of the face while the images are the default appearance of the character when entering the character creation screen. Table A-1: Neonatal measurements description. Measurements with points based on Baudouin and Tiberghien (2004) Eye measurements: (Width of eye, height of eye, space between eyes) Nose measurements: (Bridge of nose to tip of nose, left of nose to right of nose, distance between nose and mouth) Mouth measurements: (Center of top lip to center mouth, center of mouth to center bottom lip, left edge of mouth to right edge of mouth) 64 Default Lalafell Male Default Lalafell Female Eye measurements: (0.174, 0.125, 0.326) Eye measurements: (0.178, 0.125, 0.333) Nose measurements: (0.089, 0.087, 0.036) Nose measurements: (0.107, 0.089, 0.036) Mouth measurements: (0.018, 0.018, 0.174) Mouth measurements: (0.0180, 0.018, 0.178) Default Au Ra Male Default Au Ra Female Eye measurements: (0.217, 0.062, 0.239) Eye measurements: (0.200, 0.102, 0.250) Nose measurements: (0.200, 0.174, 0.092) Nose measurements: (0.163, 0.150, 0.082) Mouth measurements: (0.031, 0.046, 0.326) Mouth measurements: (0.041, 0.061, 0.225) Figure A.1: Ratios of key characteristics in male and females of the Lalafell and Au Ra races. FINAL FANTASY XIV © 2010 - 2018 SQUARE ENIX CO., LTD. All Rights Reserved. 65 Default Miqo’te Male Default Miqo’te Female Eye measurements: (0.162, 0.057, 0.297) Eye measurements: (0.200, 0.106, 0.325) Nose measurements: (0.170, 0.189, 0.076) Nose measurements: (0.213, 0.175, 0.085) Mouth measurements: (0.038, 0.038, 0.405) Mouth measurements: (0.043, 0.064, 0.330) Default Hyur Male Default Hyur Female Eye measurements: (0.166, 0.057, 0.268) Eye measurements: (0.216, 0.076, 0.297) Nose measurements: (0.151, 0.244, 0.057) Nose measurements: (0.170, 0.189, 0.076) Mouth measurements: (0.038, 0.057, 0.293) Mouth measurements: (0.038, 0.038, 0.216) Figure A.2: Ratios of key characteristics in male and females of the Miqo’te and Hyur races. FINAL FANTASY XIV © 2010 - 2018 SQUARE ENIX CO., LTD. All Rights Reserved. 66 Default Elezen Male Default Elezen Female Eye measurements: (0.191, 0.056, 0.167) Eye measurements: (0.175, 0.080, 0.275) Nose measurements: (0.222, 0.214, 0.074) Nose measurements: (0.200, 0.200, 0.080) Mouth measurements: (0.037, 0.056, 0.405) Mouth measurements: (0.040, 0.060, 0.275) Default Roegadyn Male Default Roegadyn Female Eye measurements: (0.148, 0.057, 0.333) Eye measurements: (0.220, 0.080, 0.317) Nose measurements: (0.200, 0.278, 0.057) Nose measurements: (0.180, 0.220, 0.120) Mouth measurements: (0.043, 0.043, 0.482) Mouth measurements: (0.060, 0.080, 0.463) Figure A.3: Ratios of key characteristics in male and females of the Elezen and Roegadyn races. FINAL FANTASY XIV © 2010 - 2018 SQUARE ENIX CO., LTD. All Rights Reserved. 67 Average Human Ratio Eye measurement: (0.229, 0.658, 0.251) Nose measurement: (0.210, 0.284, 0.835) Mouth measurement: (0.354, 0.557, 0.476) Figure A.4: Average ratios of key characteristics, measured from the sample provided by Baudouin and Tiberghien (2004). Double features (eyes) were averaged together. Table A-2: Table of the ratios from all races divided by average key characteristics of an average human to create a neonateness index. Males and females of each in-game race were then averaged together. M = Male, F = Female, A = Average. Eye Nose Mouth measurement measurement measurement Lalafell (M) 0.760 0.190 1.299 0.424 0.306 0.043 0.051 0.032 0.366 Lalafell (F) 0.777 0.190 1.327 0.510 0.313 0.043 0.051 0.032 0.374 Lalafell (A) 0.769 0.190 1.313 0.467 0.310 0.043 0.051 0.032 0.370 Au Ra (M) 0.948 0.094 0.143 0.952 0.613 0.110 0.088 0.083 0.685 Au Ra (F) 0.873 0.155 0.996 0.776 0.528 0.098 0.116 0.110 0.473 Au Ra (A) 0.910 0.125 0.570 0.864 0.570 0.104 0.102 0.096 0.579 Hyur (M) 0.725 0.087 1.068 0.719 0.859 0.068 0.107 0.102 0.616 Hyur (F) 0.943 0.116 1.183 0.810 0.665 0.091 0.107 0.068 0.454 Hyur (A) 0.834 0.101 1.125 0.764 0.762 0.080 0.107 0.085 0.535 Elezen (M) 0.834 0.085 0.665 1.057 0.754 0.089 0.105 0.101 0.851 Elezen (F) 0.764 0.122 1.096 0.952 0.704 0.096 0.113 0.108 0.578 Elezen (A) 0.799 0.103 0.880 1.005 0.729 0.092 0.109 0.104 0.714 Miqo’te (M) 0.707 0.087 1.183 0.810 0.665 0.091 0.107 0.068 0.851 Miqo’te (F) 0.873 0.161 1.295 1.014 0.616 0.102 0.121 0.115 0.693 Miqo’te (A) 0.790 0.124 1.239 0.912 0.641 0.096 0.114 0.092 0.772 Roegadyn (M) 2.096 0.087 1.327 0.952 0.979 0.068 0.121 0.772 1.013 Roegadyn (F) 0.961 0.122 1.263 0.857 0.775 0.144 0.169 0.144 0.973 Roegadyn (A) 1.528 0.104 1.295 0.905 0.877 0.106 0.145 0.458 0.993 68 Table A-3: Table of the ratios from all races divided by average key characteristics of an average human to create a neonateness index. Males and females of each in-game race were then averaged together. M = Male, F = Female, A = Average In-game Race Neonatal Average Lalafell 0.394 Au Ra 0.436 Hyur 0.488 Elezen 0.504 Miqo’te 0.531 Roegadyn 0.712 69 Appendix D Final Fantasy XIV Experience Character Creation When creating a character, the default character is a male Midlander Hyur. Each race and clan also has its own default model, which readers are able to see in more detail in figures A-5 through A-10. Please note that players may adjust many features to their liking, from eye shape and facial structure. Also note that, at the time of writing this, there is no slider to gradually exaggerate or soften a body part as is the case in other games. This means that the only available variations are pre-selected by the game developers. Clans are exclusive to a race and cannot be mixed with other races. For example, an Au Ra may only be of the Raen or Xaela clan; it is not possible for a player to create a Wildwood Au Ra, for example. There are many physical differences between the character races and clans. The most common distinctions are the coloration of skin or fur, eye differences and differences in height. Some dimorphism is more explicit, such as the tiny stature of female Au Ras in comparison to the larger male Au Ras, but there are sometimes differences between the same race such as the larger build of Highlander Hyurs compared to Midlander Hyurs. 70 Raen Au Ra Male Raen Au Ra Female Xaela Au Ra Male Xaela Au Ra Female Figure A-5: Visual aid for Au Ra clans, separated by gender. Note the dimorphism, in which the females are significantly smaller in size than their male counterparts. In-game, females are only slightly more than half the size of males. The clans are separated based on the Auri creation myth, in which the Au Ra were created by the Dawn Father and Dusk Mother; the bright scales of the Raen signify relation to the Dawn Father while the darker scales of the Xaela signify relation to the Dusk Mother. In addition, the Xaela are a nomadic tribe while the Raen typically live in solitude. 71 Duskwight Elezen Male Duskwight Elezen Female Wildwood Elezen Male Wildwood Elezen Female Figure A-6: Visual aid for Elezen clans, separated by gender. Females have slightly rounder faces than their male counterparts, but other than that there is minimal dimorphism. Duskwight Elezen are named as such due to their retreat into caves and caverns while the Wildwood Elezen are named as such for their love of forests and plains. 72 Midlander Hyur Male Midlander Hyur Female Highlander Hyur Male Highlander Hyur Female Figure A-7: Visual aid for Hyur clans, separated by gender. The Highlanders, differentiated from the Midlanders by their larger build, arrived on the continent first. Many Highlanders are refugees from a war. Midlanders are noted to be more educated than most other races and clans, learning to read and write from a young age. 73 Plainsfolk Lalafell Male Plainsfolk Lalafell Female Dunesfolk Lalafell Male Dunesfolk Lalafell Female Figure A-8: Visual aid for Lalafell clans, separated by gender. Like the Au Ra, Lalafells differ significantly in culture. Dunesfolk Lalafell tend to passionately pursue knowledge. They were the ones who helped make Ul’dah, a city-state located in the desert, the success that is. In contrast, Plainsfolk prefer the marine city-state of Lima Lominsa and are known for their excessive desire for conversation rather than for a pursuit of anything noteworthy. Due to the harsh sunlight from the Dunesfolk homelands, their eyes are glossier with an extra layer of protection which shields their pupils from the sun. 74 Seekers of the Sun Miqo’te Male Seekers of the Sun Miqo’te Female Keepers of the Moon Miqo’te Male Keepers of the Moon Miqo’te Female Figure A-9: Visual aid for Miqo’te clans, separated by gender. There are two major lifestyle differences between them. The Keepers are nocturnal and offer their piety to Menphina, the goddess of the moon. In contrast, the Seekers are diurnal and offer their piety to Azeyma, goddess of inquiry and keeper of the sun. Like Dunesfolk Lalafell and Duskwight Elezen, Keepers of the Moon Miqo’te have darker coloration and large pupils. Seekers of the Sun have narrow, vertical pupils like a cat and are noted to occasionally be born with heterochromia. 75 Sea Wolves Roegadyn Male Sea Wolves Roegadyn Female Hellsguard Roegadyn Male Hellsguard Roegadyn Female Figure A-10: Visual aid for Roegadyn clans, separated by gender. As one might infer from their clan name, the Sea Wolves typically live their lives on the seas. They were infamous for their piracy many years ago, but in present time most are simply sailors or fishermen. The Hellsguard live their lives in volcanic regions instead, where the little sustenance available encourages the younger children to set out toward cities to live their lives. The Hellsguard protect the volcano, believing it to serve as a gate to the underworld. It is noted that due to their size and strength, Hellsguard Roegadyn are popular hires for heavy lifting and sellswords. 76 On the character creation screen, there is a brief description of clan distinctions on the top-left of the screen. This description offers an explanation of cultural history for each clan and also explicitly tells the player what the dimorphism is between clans. There is also additional information including the starting stats of each race. In the long run, these stats are fairly insignificant but the game does not hint this to the player, thereby offering the possibility of encouraging race and clan selection based on stats rather than appearance. The default gender, clan and race is a male Midlander Hyur meaning that the image is often times the first impression players receive of the available races. The male Midlander Hyur also serves as the “player character” in promotional material related to the game. Roles Roles are a necessity in Final Fantasy® XIV and is not something a player can avoid. Even if a player chooses to never finish their class quest to gain a job (an upgraded version of a class with more abilities), a player must select a class, and thus a role, prior to beginning the game. There are three roles within the game: tanks, which maintain the attention of enemies; damage dealers (DPS), which have access to a plethora of damage dealing spells, actions or abilities; and healers, who heal the party. Players are able to switch between these rules, but only when not inside of a dungeon, trial or raid and only when not in combat. The three roles of tank, damage dealers and healers can be broken down further by job. There are three tank jobs: Paladins, Warriors and Dark Knights. DPS is broken down by types of damage dealers. Monks, Dragoons, Ninja and Samurai make up Melee DPS; Bards and Machinists make up Physical Ranged DPS; and Black Mages, 77 Summoners and Red Mages make up Magical Ranged DPS. The main difference between the jobs within a role is how they relate to the rest of the party. Some jobs are excellent for dealing damage like Samurai, but have few, if any, ways to help the rest of the party through their abilities. In contrast, a Red Mage can deal damage, but can also assist healers by healing party members or reviving party members should they fall in battle. Despite their healing capabilities, their job is still to primarily deal damage. Together, tanks, melee DPS and physical ranged DPS are known as Disciples of War while healers and magical ranged DPS are known as Disciples of Magic. In short, they are the game’s combative roles. There are also two other roles: Disciple of the Hand (DoH) and Disciple of the Land (DoL) which deal with crafting items and gathering items respectively. DoH and DoL are completely optional roles. A player may reach maximum level with all of their combative roles while never learning a single DoH or DoL class. Communities There are several communities players may join, and these different groups serve different purposes. Free Companies are guilds that allow for players to help and learn from one another. They also provide a large group of bodies so that it is easier to find enough people to do content that requires large numbers of people such as 24-man raids. Players are only able to be a member of one Free Company at a time and not all Free Companies are the same. Some focus on progressing though current content while others focus on DoH and DoH roles to make in-game money. 78 Linkshells are similar to Free Companies, but you can join more than one. Like Free Companies, they are also focused on particular aspects of the game such as tracking the location of rare monsters. Lastly, sometimes players form groups known as “statics.” As the name implies, it is a group of individuals, usually eight, with a few extra people as backup. This allows for the same group of players to progress through content together and receive loot at the completion of current content. Members who are part of a static typically do not have to face situations where they need to constantly look for new parties as statics understand multiple runs of content may be necessary. Statics also ensure that everyone progress and learns at approximately the same pace. This is in contrast to pick-up-groups (known as PUGs), where every time a party disbands, players need to find a new party and hope that everyone is aware of how the trial or raid works. Terms and Mechanics The goal of playing Final Fantasy® XIV is different for each player. Some players may want to complete the story known as the Main Scenario Quest (MSQ) within the game’s tutorials. Other players may want to level a particular job to the maximum level and equip the best combat gear. Some players may want to make a lot of in-game currency so that they can buy a house in-game and decorate it as they please. Although the game begins with pointing the player toward the first MSQ, after the first fifteen levels or so it is up to the player to decide what to do. Some content is locked behind MSQ progression and so, for example, new players are unable to immediately access Kugane, a social hub and important location very late in the game. 79 In order to progress the MSQ, players will need to do a combination of completing quests, dungeons and trials. While quests are typically completed alone, players are able to group up with other players to make eliminating enemies an easier task. There are some restrictions on dungeons and trials. Under normal circumstances, a player needs to be in a group of four, including the player. The group must contain a tank to hold the attention of the enemies; a healer to restore the health of injured party members; and two damage dealers, known as DPS to cause enough damage to eliminate the enemy efficiently. In addition, dungeons have a level range. If players exceed the maximum level range, the player’s level will be scaled down to match the dungeon level. This means players cannot use skills learned past the dungeon’s maximum level. However, there are some exceptions such as choosing to run the dungeon unsynced. By running the dungeon unsynced, players can either go through the dungeon alone at a higher level or have another player who is higher level join their party and clear the dungeon for them which is known as a “carry.” When partied together, players are expected to perform their roles. However, that does not mean players cannot perform other roles as well. For example, a healer is expected to restore the health of their party members. However, a healer typically does not need to heal constantly and may decide to use their downtime to damage the enemy instead. Likewise, if a healer has died in combat, some DPS are capable of reviving the healer so they may reenter combat. Tanks are also able to deal damage, although usually not to the same extent as a DPS, whose job is specifically designed for damaging the enemy. Together, the party balances each other out. 80 When players go through a dungeon, there are many approaches. When content is either newly released or new to a particular player, groups may decide to move slower throughout the dungeon, although this is not always the case. The game indicates when a dungeon has not yet been completed by a player in the party, but does not specify which player or take into consideration if the dungeon has been completed by the player on another character. Some groups prefer to charge through the dungeon, gathering as many enemies as possible so that the DPS can efficiently eliminate multiple enemies at once. Within dungeons, there are several dangerous enemies known as bosses. These bosses reside in boss rooms which are clearly marked by glowing purple lines. Crossing the purple line does not necessarily engage the boss. However, some spells have a large range and if the player stands too closely to the boss, they may accidentally engage it. Once the boss is engaged, a fifteen second timer begins. When the time elapses, players are locked into the room and may not leave until either the boss is defeated, the party is defeated, or the player is defeated and chooses to respawn. By respawning, their character’s body leaves the boss room and reappears at the entrance to the dungeon. Depending on how many bosses have been defeated, the player may often utilize shortcuts to return near the boss room. If the boss encounter is still in progress, players who have respawned will not be able to enter the boss room, but can still watch the encounter. It is expected that if a healer dies, a DPS capable of reviving the healer should cease damage long enough in order to revive the healer. If no one is able to revive the healer, the healer may either respawn or choose to watch the rest of the encounter and then respawn. If the tank or any of the DPS are defeated in combat, it is expected that the 81 healer will revive them, although healers must often prioritize when to revive fallen party members. For example, the healer must often decide to revive the tank or one of the DPS. In making this decision, the healer must take into consideration their own chance of survival. Tanks must often decide to use their strong attack combinations or to use their attacks which attract the enemy’s attention, which is known as gaining aggro. Choosing strong attacks over attacks that cause aggro may cause the tank to lose the attention of the enemy, who may then attack and possibly defeat the other party members. Actions a boss or other enemy uses in order to interact with the party are known as mechanics. These mechanics differ from enemy to enemy. Some mechanics include a specific order of attacks, conditions that cause a certain attack to occur or a change in the environment of the boss room. Other mechanics rely on trusting other party members to deal with them. One example is the infliction of a debuff. A debuff affects the player character in a way that interferes with gameplay. A player who is inflicted with “Slow” will, as the name implies, be forced to move slowly. This can be dangerous if a player needs to reach a certain location in time for an upcoming mechanic. It is in cases like these where the player must usually rely on others to help, such as a healer casting an “Esuna” spell on them to remove the effect. There are times when players may remove the debuff themselves, but in general the gameplay encourages players to help one another as combating stronger enemies is easier done with the entire party alive in most cases. In contrast to debuffs, there are also buffs which are tools a player may use to boost some characteristic of their own character or other player characters. A very common spell is “Protect,” a spell restricted to healers. The Protect spell provides a 82 defense bonus to everyone within range when it is cast. Other spells increase speed or absorb a certain percentage of damage. Some of these buffs manifest in different ways. One of these ways is known as a tether. Players are often tethered together when certain buffs are activated. The tether serves as a physical representation of the buff as well as an indicator of whether the players are close enough for the buff to remain in effect. Players who stray too far from one another will break the tether, which in turn breaks the effect of the buff. One final important dungeon mechanic is the Limit Break. It is a standard RPG mechanic in which after enough power is stored, some kind of strong attack can be unleashed. In Final Fantasy® XIV, Limit Breaks have different effects depending on the role of the player who activated it. When activated by a tank, damage taken is reduced for a certain amount of time. When activated by a melee DPS, a strong, single-target attack is unleashed. When activated by a ranged DPS, a potentially multi-target attack is unleashed, although it is not as strong as the melee Limit Break. Because the attack is in a straight line from the player character to the enemy, it may be difficult for the player to have a good enough alignment to damage all enemies. When the Limit Break is activated by a magical DPS, significant damage is inflicted within a certain radius. Both ranged and magical DPS have an indicator of where the damage will occur which is visible to all other players. Limit Breaks activated by the healer have varying effects depending on the percentage of the Limit Break gauge that is filled. When all three bars are completely filled, healers are able to revive their entire party at the expense of being immobile for several seconds. Fallen players must be within a certain radius of the player initiating the 83 Limit Break in order to revive. When one or two bars are filled, the Limit Break simply acts as a healing spell. In the majority of content, players who are grouped together are all real life individuals. That is to say, a group of four players will typically not contain non-playable characters (NPCs), which are characters that are programmed to behave in certain ways. There are some instances in which players may be joined by NPCs, but these instances are made obvious to the player. Sometimes, the player characters are not played by a real person. These characters are known as “bots” and perform actions automatically. Often times, these bots will take on a gathering role to collect materials and sell them on the market board, an object that is available in several city-states which allows players to buy and sell items instantly. The economy of Final Fantasy® XIV can be affected by these bots, who are able to quickly amass materials to sell that real players typically cannot collect as quickly. Bots are not restricted to gathering roles and are capable of going through dungeons with real players. As bots are usually silent in dungeons, some confusion may occur as to whether a party members is being played by a real person or if they are a bot. Finally, players are able to create a “macro” which can have several functions. Players can use it so that they may post several lines of dialogue with a single button press. This can be useful if a player wants to explain the mechanics of a dungeon to the rest of their party. Macros can also be used to display text along with a certain action. This is useful for healers in particular, whose revive spells can overlap one another. By making a macro with the action to revive a targeted player and the text “Reviving 84 produced text to draw players’ attention to the chat box. Macros can be as simple or as complex as players want them to be, and so while some players use them in the above scenarios, other players take it further and can completely automate crafting roles. This means that a player can complete a craft that would take fifty button presses in perhaps five or six macro buttons, eliminating potential mistakes from manual button presses. 85