Fandom and identity construction: an analysis of Thai fans’ engagement with Twitter

In our digital era, fandom has become a social and cultural phenomenon, notably in Thailand. Fans are dedicated, and creating fan text (i.e., text production made by fans about their object of fandom) is one way of showing dedication and passion to the fan base. This article explores how Thai fans engage with fan text on popular social media platforms such as Twitter, and how fandom relates to identity construction among Thai fans who are online media users. The results from a selected sample comprising 100 fan tweets from four different sources, suggests five types of fan tweets, including: hypothetical interpretation, fan art, narrative concerning an anecdote regarding the source text, expression of personal opinions and feelings, and fan parody. Moreover, this article discusses fans’ shared lexicon called ‘fan talk,’ and how fans position themselves as relatives and friends of the source texts. This article further discusses the humorous nature and the transcultural elements found in fan tweets, especially the ‘Thai-ifize’ method that fans use in creating fan tweets.


Introduction
G ray et al. (2007) stated that, "most people are fans of something. If not, they are bound to know someone who is" (p. 1). At present, fandom has become a phenomenon that is very influential and global (Black, 2008;Booth and Kelly, 2013). It redefines popular culture as something that is more diverse and complex than the culture of the "mass" (Lewis, 1992), and it reflects a contemporary culture and society, where the Internet is at the heart of communication and consumers now are parts of the production (Gray et al., 2007(Gray et al., , 2017Jeewa and Wade, 2015).
The media industry itself has turned its strategy from driving not only breadth but also depth of engagement, and shifting their target from audiences to fans (Kresnicka, 2016). Moreover, mostly driven by the use of social media, younger generations are increasingly eager to participate in fan cultures; therefore, this phenomenon does not seem like it will subside or end anytime soon. On the contrary, fandom has only grown bigger and is expanding all over the world. There have been many fan conventions held in many countries (i.e. Comic-con, D23 expo, KCON). Many people have been participating in fan activities such as cosplay and cover dance. Numerous websites in different countries provide spaces for fans to write fan fiction in their own languages (i.e. FanFiction.net, Quotev, Wattpad, Asianfanfics, and Dek-d.com).
Moreover, in the present online era, most people tend to use the Internet on a daily basis. It is the same for fans, as they often follow information and communicate with one another on internet platforms like Facebook and Twitter rather than frequently meeting up in real life. Thus, many fan activities have moved from the real world to the online world, for example, fan fiction, games, and group chat. In addition, according to Matusitz (2005), there are more chances for a greater variety of identities to emerge in the online world. As a consequence of these fan activities, recently more and more researchers have turned to explore fandom in online contexts (Baker, 2009;Black, 2008;Booth and Kelly, 2013;Booth, 2008). In other words, understanding fans and their fandom has become the bridge to understanding contemporary life and society driven by the power of online technology.
As a consequence, this study investigates how Thai fans construct their identities mainly through their language used in popular online contexts like Twitter.
Fan discourse and convergence culture. Traditionally, media has been viewed as the shapers of society, in which they have the power to influence people (intentional approach), or has been viewed as the mirror that reflects society (reflective approach) (Hodkinson, 2017). However, to the constructivists, media does not simply shape or reflect society. As Hall (1982) argued "Representation is a very different notion from that of reflection, It implies the active work of selecting and presenting, of structuring and shaping not merely the transmitting of an already-existing meaning, but the more active labor of making things mean" (p. 64, as cited in Croteau and Hoynes, 2003, p. 168). In other words, they give us a selective and manufactured set of representations of the reality of the world. This means that they have power to emphasize or exclude certain contents to the audiences.
Stuart Hall's reception model (1980) is one of the four prominent models of communication (the others include a transmission model, a ritual model, and a publicity model) (as cited in McQuail, 2010). The reception model is rooted in critical theory, semiology, and discourse analysis. Hall challenged the traditional model of passive audiences by arguing that meanings in media texts are always open and polysemic (have multiple meanings). Hall stressed the power of receivers. He stated that receivers interpret media texts according to their context and culture. He called this process "encoding-decoding" (Hall, 1980;Hodkinson, 2017;McQuail, 2010). Meaning is encoded by the sender and decoded by the receiver in which the end results might not be mutual. Furthermore, Hall argues that audiences' decoding of media texts are related to socio-economic context and could be categorized into three types: • The dominant-hegemonic position: This position refers to audience interpretations that are equivalent to the meaning encoded into the media text. The text is assumed to be created by the dominant class and the audience is also a part of that dominant cultural order (has a dominant point of view).
• The negotiated position: This position refers to the acknowledgment of the dominant code in the media text but not entirely accepting it. The audiences accept some preferred meaning and resist what is opposed to their experiences or likings.
• The oppositional position: This position refers to audiences' rejection of the dominant code. The audiences understand the intended meaning encoded in the media text but completely reject it or alter it to suit their experiences.
Moreover, according to Jørgensen and Phillips (2002) "language is structured in patterns or discourses-there is not just one general system of meaning as in Saussurian structuralism but a series of systems or discourses, whereby meanings change from discourse to discourse" (p.12).
"Discourse" is the term that broadly refers to language used in contexts. Its concept goes beyond just linguistics, focusing on how people use language to express themselves or make any action or activity (Stubbs, 1983). Discourse could be varied from the work place, the courtroom, to the Internet. Each discourse is interrelated with and dependant on one another. In discourses, other than informing, language allows people to "do things" and to "be things" as well (Gee, 2011). When we say things, at the same time, we are engaging in action (to promise, to pray, or to perform something) and adopt different identities in society. Identity is an area of reality that people construct when using language.
In addition, Michel Foucault (1972) has played an important role in the field of discourse analysis by connecting discourse to power relation in society. Foucault suggested that power is diffused and embodied in discourse. Discourse is not an exact reality, but rather a culturally constructed representation of it. Each discourse is a representation of a society, situation, or culture. Power circulates through discourses and governs the way we think, talk, and understand. It is framing us into the norms of what is acceptable and unacceptable through the power within political, economic, institutional, and technical conduct (as cited in Jones et al., 2011;Jørgensen and Phillips 2002;Rabinow, 1984;Wandel, 2001;White, 1986). Moreover, these social norms are called "discursive practices" which come from the process of how knowledge (or meaning) is governed by certain structures and (dominant) reality is promoted by the power of discourse or what Foucault called "discursive formation" (Rose, 1998, as cited in Zembylas, 2005Foucault 1972;Mills, 1997). Furthermore, people are always subject to discursive practices. Thus, one's identities are mostly governed by discourses and the discursive practices that they are connected to (Baxter, 2016).
Likewise, the word 'fan' and 'fandom' cannot be understood by just linguistics. As fandom is socially constructed, it can be considered one kind of discourse. In the past, fandom was discursively constructed as cultist and dangerous (Stanfill, 2011).

ARTICLE
HUMANITIES AND SOCIAL SCIENCES COMMUNICATIONS | https://doi.org/10.1057/s41599-020-00653-1 To illustrate, the first wave of fan studies, which began in the late 1980s, was primarily involved with questions of power and representation. According to de Certeau (1988), consumers are not passive. He stated that ordinary people have ways in which they resist the producers of culture or the ruling class in everyday life. Influenced by de Certeau's work, the early scholars of fan studies also viewed fandom as a power-struggle against the media industries.
Fans were originally portrayed as abnormal people who were heavily stereotyped as pathological individuals with mental illness. According to Jensen (1992), fans were constantly characterized as "potential fanatics." Their behaviors were viewed as excessive and almost-insane. As Fiske (1992) stated, fans were then "associated with the cultural tastes of subordinated formations of the people," which were degraded by the standard dominant value system. Arguing against the common myths of what fandom is, first wave scholars described fans as "active" consumers who often responded, retorted, and poached (Gray et al., 2007(Gray et al., , 2017. Despite other pioneer scholars like Michael de Carteau, John Fiske, and Lisa A. Lewis, it was Henry Jenkins who published the first book of fan studies that explored fandom from a cultural studies perspective. Textual Poachers (1992) still heavily influences fan studies research which examine fan communities and fan text production (Coppa, 2014;Booth, 2009). This wave is referred to as "Fandom is Beautiful" by Gray et al. (2007Gray et al. ( , 2017. However, after the power of mainstream media has been reduced by the invention of the Internet, fandom has been reinterpreted as something more positive. Today, with the evolution of technology, new media has altered the circumstances of producers and consumers to a great extent. We are now in an era where old and new media meet, or what Jenkins (2006) refers to as "convergence culture." Convergence culture maps together three concepts of media convergence, participatory culture, and collective intelligence. According to Jenkins, "convergence" refers to, "the flow of content across multiple media platforms, the cooperation between multiple media industries, and the migratory behavior of media audiences who would go almost anywhere in search of the kinds of entertainment experiences they wanted" (p. 2).
"Participatory culture" is a term that Jenkins used to point out the new rules of media producers and consumers. Instead of playing the opposite role of traditional producers and consumers, Jenkins sees individuals as participants who interact with each other "according to a new set of rules that none of us fully understands" (p. 3). At present, convergence culture is being defined in both ways: top-down, and bottom-up. The flow of media content is as much shaped by the decision of teenage consumers as the decisions made by the company board of directors. These two sometimes reinforce each other, creating a more satisfied relationship between the producers and consumers, but sometimes they contradict, causing negotiations of power between them (Jenkins and Deuze, 2008). In addition, while fans may engage in the productive but piracy act, such as writing fan fiction and making fan art, they can also be the most loyal consumers at the same time (Wood, 2013). That is why even though media corporations still hold a lot of power, in convergence culture fans also have some leverage over the production of media content. As a consequence, convergence culture is not defined by merely a shift in technology, but also as the revolution of power relations between media producers and consumers. In addition, "fan media of various kinds is a longstanding aspect of media reception that Media Studies hasn't paid sufficient attention to" (McDougall, 2012, p. 152). It is the rise of the Internet that has made fans become more apparent, and fan studies have become more significant. Fans are one of the best illustrations of the new media users and convergence culture.
In addition, fan discourse is found to influence other discourses as well. According to Berger (2010), fans have long known to influence star discourses, including the influence on star's fame and fortune. Likewise, in political discourse, political supporters and activists have recently been commonly referred to as fans (Gray et al. 2017). These examples illustrate that fan discourses might not be that small or powerless after all.
Fan identities and the Internet. Booth (2010) suggested that, "whatever we are fans of, we base part of our identity on our appreciation of that fandom" (p. 20). Similarly, Sandvoss (2005) stated that a media object is part of the fan's sense of self. In fan studies, identity is and possibly will always be the central focus of research. Some fans use their identity as a way to differentiate themselves from ordinary media audiences. Thus, fan studies help us, "explore some of the key mechanisms through which we interact with the mediated world at the heart of our social, political, and cultural realities and identities" (Gray et al., 2007, p. 10). Furthermore, the more important fans are to media producers, the more fan studies become important to understanding mass media (Booth, 2010).
In addition, fandom studies tend to increasingly focus on fan culture as a space to explore fans' gender and sexual identities. According to Butler's "performative theories of identity" (1990), gender is not innate, but rather emerges from gender identities being repeatedly acted out or performed until it becomes naturalized (as cited in Jeewa and Wade, 2015). When we apply a performative approach to the study of fandom, we could see its social implications clearer in the way that people perform sexual preferences, gender, ethnicity, or even political involvement (de Kloet and Zoonen, 2007).
According to Coppa (2006), the activity of writing fan fiction fits into a performative criteria rather than a literary one. Likewise, Thomas (2006) suggested that fan fiction is a space for fans to self-reflect and explore issues of identity and empowerment. This statement holds true, especially for the slash fan fiction genre.
Another practice of fandom that is connected to the performance of identity is "role playing." In the study of online fan role playing, Jeewa and Wade (2015) mentioned that with the help of the Internet, fans can imagine themselves in roles they want to be in this world through the aid of virtual reality. Also, because these roles are chosen and not imposed, fans speak from their passion which has a value that is very rich and emotional.
As much of the discussion is primarily involved with online culture, it is noteworthy that fandom has only recently moved online. Despite the drastic change, providing fans with multiple platforms for creativity and global communities, fan identities and practices have not changed much from the pre-internet era. Booth and Kelly (2013), in their study of Doctor Who fans, discovered that the Internet has not changed, but it has augmented fan practices and behaviors. Some fans with more financial and social capital continue to prefer offline practices like attending conventions, because they can afford to travel. On the other hand, fans with fewer resources can still be part of the communities through the use of online technology. Online fandom has made fandom more visible as a whole, and thus made it more acceptable as a cultural identity (Booth and Kelly, 2013). Thus, the Internet has also broadened the scope of fandom, challenging researchers to engage in new platforms and new methods of analysis.
Fan text. In opposition to an early belief of passive audience, Stuart Hall (1980) argued that the media audience does not just passively accept the media text, but they interpret it based on their cultural and background knowledge. Through the process of encoding and decoding, the media can encode any meaning they want to convey or influence, but it is up to the audiences on how they decode or interpret that media text, in which the end result might not be mutual. This idea is further developed by fan scholars stating that media audiences, especially fans, not only have the power to negotiate the meaning of media text, but they also have the power to reproduce or redefine it.
Rooted in Stuart Hall's notions of encoding/decoding, fan studies see media audiences as members who engage in different degrees of semiotic productivity, in which they produce meanings and pleasures that pertain to their social situation (Fiske, 1992). Thus, rather than passively taking in media content, media audiences freely shape those meanings which in part are their own (Fiske, 1992;Lee Cooper, 2015;Lindlof et al., 1998). Especially in the age of the World Wide Web, it is hard for media producers to control meanings or uses of their texts in an online context (Soukup, 2006). Sometimes, the media text itself could be altered by fans reconstructions of meaning.
One of the key works of fan studies that explored the cultural dimension involving fans production is Jenkins (1992) Textual poaching, a notion first introduced by Michel de Certeau in The Practice of Everyday Life (1984) and later developed by Henry Jenkins in Textual poachers: Television Fans and Participatory Culture (1992) (as cited in Booth, 2009;Bothe, 2014;Gonzalez, 2016;Jeewa and Wade, 2015). Following Stuart Hall's encoding/ decoding model of communication (1980), De Certeau argues that audiences are not passive consumers but instead active interpreters. He links audience members to poachers by mentioning how they, "move across lands belonging to someone else, like nomads poaching their way across fields they did not write, despoiling the wealth of Egypt to enjoy it for themselves" (de Certeau, 1988, p. 174). Jenkins then applied de Certeau's theory to fan cultures in which the fans "poach" from source texts to create new texts (for example, fan fiction) to fill in further details that did not exist in the original text. Moreover, today the Internet has made fans' practices of textual poaching more widespread and easy to access online.
Fans are dedicated, and creating a fan text is one way of showing their dedication and passion to their fandom. Aside from other negative aspects of their reputation, it is hard to deny that fans are productive and creative. Moreover, in this age of rapid technological change, fan texts are easier to produce, access, and exhibit.
Fans celebrate the "repetition" of their beloved text as a way of producing creative works. Some fans show their love and support for the source text by complementing or extending the storyline (Hellekson and Busse, 2006;Jenkins, 1992). Some fans only borrow their beloved characters and scenes from the source text as a discursive referent in their new art work (Busse, 2017). In other words, fans 'perform' their fan identities through creating fan text.
In addition, according to Bothe (2014), one way that different interpretations of a source text are evident in fandom is through "shipping." It is derived from the word "relationship", which implies the emotional, romantic, or sexual pairing of characters by the fans. Most fan fiction involves "shipping" (Rössler, 2017) between both source text couples (official couples in the source text) and non-source text couples (couples that are unofficially paired by the fans). Through this shipping, not only are heterosexual couples paired, but couples of the same sex are also paired by media fans, especially female fans.
In addition, fan texts in numerous studies include the production of fan fiction as the main focus (Franceschi, 2017;Jenkins, 1992;Storey, 1996). Other subordinate fan texts are songs, poems, visual arts, and videos (Hills, 2013; Storey, 1996).
However, Hills (2013) argued that in the age of the Internet, what could be counted as fan textual productivity could cover a wider range of practices rather than what is traditionally known in this field. As a consequence, online comments, tweets and status could be considered fan textual productions. Nevertheless, Hills was aware that his argument might overly expand the definition of fan textual production in which a fan tweet should not be considered as fan text similar to fan fiction. Fan fiction is an artistic production, while a fan tweet or fan post is not. However, he also stated that online fan productions appear in a wide range of forms and genres. To illustrate, some fan textual productions preserve the old esthetics of traditional hand-drawn fan art, while later versions include new esthetics of photo-editing. In this sense, we argue that the varieties of fan textual productivity keep expanding as technology progresses, and the notion of fan text should not be fixated to any particular kind, quality, or genre. Therefore, in this study, a fan tweet is considered one type of fan text, which is created in popular online platforms like Twitter. Many people engage in creating fan tweets because it is easy to write and be seen by others. However, fan tweets are not widely explored in fan studies as much as fan fiction, despite its popularity. Thus, this study aims to broaden the scope of fan text and explore other types of fan text than fan fiction, namely fan tweets.
Fandom in Asia and Thai context. Although fandom is a universal phenomenon, it is also local. Because fan practices are not the same in all places, it is vital to take into consideration characteristics of each culture. At present, there are far more researches on western fan culture than Asian fan culture, especially in English language (Chin and Morimoto, 2013).
Most Asian fan studies, especially in the East Asian context, focus on fans of Japanese popular culture and Korean popular culture (Chin, 2007;Chua and Iwabuchi, 2008;Williams and Ho, 2016). Moreover, unlike the Western fan studies which focus on the power resistance of mainstream media by fans and the promotion of participatory culture, Yano (2004( , as cited in Chin, 2007 proposed that East Asian fandom is tied to the concept of 'intimacy' between fans and their object of fandom, which "impels individuals to act in ways that go beyond the bounds of self to seek greater communion with the object of their adoration" (p. 44).
Likewise, in fan studies, 'transculturalism' is a concept derived from Matt Hill's work of 'transcultural homology' on Western and Japanese otaku fans (Chin and Morimoto, 2013). Chin and Morimoto (2013) further employed this concept to explore why fandoms cross borders where there are social and cultural differences. They stated that the concept of transculturalism frees fandom from "the constraints of national belonging, reinforcing our contention that fans become fans of border-crossing texts or objects not necessarily because of where they are produced, but because they may recognize a subjective moment of affinity regardless of origin" (p. 99). In other words, fans became fans mostly because of the moment of affinity rather than the cultural or national differences or similarities between fans and the media object.
For fan studies in a Thai context, despite the numerous and still growing fan practices and fan communities in Thailand, there has not been much research in the field of fandom in this particular context. In addition, those studies often focus on sport fans or would be analyzed through the marketing viewpoint.
According to Kaewthep et al. (2012), the first stage of the music fandom phenomenon in Thailand, which was intensified by the growth of cultural industries around 1960, was fans of popular Thai singers such as the fans of Thongchai McIntyre, famously known as 'Bird.' Later, the influence of J-pop (Japanese pop) and K-pop (Korean pop) media became popular in Thailand and created the fandom of these two countries. Thus, many fan studies in a Thai context at that time focused on these two fandoms. Nevertheless, Thai fan studies switched back to Thai fandom as reality shows such as 'Academy Fantasia' became popular.
However, at present, the focus of Thai fan studies changed back to mainly K-pop fandom. The global impact of South Korean popular culture began in the late 1990s (Jung, 2010). K-pop achieves its popularity through a highly rationalized industry via music, television, and other entertainment channels (Williams and Ho, 2016). Especially in Thailand, the influence of K-pop is evident in popular fan practices like cover dance and fan fiction.
'Boomcassiopeia', one of the Thai pioneer Kpop (Korean pop) fans, mentioned that Kpop fan communities in Thailand have grown very fast. He said that it started out as a small group. He did not know that there were also other fans like himself. However, now they can come together and everyone can see how big it is with the help of social media (Panyalimpanun, 2017). This shows how fandom has also become a major social phenomenon happening in Thailand and seems likely to continue in the future. Its rising significance is indicated in the growing number of fan sites, fan pages, and fan events, suggesting that Thai fans are one of the most dedicated groups of fans in the world. Nevertheless, the phenomenon has barely been explored in the Thai context, unlike in western cultures where there has been research interest in this topic for over a decade. Therefore, this study can serve as an additional insight into fan studies in Asian cultures, and more specifically in Thailand.
As a consequence, two research questions were asked. First, how is the source text portrayed and rewritten in fan tweets? Second, how does fandom relate to identity construction among Thai fan online media users?
The first question explores how Thai fans engage with fan text in popular online platforms such as Twitter. The second question explores Thai fans' identity construction through their creation of these fan tweets.
Furthermore, even though there are other kinds of fans (e.g. sport fans and brand fans), this study only focuses on media fans. Nevertheless, media fandom is somehow entangled with other fandoms, as media, both mainstream and online, could be said to also be a part of many cultural objects and the acceleration of their fandom.

Methods and analytical framework
Methods. The unit of analysis is obtained from fan tweets, which is a kind of fan post on the social media platform Twitter. In Thailand, this platform is very popular and presently has more than a million users (Statista.com). Twitter provides space for people to express their thoughts through tweets, where fans can communicate and follow one another via the Internet. Fan tweets include both linguistic texts and pictures, some of which could be considered fan art. Thus, it could be said that Twitter is a site that provides a rich source of fan production.
Four popular media objects in Thailand were selected as source texts, namely a Korean boy band 'Wanna One', a Thai popular drama 'Buppae Sunniwas', a popular Chinese survival show 'Idol Produce', and Marvel's The Avengers. These media objects originated from different fandom cultures, namely Korean fandom, Thai fandom, Chinese fandom, and Western fandom. The mix of fandom cultures is due to the intention to include diversity of data. The criterion for a Twitter fan tweet is the recentness and popularity of the tweet. Twitter can track recent trends by using Hashtags (#) and also by following related fan topics.
To answer research question one, content analysis was used as an analytical tool. According to Krippendorff (2004), "content analysis is an empirically grounded method, exploratory in process, and predictive or inferential in intent" (p. xvii). Content analysis is used as a method to systematically transcribe and organize a large amount of text into a 'concise summary of key results' (Erlingsson and Brysiewicz, 2017). Content, in this case, "can be anything produced by people for various purposes, usually not originally intended for research" (Lune and Berg, 2017, p. 182) and content analysis examines a discourse by looking at patterns of language used in the social and cultural contexts. The followings are general steps of conducting content analysis: a. Data is collected and made into text or otherwise organized to be "read" (e.g., field notes, transcripts, image sequences, news reports). b. Codes are analytically developed and/or inductively identified in the data and affixed to sets of notes or transcript pages. c. Codes are transformed into categorical labels or themes. d. Materials are sorted by these categories, identifying similar phrases, patterns, relationships, and commonalties or disparities. e. Sorted materials are examined to isolate meaningful patterns and processes. f. Identified patterns are considered in light of previous research and theories, and a small set of generalizations is established (Lune and Berg, 2017, p.184).
According to Berg (1983, as cited in Lune and Berg, 2017), the content could be divided into seven major elements or units to be coded, namely words or terms, themes, characters, paragraphs, items, concepts, and semantics. This study uses the elements of concepts and themes as units. Concepts are units of conceptual clusters of words in relation to certain ideas. For example, words such as love, like, and my love are clustered around the conceptual idea of affection toward fans' object of fandom. Themes are units of shared underlying meaning, or latent content, which could be expressed in words, phrases, sentences, or paragraphs. After defining the units of analysis, the coding categories were constructed by using an inductive approach, or using the data as basis for developing meaningful and relevant categories and sub-categories.
In addition, the size of the unit of analysis is 100 tweets (25 fan tweets per each source text). According to Bengtsson (2016), "there are no established criteria when using content analysis for the size of a unit of analysis, neither the number of informants or objects to study, nor the number of pages based on the informants' own written text or transcribed data" (p. 10). Thus, it is up to the researcher as to whether he/she feels that the size of a unit of analysis is appropriate for answering the research question.
Analytical framework. To answer research question two, semiotic analysis is used as analytical framework in exploring identities in narrative texts such as fan tweets. Modern semiotics suggested that there are connections between signs and identities. According to Berger (2012), objects and artifacts, such as hairstyles and clothing are signs intended to convey certain meanings of what a person is like. He also stated that semiotics concerns itself with how meaning is generated in texts, through the use of signs, especially in narratives or stories. As a result, many researchers have adopted this approach to their study of identities (Martinec, 2000;Matusitz, 2005;Pollock, 1995;Singh and Kumar, 2016).
Roland Barthes (1915Barthes ( -1980 was one of the contemporary semioticians who turned from structuralism to post-structuralism (Berger, 2010;Posner, 2011). Barthes rejected the idea of preexisting sign systems and suggested discourse as a source of meaning comprehension (Posner, 2011). Barthes (1972, as cited in Berger, 2012 introduced two systems of meaning, "denotation" and "connotation". While denotation refers to the literal meaning of words or phenomena, connotation refers to the cultural meanings and myths that are attached to words or to things. To illustrate, the sign Rolls-Royce would denote a particular kind of car. On the other hand, from our social experience, we know that Rolls-Royce cars are expensive. Thus, it connotes wealth and luxury (Bignell, 2002). To Barthes, this social phenomenon is the making of "Myth", or the shaping of people's ways of thinking.
Likewise, Bakhtin (1981, as cited in Berger, 2012;Nayar, 2010) suggested that language is "dialogic" meaning that what we say is connected to things that have been said before and that is expected to be made in the future. Built on Bakhtin's dialogism, Julia Kristeva coined the term "intertexuality", indicating that "one particular type of context which may influence the reading of a discourse consists of previous discourses" (Posner, 2011, p. 22). In other words, the creation of text (linguistics and nonlinguistics) is influenced by other previous texts (Berger, 2012;Elkad-Lehman and Greensfeld, 2011;Worton and Still, 1991). Furthermore, as people read, listen, or see texts, it is interwoven with their past experiences with other texts. Thus, this makes one's interpretation of text intertextual. In addition, influenced by Kristeva's work, French theorist, Roland Barthes also suggested that literary work, in its nature, is intertextual. Thus, readers could find multiple meanings in a single work. In this case, the "author" has no authority or power over the reader, which Barthes considered "dead" (Allen, 2011;Nayar, 2010).
Kristeva's notion of intertexuality goes together with the nature of fan texts. As a consequence, intertexuality has been one of the important parts of fan studies and has been mentioned or adopted by many researchers in their studies of fandom (Busse, 2017;Gray et al., 2017;Jenkins, 1992;Piper, 2015;Wilson, 2016).
In addition, Singh and Kumar (2016) stated that literature with multicultural elements reflects the world's countless voices and perspectives. Narratives/story telling does not merely build "identity" and "multiculture," but it also helps audiences to accept these two terms by putting oneself in the shoes of others in an imaginative way. Likewise, it can be said that cultures are interwoven within the production of fan texts. As fandom is a global phenomenon (Gray et al., 2007(Gray et al., , 2017, it is built with multiple cultures. As fans are always involved in creating, sending, receiving, and interpreting fan texts with multicultural elements, Berger (1995) suggested that "what semiotics and semiology do is provide us with more refined and sophisticated ways of interpreting these messages-and of sending them. In particular, they provide us with methods of analyzing texts in cultures and cultures as texts" (p. 100). Therefore, a semiotic analysis is considered suitable and useful analytical frameworks for investigating the online identities of Thai fans in this study.

Data analysis and results
Fans' portrayal and rewriting of source texts in fan tweet. To address the issue of the portrayal and rewritten source text in fan tweets, 100 fan tweets were collected and analyzed by using content analysis as a tool. Top tweets of four source texts: The Avengers, Idol Producer, Wanna One, and Buppae Sunniwas were retrieved on 13 June 2018 from Twitter.com. There was a total of 100 tweets, 25 tweets per source text.
To fully understand fan texts, four selected source texts' official information are presented below. The final 9 formed as a group called Nine Percent, which will be active for 18 months. Afterward the group will be disbanded and the members will return to their respective companies ("Idol Producer"; "Nine Percent").

d. The Avengers
Marvels' The Avengers (2012) is a very popular and successful superhero movie. It is based on superheroes from Marvel Comics of the same name and now includes four sequels. The Avengers is one of the films in 'Marvel Cinematic Universe' (MCU) which is a media franchise of a shared universe including films, television series, etc. The franchise is produced by Marvel Studios and distributed by Walt Disney Studios Motion Pictures, which started with the movie Iron Man (2008) as its first plotline. The storyline in each MCU story is somewhat complex and intertwined with one another. The last movie, Avengers: Endgame currently is the highest grossing movie of all time. The other sequels are ranked as followed: Avengers: Infinity War, ranked number five; Marvel's The Avengers, ranked number eight; Avengers: Age of Ultron, ranked number 10. Despite many storylines in-between the franchise, the main plotline is the Avengers saving the universe from supervillain Thanos (Marvel cinematic universe wiki.).
Twitter provides a filter that can track recent trends by using Hashtags (#). Therefore, the selected top tweets were the most recent and popular tweets at that time.
Two source texts, The Avengers and Idol producer, were searched by the name of its' own title, #Avengers and #Idolproducer. The other source text, Wanna One, was searched by the name of the source texts' two main pairings, #Linhoon and #Nielong. Lastly, Buppae Sunniwas was searched by the name of both main pairings and its title, #เดื ่ อ เกด (#Duerkade) and # บุ พเพสั นนิ วาส (#Buppaesunniwas).
The data collected of fan tweets could be categorized into five types: hypothetical interpretation, fan arts, narrative of a personal anecdote regarding the source text, expression of personal opinions and feelings, and fan parody.
Hypothetical interpretation. There are 25 fan tweets in this category. The tweets are mainly about the hypothetical interpretation which means making hypothesis or assumption of the elements in the source text. In other words, fans were cracking the hidden messages in the source text photos or videos. The interpretation is often about 'unspoken' romantic relationships of a certain source text's characters or personalities. All of these fan tweets come with a hashtag (#) of the pairing of unofficial couples by fans. For example, the Korean boy band Wanna One's unofficial pairing includes #Linhoon, referring to the pairing of Lai Kuanlin and Park Jihoon. Fans took parts of their names, Lin from Kuanlin and hoon from Jihoon, and combined them together, similar to slash fiction's Kirk/Spock.
The tweets in this category tend to establish the possibilities of their unofficial pairing by pointing out the sometimes obvious sometimes subtle evidence of their 'unspoken' affection in the source text photos or videos. To illustrate, an example of @tweet no. 1 tweeted: The tweet is attached with a video clip of Kuanlin hugging Jihoon in a television show in the background (see Fig. 1). As in reality, we do not know whether the couple is really in a secret relationship or if they are just friends. Whether it was actually a romantic gesture or just a friendly hug, this fan pointed out that she saw a romantic relationship between the two idols, even though they were trying to hide it. Thus, this tweet could be considered a hypothetical interpretation of @tweet no.1.
Another tweet is from @tweet no. This tweet started out by stating that he/she did not speak from the shipper's (fans who 'ship' a certain couple of the source text) point of view, or try to view it in romantic way. Nevertheless, @tweet no. 2 concluded that it is obvious that Kuanlin and Jihoon cannot be anything else but a couple. Similar to the previous example, we could assume that @tweet no. 2 was making an assumption that the two idols are in a relationship based on how they treated each other.
Another example is from the other pairing of Wanna One, Kang Daniel and Ong Seongwu, by @tweet no. 3. Although this tweet was not trying to obviously prove the relationship of the pairing, it hinted at their sexual identity and pairing position (see Fig. 2 The tweet is attached with a picture of Kang Daniel and Ong Seongwu doing the same dance choreography, but Seungwu (right) seems a little bit more feminine than Daniel (left). The word 'King' and 'Queen' in this context have the connotation of 'masculine gay' and 'feminine gay' which hinted not only the idols' sexual identity, but also the pairing position. It is obvious that @tweet no. 3 assumed that Seongwu is the queen, taking the similar position of a female, because of his more feminine gestures, while Daniel is the King who takes the position of a male.
Fan art. The second type is the creation of fan art. Fan art is the unofficial artwork created by fans of their object of fandom, including traditional hand-drawn drawings, digital art, crafts, etc. There are eleven pieces of fan art included in the selected fan tweets. These fan art pieces include traditional hand-drawn drawings and paintings and edited photos. From the data collected, the fan art are drawings with verbal descriptions, which is sometimes comprehensible to only fans. Moreover, some fan art portray shipping couples such as Daniel and Ong Seongwu, Lai Kuanlin and Park Jihoon, and Captain America and Black Widow.
Narrative of a personal anecdote regarding the source text. This type of fan tweets is statements or narrative of an anecdote regarding the source text. This includes users own experiences toward the media objects and the media objects' pictures or videos they saw. The intention of this type of tweet is likely to share information, stories, or their experiences with other fans. Also, this type of fan tweets mostly includes some specific fan words that are often understood within the fan communities. As a consequence, it could somehow automatically seclude these narratives or experiences from non-fans.
For example, @tweet no. 4 tweeted: This fan tweeted about his/her experience with a 'non-fan,' or in this case, the teacher. While @tweet no. 4 was making a reference to one specific memorable scene in The Avengers: Infinity War, supposedly making a humorous excuse for his/her absence. @tweet no. 4 quoted the line "I don't feel so good" from the scene which Spider-man said this before he vanished. Without enough knowledge of the movie, the Filipino teacher was a little confused, but then moved on.
Expression of personal opinions and feelings. Most of the selected fan tweets (43 tweets out of 100 tweets) are expressions of emotions and personal opinions toward source text pictures, videos, or personal fan experiences. Unlike narratives of anecdotes, this type of tweet offers insight into the fans' emotion and feeling towards the source text. Fans gave more personal thoughts or opinions, and mostly with more intense emotional expression, including love, lust, curiosity, and desperation. For example, @tweet no. 5 tweeted: 100ธอร หล อ หล อโคตรๆ หล อเหี ้ ยๆ #CaptainAmerica #Thor #TonyStark #Spiderman #BlackPanther #DoctorStrange #Avengers #InfinityWarTH #AvengersInfinityWar [Review #Infinity War 1. Tom is cute 2. Captain is handsome 3. Doctor Strange is handsome 4. Tony is handsome 5. Your highness is handsome 6. Thor is handsome 7. Thor is handsome … 100. Thor is handsome. So damn handsome, fucking handsome. #CaptainAmerica #Thor #TonyStark #Spiderman #BlackPanther #DoctorStrange #Avengers #Infi-nityWarTH #AvengersInfinityWar] This tweet is a personal review of, obviously not the content of the movie, but the male characters' appearances as @tweet no. 5 used only 'handsome' multiple times to describe them. It seems that @tweet no. 5 was biased toward Thor, as his name was mentioned more times than others and used vulgar words like 'damn' and 'fucking' to show a degree that is more than handsome. This favoritism might come from @tweet no. 5's own preference of beauty standard. This example is also created for a humorous purpose by using the method of exaggeration.
Another example is from @tweet no. 6: According to the tweet, @tweet no. 6 is a fan fiction writer stating that he/she wants to write new fan fiction about a Thai important historical figure, Phraya Tak or King Taksin the Great. However, he/she also expresses concern as most Thais believe that it is disrespectful to represent royal figures as characters in fiction, and what he/she is writing now is risky (because Duer was also King Sanphet VIII of Ayutthaya). @tweet no. 6 used words like 'hell is calling,' which is a Thai idiom that means karma from doing sinful things, and 'welcome to avejee' which means hell. This fan exaggerated the Thai belief of heaven and hell to create humor.
Fan parody. The last type of fan tweets is fan parody, with a total of 7 tweets. A fan parody is a fan's intention to express their ideas and opinions in a humorous and mostly exaggerated way about the source text content through visual and verbal means, including characters, plotlines, or even merchandise. The selected fan parody tweets include both their own creations and sharing of others' fan parody.
For example, @tweet no. 7 tweeted: This tweet is a parody of The Avengers characters, Thor, who is the God of Thunder, being compared to Toothless, a dragon from the animation movie How to Train Your Dragon (see Fig. 3). Moreover, 'little bear' comes from the nickname that Thai fans called Chris Hemsworth, the actor who plays Thor, as his appearance is similar to a bear. @tweet no. 7 added the word 'little' to make it sound more adorable.
In conclusion, the result suggested that fan tweets are more of a reinterpretation or a restatement rather than a re-writing or replotting of the source text. Most fans were trying to voice their stories, experiences, emotions, and interpretation about the source text through various types of text such as words, images, drawings, and videos. They also used specific fandom words and source text references. Moreover, in terms of writing style, most fan tweets used a large quantity of Internet language, namely slang, symbols, emoticons, emoji, stickers, and hashtags (#).
Fans' identity construction in relation to their fan text creation Fan talk: the shared lexicon. The result suggests one way that fans distinguish themselves from non-fans and identify themselves with other fans is through a shared lexicon called 'fan talk.' Fan talk includes 'special codes' and particular communication styles that are shared within the fandom to create the feeling of fellowship among fans and also alienation from non-fans. In other words, fan talk is one of the ways that fans hint about their fan identity with one another.
There is no dictionary, notes, or handbook for this kind of code and communication style. It is rather gradually built-up lexicon that has been created through different events that happened in their fandom. In order to understand the words, fans have to understand those "events" as well. Thus, it could be said, the longer they stay in the fandom or the more effort they put in (i.e., researching), the more codes they know and the more they become a part of their fandom.
One of the examples mentioned earlier is the nickname 'bear' of Chris Hemsworth, or the actor who plays Thor. The word 'bear' in this context could be comprehended in mainly two ways, first is Chris's nickname, and second is a wild animal, which is not quite relevant to the context. Thus, the word 'bear' would make sense for fans of Chris Hemsworth more than non-fans. Moreover, this nickname is mostly used only by female fans of Chris Hemsworth in Thailand. This indicates that these codes might vary according to, not only region, but also gender.
The second example is shown in the position in the pairing after a hashtag (#). The name that goes first represents the dominant position and the latter is submissive. To illustrate, #Nielong is an unofficial pairing of Wanna One, which Kang Daniel is in the dominant position and Ong Seongwu is in the submissive position. Therefore, #Nielong is not the same as #Ongniel, meaning that even though some fans 'ship' the same pairing, they might favor the position of that pairing differently. This kind of word might be incomprehensible to non-fans, so the meaning could vary according to something else or could appear as random words with no meaning at all.
Another example is a particular communication style of Buppae Sunniwas fans. This communication style is not like a fan special code, but it is somehow not common in everyday speech. This style comes from the attempt to imitate the source text's interaction, which is situated in a historical period. Thus, this communication style turns out to be a combination of both modern and old Thai language. For example, fans used the word 'เจ าคะ' (jâoká), which is an ancient word, after questions. Moreover, they used 'ออเจ า' (ɔɔjâo) which is also an ancient word that was very iconic in the drama, instead of other modern Thai words referring to 'you.' On the contrary, as these fan talks and communication styles could be considered a way of expressing group identity, it is simultaneously a way of differentiating themselves from non-fans. As a result, fan talk is a method that creates both a bond within the group and a boundary that separates them from the world and makes it difficult for outsiders to identify with.
Fans as relatives and friends. In term of fans' relationship with their object of fandom, many of them often positioned themselves as a 'family' or a 'friend.' For example, many fans used Thai personal pronouns for friends and family members, including 'พี ่ ' (phi/ senior), 'น อง' (nong/ junior), 'เด็ ก' (dek/ kids), 'ลู ก'(luk/ daughter or son), etc. to refer to the media personalities of the source text. To illustrate, some Idol Producer fans called Cai Xukun 'พี ่ คุ น' (phi-kun) and The Avengers fans called Tom Holland, the actor who plays Spider-man, 'น องทอม' (nong-tom). Despite the anonymity in online text, the age of the writers could be roughly assumed by the pronouns they used to refer to their source texts. Moreover, this kind of positioning seems to be a way for fans to express fellowship and intimacy between fans and their source texts, which are greater than being just a fan.
The multicultural elements. The results suggest that Thai fans obtained some cultural aspects from the source text, namely the source text language. On the other hand, they also integrate their own culture into their creation of fan tweets.   Another example from @tweet no. 9 includes Chinese language, which is the source text language of Idol Producer. # 范丞丞 reads Fan Chengcheng.

เคย
In addition, obtaining source text cultural aspects does not only refer to national culture. This also includes the culture of the specific period which is presented in the source text. To illustrate, Buppae Sunniwas fans often used words and communication style of the source text, which previously appeared particularly in the Ayutthaya period, in their fan texts.
On the contrary, Thai fans also integrate their own cultural identities into their creation of fan tweets, mostly in a parody style.
For example, @tweet no. 10 shared a link from a Facebook fan page called 'วิ วาทะใต หน ากาก' which means 'speech under the mask.' The tweet presented the Avengers characters' names, which were changed to funny outdated Thai names but still resemble the original pronunciation, for humorous purposes. To illustrate, the name of Steve Rogers or Captain American was altered to 'สั นติ ' (săn-dtì) which has a similar pronunciation as 'Steve' but is considered mundane name in Thailand.
Another example is from @tweet no. 11, an Avengers fan who tweeted: This tweet is a parody of The Avengers characters, Thor and Loki (see Fig. 4). According to Thai culture, people usually pay respect to a statue of worship by hanging flowers or wreaths on it. In this tweet model, figures of Thor and Loki were equivalent to statues of worship in a mocking way. This is not only a parody of The Avengers characters, but also of Thai beliefs in gods and holy spirits. By combining the object of fandom and Thai culture in an extreme and mismatched way, this fan created humor according to his/her own interpretation and cultural background, in which only people with similar cultural background would understand the joke.

Discussion
Twitter can be considered one of the most popular social media platforms in Thailand. A fan tweet is a post on Twitter by fans regarding their beloved media object. The evidence of many fan interpretations of source text could be seen in the selected fan tweets. Answering the first research question, the result shows that there are five ways Thai fans engaged with fan tweets: hypothetical interpretation, fan art, narrative of an anecdote regarding the source text, expression of personal opinions and feelings, and fan parody. The type that fans mostly engaged with is 'expressions of emotions and personal opinions toward source text pictures, videos, or personal fan experiences.' This shows that twitter could be considered a platform that resembles a bulletin board for fans to express themselves toward their beloved media object rather than a community. In other words, fans used twitter to voice their stories, experiences, emotions, and interpretation about the source text through various types of text such as words, images, drawings, and videos.
Thus, fan tweets are not that similar to fan fiction. Fan fictions are often a re-writing or re-plotting of source text, while fan tweets referred to a real situation involving the source text or the official plotline of the source text, which could be considered more of a re-statement of the source text according to fans' own interpretations.
Moreover, fan tweets are argued to be characterized by humor. The results show that most fan tweets were written in a humorous tone. It is not clear whether the humorous factor used heavily in this type of fan text relates to the nature of Twitter text, fan culture in Twitter, Thai culture, or all of the above. It is found that even fan art in the selected fan tweets have a humorous tone.
Furthermore, regarding research question two, the data collected shows that there is evidence of heteronormativity, or the belief of heterosexuality as the norm, especially in the homosexual relationships. Heteronormativity suggests that people fall into two specific biological binary genders, male and female, and an appropriate pairing in terms of marriage (Francis, 2012). In most cultures, this norm is constituted by gender order or "the patterns of power relations between men and women that shape norms for femininity and masculinity by defining what is genderappropriate in arenas, such as romantic partner selection, occupational choice, and parental roles" (Schilt and Westbrook, 2009).
The evidence of heteronormativity is shown in the 'shipping' position of the pairing of the homosexual couple. To illustrate, fans put a male idol whose appearances and characteristics 'look' more masculine and powerful in the 'dominant' position, and put another male idol whose appearances and characteristics look more feminine in the 'submissive' position. This could be seen in one of the examples mentioned earlier of Kang Daniel and Ong Seongwu's 'king' and 'queen' position. We do not know for sure whether these fans actually believe in these pairings or not, but we could assume that they are 'performing' and playing with gender identities through their beloved idols. Fans depict their idols as characters, imagining and molding them into the images they desired. Nevertheless, the heteronormativity in their homosexual pairing could signify that Thai fans are still influenced by or adhere to typical society's gender order.
In addition, the results of the selected fan tweets indicate the use of fan talk, which requires source text knowledge to be able to completely understand the meaning of it. As the Internet continues to blur the line between media users and fans, the fans and the fan communities are trying to distinguish themselves from others through 'fan talk.' In the online world where there is no face-to-face interaction and no clues from fan accessories, language and sign is the only way to express and detect one's fan identity. Fans create a shared lexicon called 'fan talk,' in which it creates both feeling of fellowship and alienation. This fan talk includes specific vocabularies and communication styles of certain fandom, which more often require some knowledge about the source text in order to crack the code. Thus, it would be difficult for non-fans or even casual fans to interpret a fan tweet.
Furthermore, this 'fan talk' consists of various cultural elements including fan culture or the source text, group culture or the fan community, and the national culture or the country fans live in. In other words, the intertextuality of fan talk could vary between fans of the same source text but of different fan communities or different nationalities.
In this study, three of the selected source texts originated from various cultures other than Thai, but are very popular in Thailand. Likewise, many fan scholars have explored this kind of cross-border fandom, especially between Eastern and Western culture. Yano (2004( , as cited in Chin, 2007 proposed that East Asian fandom is tied to the concept of 'intimacy' rather than fan resistance proposed by the Western fan scholars. This feeling of intimacy comes from the relationship between fans and the star or the idol rather than a specific cultural text (i.e., film or genre). Chin (2007) also pointed out that it is not to state that this factor shows the difference between the two sides, but that the national and cultural structure of the industries could influence how fans perform their identities differently. Similarly, the result from the fan tweets of Thai fans shows intimacy and affection more than a resistance to the power of mainstream media.
In addition, in their creation of fan tweets, there are mainly two ways that Thai fans handled the differences of their own culture and the source text. First, Thai fans obtained some of the source text's cultural elements and integrated it in their fan tweets. Second, Thai fans also integrated their own cultural elements with the source text in their fan texts. This result indicates the crossing of boundaries between cultures and its' complex connection, which is parallel with the notion of 'transculturalism.' According to Welsch (1999), "cultures today are in general characterized by hybridization. For every culture, all other cultures have tendencially come to be inner-content or satellites. This applies on the levels of population, merchandise, and information" (p. 199). He suggested that, in this era of the Internet and global networks, cultures are entangled, intermixing, and promote exchange and interaction. He also noted that we are cultural hybrids and our cultural formation is transcultural.
The result suggests that some fans included Thai cultures into their fan tweets. This kind of 'Thai-ifize' method, or method of merging the source text into Thai context, is highlighted mainly by the use of locality and humor. The use of locality includes, changing or inserting of local linguistic structures, culture, beliefs, norms, and values. The other is the use of humor. Most Thais would agree that humor is one characteristic of Thai culture (Chatvikanet, 2010;Zoom, 2018). According to Chatvikanet (2010), humor has been the basis of Thainess or our 'structure of feeling' for a long time. Therefore, Thais include humor in many aspects of life, such as, work, ceremony, and entertainment, which is different from the 'structure of feeling' of Western culture. Thai fans know how to use humor with local elements to create fan texts that are unique in a Thai way.
Moreover, the result also shows that transculturalism does not only apply in the culture of various places, but also of different times. One example is how Buppae Sunniwas fans used specific words and communication style of the Ayutthaya period. These words and communication style are outdated in this modern era and unfamiliar to Thais. However, Buppae Sunniwas fans adopted this Ayutthaya period's culture in their fan tweets, making it transcultural across time as well.
Therefore, fan texts could not be solely characterized or owned by a particular source text, a particular nationality, or a particular culture. Fan text is transcultural. Fan identity is also transcultural. Moreover, we believe that a person's identity is transcultural in some ways, either he/she is a fan or not. However, this study shows that fandom could be one of the major parts of tranculturalism, which is one of the reasons of hybridity in fans' identities.
In addition, the reasons for fan tweets' popularity are still not clear. Nevertheless, from the humorous nature of fan tweets, the expressive feature, and the space for identity play, it seems like one of the reasons for its popularity could be that fan tweets is an easier way that fans could integrate themselves with their source text than other types of fan texts. Fan tweets are short, direct, and fans can make as many as they desire.

Conclusion
This study had proposed two research questions regarding Thai fandom. The first question asks 'How is the source text portrayed and rewritten in fan tweets?' Research question one explores how Thai fans engage with (in other words, portray and rewrite) fan text in popular social media platforms such as Twitter. The results suggest a reinterpretation or a restatement rather than a rewriting or re-plotting of the source text. These results include hypothetical interpretation, fan art, narrative of an anecdote regarding the source text, expression of personal opinions and feelings, and fan parody. The second question asks 'how does fandom relate to identity construction among Thai fan online media users?' The results present mainly four topics: the hint of heteronormativity, the use of fan talk, how fans position themselves, and the multicultural elements found in fan tweets. The paper further discusses the characteristics of fan tweets, including the humorous nature and transculturalism aspects.
Furthermore, the study of fans and social media not only contributes to this specific field of fandom, but also in larger fields such as the study of media and society. This study has, more or less, expanded some knowledge of social media users and audiences who called themselves a 'fan' in this late era of modernity, especially in the Asian context. Furthermore, this study has investigated fan tweets, which are not often mentioned in fan studies but are as popular as any other types of fan texts and how fans engage with them. Likewise, this study has shed light on the connection between local and global fandoms, stressing the transcultural elements within fans' productions and fans' identities.

Ethical approval
All procedures performed in studies involving human participants were in accordance with the ethical standards of the institutional and/or national research committee and with the 1964 Helsinki Declaration and its later amendments or comparable ethical standards.

Data availability
The datasets generated during and/or analyzed during the current study are not publicly available due to restrictions, e.g. data include original works and photos, but are available from the corresponding author on reasonable request.