Dating back to 1975 when Hunt ^[10] firstly proposed the concept of “tourism destination image”, it has received extensive research attention from multiple disciplines for nearly five decades. Most literature focuses on exploring tourists’ perceived image, i.e., the mental construct of tourist’s individual impressions of and ideas about a certain tourism destination ^[11]. A large quantity of work has been accumulated on TDI, including the examination of its conceptualization and dimensions ^{[11][12][13][14]}, the formation process and influencing factors ^{[9][13][15][16][17][18][19][20][21]}, assessments and measurements ^[14][20][22], and its impact on tourists’ decision making and behavior ^{[17][23][24][25][26]}. For example, Kislali et al. divide tourist perceived destination image into the primary and secondary image, with the former referring to a tourist’s destination image after real visitation and the latter referring to the image generated from secondary sources of information beyond one’s own experience ^[9]. The secondary image is further divided into organic and induced images according to whether the source of information is not directly aimed at promoting the destination (organic) or intentionally promoting this destination (induced) ^[15][27][28]. Lai and Li summarized the three classical models of the TDI structure. They are the three-dimensional model (from functional to psychological characteristics, from common to unique image attributes, and from individual attributes to holistic impressions) ^[29][30], the causal-networking model (mainly referring to the ‘‘cognitive-affective-conative model’’ ^[31], the ‘‘cognitive affective-overall model’’ ^[13]), and the core-peripheral model ^[12][32]. Even though there are continuous debates about TDI structuring, another group of scholars contend that TDI should be understood as a holistic construct or gestalt experience ^{[9][33][34][35][36]}.

Besides the tourist’s perceived image, a group of literature distinguishes TDI according to different subjects that give birth to relevant concepts such as residents’ destination image ^[16] and the image projected mainly by DMOs ^[8][37][38]. Before the Web 2.0 era, the distinction between the projected and the perceived TDI is relatively clear, since the expense of publishing the printed brochures and guidebooks or shooting the official destination-marketing advertisements “imparts a certain degree of authority on the marketers’ behalf” ^[1] (p. 222). Following this line of thought, many scholars are examining the impacts of destination marketing strategies or TDI projection on the tourists’ image perception ^[27][39], or comparing the differences between the projected images from the destination supply side and the perceived images from the demand side ^{[3][5][6][8][40]}.

However, in the increasingly developed Web 2.0 world, the boundary between the official projected TDI and tourists’ perceived TDI becomes more and more blurred ^[1][41][42]. The tourism officials and organizations can utilize social media platforms to promote the destination image ^[43]. Individual tourists can also freely express and publish their evaluations of or visiting experiences to the destination, relying on multiple sources of online platforms such as social media and online travel agencies (OTA), creating User-Generated Content (UGC) ^[44]. Recent studies have found the obvious influence of UGC on the formation of perceived TDI among potential tourists ^{[17][45][46][47][48]}. In other words, UGC is increasingly implicated as an important alternative source away from traditional DMOs and destination marketers to project the TDI. Some studies even indicate possible inter-influences between UGC and the DMO-published content in cocreating a successful TDI in this globalized internet environment ^[49][50][51]. Generally, the study of TDI experiences a paradigm shift from a stable and fixed concept to a dynamic and accumulative process under the constructivism perspective, especially after stepping into the Web.2.0 era ^[9][52][53].

In this context, a new question emerges: How do the official DMOs and the tourists influence each other’s TDI projections through their online generated content? Is it possible for DMOs to refer to the tourists’ UGC when exploring a TDI that is attractive and sustainable enough to attract potential tourism markets? What are the crucial influencing factors in this process? The answers to these questions are important to the sustainable development of the tourism destination and would shed light on its marketing and management practices. Currently, relevant diachronic process analysis of the inter-influences between DMO and tourists concerning their online generated content to construct the projected TDI is lacking in literature.

3. The Theoretical Perspective: “Circle of Representation”

It was initially proposed by Urry ^[54][55] in his seminal book The Tourist Gaze, complemented by the work of Hall, Butler and Jenkins ^[56][57][58]. It describes the cyclic process of TDI representation in four stages ^[58]: The first is the stage of image projection. TDIs are projected by mass media collectively. The second is the stage of image perception. The projected TDIs are perceived by individuals and may inspire their visitation to the destination. The third stage is the visitation of tourism icons. At the destination, the tourists will probably visit the tourist icons or sites seen in the projected TDIs. The fourth stage is the reproduction of TDIs. Tourists visiting the tourism icons will photograph their experiences at the site as proof of their visit. These proofs are displayed and spread to others after the trip as another form of image projection. It begins the cycle again by influencing the perceived image of other potential tourists.

Jenkins proved the existence of the cyclic process of TDI representation through an empirical study comparing the images of travel brochures promoting Australia and the pictures taken by the Canadian backpackers to that country ^[58]. Based on his findings, he further conceptualizes “the circle” as an outward spiral anchored to a central tourism icon. Each whirl of the spiral adds new audiences to the TDI, and they may participate in updating the projection of this tourism icon through producing their own insights. In this way, the TDI is continuously in evolution, with circuitous cultural production and reproduction processes.

Jenkins’ thoughts consider the agency of tourist audiences and possible inter-influences between different agents (e.g., DMO, mass media, tourists) to project TDI. Wijngaarden also proves the agency of tourists in TDI (re)production ^[59]. Therefore, Jenkins’ developed model of “circle of representation” is more fitting to explain TDI in the Web 2.0 context, where tourists are empowered to use their agency to influence the TDI projection through UGC. On the other hand, the Web 2.0 era also creates a new environment where this model could be tested in the literal context. Through the internet and different social media sources, tourists have more choices for language use to represent their actual or imaginary impressions of the destination. Besides visual materials, such as photography and short videos, literal texts such as microblog texts, travel notes, travel tips, and comments are also widely used ^[60]. Especially during the recent decade, many studies have proved the function of online literal texts in TDI representation ^{[1][42][50][61]}.