Curiosity as Brazilian tourist motivation in visiting Europe

Although the theme of push and pull motivations has received increasing attention in tourist behavior literature, little attention has been devoted to the investigation of the interaction between single push motivations and visitor loyalty and other relevant variables influencing tourist behavior. Given its undoubtable relevance in motivating human behavior, we propose curiosity as a single push motive by examining its causal relationships with destination attributes (evaluated in holistic way), attitude toward destination, and loyalty. In particular, we tested a new research model on a sample of 273 potential Brazilian travelers to Europe by using a structural equation modeling approach. Sample size is in line with the state-of-theart in literature (Ciasullo et al., 2017). The data moderately well fitted the “curiosity model” and the findings highlighted that curiosity plays a crucial role in shaping attitude and pull motivation, and in influencing tourist loyalty. Consequently, destination managers or European Union institutions should magnify the role of curiosity, attitude towards destination, and pull motivations in terms of marketing policies.


Resumen
Si bien el tema de las motivaciones push and pull ha sido cada vez más considerado en la literatura sobre el comportamiento de los turistas, se ha prestado poca atención a la investigación de la interacción entre las motivaciones de un impulso único con la lealtad de los visitantes y otras variables relevantes que influyen en sus comportamientos. Dada su indudable relevancia en motivar el comportamiento humano, proponemos la curiosidad como

INTRODUCTION
The World Travel & Tourism Council's (WTTC) annual report indicates that the growth of the travel and tourism sector in 2015 (2.8%) overtook that one of the global economies (2.3%) for the fifth successive year, generating 9.8% of global GDP and supporting 284 million jobs. Similarly, despite many challenges faced by travel and tourism in Europe starting from the end of 2015 (e.g., terrorist attacks, the economic crisis, Brexit, etc.), the sector is still expected to grow by 3.1%, confirming tourism as one of the services industries remarkably resilient in times of economic recession (WTTC, 2016).
Among various foreign destinations, Europe is the continent with the highest tourism demand (Sheth, 2011), especially regarding emerging countries and, particularly, South America. Departures from South America to Europe, in fact, amount to 26%, followed by 23% in North America, 32% in other South-American nations and 19% in the rest of the world. Particularly, the most likely to visit Europe are, above all, Brazilians (Euromonitor International, 2012).
In the current hyper-competition among tourist destinations, a thorough analysis of tourist motivation and its relationships with loyalty and attitude toward destination is crucial for developing adequate policies able to sustain tourism flow within the destination. Especially, research on tourist's decision examining the behavior of emerging markets travelers attending Europe (such as Brazilians) could represent an interesting marketing challenge, since it can contribute to increase loyalty, intercept new tourism segment, and design adequate tourism policies in line with a sustainable vision.
Scholars (Crompton, 1979;Dann, 1977;Uysal & Jurowsky, 1994) demonstrated that tourists travel because they are "pushed" to adopt a specific behavior toward the destination by their psychological factors; at the same time, they are "pulled" within the destination by its characteristics. An analysis of literature data shows the dichotomy push-pull motives in explaining tourist behavior has been generally accepted (Chen & Chen, 2015;Prayag & Ryan, 2011;Yiamjanya & Wongleedee, 2014;Yoon & Uysal, 2005). Thus, the adoption of push and pull framework requires a simultaneous analysis of both visitors' internal desires and core destination attributes (Caber & Albayrak, 2016).
Although numerous studies have investigated push and pull factors holistically and globally, very few have performed an analysis of push (or pull) motivation impact on tourist behavior individually. Moreover, previous studies on tourist behavior have rarely analyzed the relationship between the level of the visitor inner cognitive stimulation (i.e., cultural knowledge gap) and consumer behavior (Botti et al., 2015a;2015b). Among the numerous push factors capable of explaining either the creation of a specific feeling (i.e., attitude), the adoption of a precise conduct or the relation with other reasons for an individual interest in traveling (i.e., pull motivations), undoubtedly one of the possibilities may be represented by curiosity and curiosity knowledge gap. In fact, psychologists underlined that, when we feel curious to discover something, we also have feelings for engaging with novel stimuli or adopting a certain behavior (Kashdan et al., 2009, Dalli 2015. More clearly, it has been well pointed out that "It is hard to deny the power of curiosity as a force for motivating human behavior" (Hardy et al., 2017, p. 230).
To address the aforementioned gaps, the goal of this study is to investigate the causal relationships among curiosity (a well-outlined push motivation), destination attributes (pull motivations), attitude toward destination (a well-known way of thinking that affects a person's behavior) and loyalty. In particular, we developed a new research model (the Curiosity model of Tourist Behavior -CTB) to examine the relationships among these constructs in the context of potential Brazilian travelers to Europe, by conducting a survey and analyzing data using Structural Equation Modeling (SEM).
The objective of this research is twofold: first, we propose a theoretical advance in tourist motivation research, starting from which future in-depth research stream with curiosity as centerpiece could be explored. Second, the study helps destination marketing organizations of European Union or European Country to effectively use curiosity in their marketing campaigns. Practically, the study summarizes the impact of curiosity on the most relevant destination attributes (evaluated in holistic way as recently proposed by Leong et al., 2015), on attitude toward destination, and on behavioral intention.

Tourist Motivation
Motivation, understood as an altered state leading to behavior directed toward a specific goal, represents a widely debated topic in the tourism literature (Su et al., 2018;Fieger, Prayag & Bruwer, 2019) and, specifically, in marketing studies, since the 1940s (Albayrak & Caber, 2018). It consists of needs, feelings, and desires driving people to a certain behavior. Motivation is the starting point for consumers' decision process and an important construct for understanding tourist behavior (Pereira & Gosling, 2019).
In tourism research, motivation is an important area of study because it represents a fundamental construct for understanding tourist behavior, being at the basis of its decision-making process. Furthermore, motivation is an important predictor in evaluating tourists' attitude (Huang & Hsu, 2009;Lee, 2009a). According to Murray (1964, p. 7), a motive is "an internal factor that arouses, directs, and integrates a person's behavior". A definition of motivation in the tourism and travel context was offered by Dann (1981, p. 205): "a meaningful state of mind which adequately disposes an actor or group of actors to travel and which is subsequently interpretable by others as a valid explanation for such a decision".
Studies on tourist motivation provide various frameworks and scale for measuring motivation. In this regard, Valls et al. (2018) point out that the influence exerted by psychological factors on tourists when choosing a destination has long been studied and acknowledged in literature. Consistently, Park et al. (2019) highlight that the analysis of tourist motivation is fundamental for the understanding, explanation, and conceptualization of travel behaviors, since travel motivation influences tourists' attitudes, perceptions, and involvements. Plog (1974Plog ( , 2001 proposed an allocentric/psychocentric model which explains why different people tend to travel to different destinations; allocentric people are venturesome and self-assured, while psychocentrics have some common personality tendencies (such as territory boundedness, generalized anxieties, and sense of powerlessness) (Hsu & Huang, 2008). Iso-Ahola (1982), Mannell and Iso-Ahola (1987) proposed a social psychological model of tourism motivation based on escape-seeking dichotomy (Matheson et al., 2014).
Based on Maslow's (1970) hierarchy of needs, Pearce (1988) developed the Travel career ladder model. The main argument of this model is that human needs tend to ascend higher levels of the career ladder as they keep on doing more and more travel experiences being motivated by sophisticated factors.
In studies on tourist motivation, the push and pull framework elaborated first by Dann (1977;1981) and then extended by Crompton (1979), perhaps represent the most widely accepted paradigm (Jang et al., 2009;Jang & Cai, 2002;Kim & Lee, 2002;Kim et al., 2003;Prayag & Hosany, 2014) for understanding tourists' needs and willingness to enjoy. Push factors reveal the psychological factors of behavior (Wu & Pearce, 2014) such as the desire to escape from everyday environment, adventure, relaxation, and prestige. Push factors are the reasons for and direction of behavior (Iso-Ahola, 1982). Pull factors include specific destination features and attributes influencing when, where, and how people travel (Mill & Morrison, 2002;Prayag & Ryan, 2011). In this regard, pull factors stimulate consumer to travel and represent tourist's generic desire to travel (Yang et al., 2011).
Research in tourism has used the push-pull paradigm for three main purposes. The first one is to explore personal motivations that direct people towards specific behaviors. In this context, some studies attempt to clarify the motivational differences in relation to demographics (Kim et al., 2003). The second one is market segmentation (Frochot & Morrison, 2001) in which the most implemented criteria are the following: segmenting tourists from a specific source market, tourists to a specific destination, tourists traveling for a specific product within a destination, or any combination of the three ways mentioned. Finally, researchers have investigated the relationships between motivations and satisfactions (Huang et al., 2014;Yoon & Uysal, 2005). In particular, Yoon and Uysal (2005) found that tourist satisfaction in turn connected to loyalty, is directly related to authentic experiences.
Literature review on tourist motivation in an emerging market is still in its infancy. In fact, while western tourists' destination perceptions of western destinations are well researched (e.g., Beerli & Martin, 2004;Chi & Qu, 2008;Prayag & Hosany, 2014), travel motivations and perceptions of tourists from emerging markets toward western destinations is fairly recent (Li & Stepchenkova, 2012;Ryan & Mo, 2002).
2.2 Curiosity as push motive Litman and Spielberger (2003, p. 75) define curiosity "as a desire for acquiring new knowledge and new sensory experience that motivates exploratory behavior". Basing on previous Berlyne's work (1954Berlyne's work ( , 1960, Voss and Keller (1983, p. 17) similarly stated that "curiosity is a motivational prerequisite for exploratory behavior". Daniel Berlyne (1954Berlyne ( , 1960, representing perhaps the most authoritative mentor of exploratory behavior, in fact, distinguished between two types of curiosity (perceptual and epistemic) and two types of exploratory behavior (diversive and specific). Recently, knowledge has been strictly linked with curiosity: Loewenstein (1994) argued that exploratory behavior would increase when manageable levels of a knowledge gap existed. Menon and Soman (2002) more clearly evidenced that knowledge gap indicates a difference between what people know and what they want to know. Although the theoretical foundations of curiosity here highlighted are not always invoked, numerous studies use the push and pull framework considering among push motivations curiosity, knowledge gap, novelty or need for cognition (e.g., Wong et al., 2013;Chen & Chen, 2015;Bansal & Eisel, 2004;Wang et al., 2016). In other words, many tourist studies discuss motivations to travel adopting different terms that seem to bring back to the conceptualization of curiosity as outlined so far. For instance, Kim and Lee (2002), analyzing motivations in attending festivals, enumerate five categories including "curiosity". These results are in line with the previous work of Scott (1996) that propose "curiosity" as one of the motives that push toward festival: the study demonstrates that curiosity discriminates among first-time and repeat visitors. Dunn Ross and Iso-Ahola (1991), evaluating motivation dimensions of a sightseeing tour, employ a "general knowledge" motive adopting items as "To see the famous sites" or "To visit the places I especially want to see". Similarly, Cha et al. (1995), studying the motivations of Japanese overseas travelers, include "knowledge" intending for that the need of experiencing a foreign destination or traveling to historical places, the willingness to see as much as possible and to learn new things. Xu et al. (2013), using a qualitative method to explore the motivation of tourist players, identify six factors (namely, curiosity, exploring the destination, socialization, fun and fantasy experience, challenge, and achievement), putting at the basis of the motivational pyramid only curiosity. Particularly, curiosity emerged as the most popular factor: in their study, several groups of respondents mentioned curiosity as their first motivation that influences what they could do in the destination in terms of shopping, food, etc.
Recently, curiosity has been analyzed for its impact on sport consumer behavior  and, more generally, on purchase motivation (Hill et al., 2016).
Although several studies include curiosity among the push motivation and curiosity has been investigated individually in terms of influence on consumer loyalty, to the best knowledge of the authors there are no studies focused on the relationship between curiosity and consumer behavior in the tourism field.

Destination attributes as pull motive
Pull motivations are considered as external, situational, or cognitive drivers influencing consumer behavior (Dann, 1981;Devesa et al., 2010), deriving from the perception of destination characteristics. Since this kind of construct is strongly correlated not only with the destination type (big city, historical town, sun and beach village, etc.) but also with the specific destination, authors tend to generate highly detailed list of the specific destination attributes (Table 1). Consequently, a general theoretical construct of pull motivation is still lacking.  Sung et al. (2015) Attitude and quality of service Without the aim to propose an exhaustive categorization of pull motivations but willing to avoid high bias that could over-shadow the focus of the study, we tried to select the most relevant. In fact, in investigating traveler behavior literature, some common macro-areas of pull motivation emerge.
The construct "pull motives" was measured through different studies, differentiating from each other via discriminant validity. To do that, we have assessed the construct validity by estimating a confirmatory factorial model (Wong & Cheung, 2005). Moreover, further preliminary indications for the purification of the measurement scale by performing the exploratory factorial analysis: factor loadings, extracted variance, possible factorial structure between the several dimensions. Exploratory factorial analysis was applied to analyze the relationships between observed variables to identify a latent structure. The objective was to summarize a number "m" of items in "n" factors (or components), with m>n. In fact, in the process of developing the measurement scale, the exploratory factorial analysis allows to have a first estimate of factor loadings and to verify the opportunity for further purification of the scale (Tan, 2001). The application of the EFA made possible to assess the factorial structure of interest, purifying the measurement scales, and excluding the indicators with low factor loadings on the expected factor or substantial cross loading. In this way, it was possible to obtain a thrifty structure to be submitted to the confirmatory test.
In particular, four categories of destination attractiveness can be distinguished: 1) a qualitative dimension (regarding the service offering proposed by accommodations); 2) a cultural dimension (related to the relevance of cultural heritage in choosing a destination); 3) a leisure activities dimension (namely, nightlife and entertainment, shopping and how to spend free time during destination staying); 4) accessibility and transportation destination convenience dimension (Yoon & Uysal, 2005;Prayag & Ryan, 2011).
In detail, the first category refers to a set of service infrastructure offered by a destination (accommodations, food, shopping, recreation) influencing consumer decision-making process (Ritchie & Crouch, 2003). Particularly, studying the Mauritian case, Kassean and Gassita (2013) find that accommodation services are the main factors leading tourist to visit the destination.
The second dimension, on the other hand, represents cultural, historical, and natural resources which increase destination attractiveness (Casarin & Iasevoli, 2012). Moreover, several studies stress the relevance of cultural motivations mechanisms in influencing tourist's needs, wants, and preferences. For instance, in a study on Korean national parks, Jeong (1997) reveals that visitors perceived natural, historical, and cultural resources as the most important attractions. They represent and constitute a key driver for fostering the peculiarity of each single park, enhancing its identity. Similarly, Kim et al. (2003) confirm that cultural and historical resources drive visitor's decision, whereas Yoon and Uysal (2005), analyzing tourist pull motivations, focus on customer's willingness to experience a "different culture".
Regarding the third dimension (nightlife and entertainment), a fragmentation seems to emerge. Although, nightlife seems to be one of the most common sub-dimensions adopted in defining pull-motives of this category, Yoon and Uysal (2005) determining pull motivation attracting tourists to Northern Cyprus, adopt the unified dimension of "nightlife and local cuisine", whereas  refer only to entertainment. In any case, a common reference to nightlife, entertainment and how to spend free time, seem to characterize the most relevant literature.
Finally, the fourth pull motivational factor representing a common thread among the different works on traveler motivation is "accessibility to the destination". Kim et al. (2003) find a significant correlation between "accessibility and transportation" and push motivations, particularly regarding "natural resources and health" dimension, showing that good destination accessibility influences visitors' willingness to experience nature. Among the others, Sung et al. (2015) identify a cluster of travelers whose primary travel motivations are convenience and ease of travel.
The four categories of destination attractiveness, their abbreviations and the references used to elaborate measurement items are shown in Table 2.

Attitude toward destination and loyalty
Tourist attitude qualifies the psychological orientation expressed by the favorable or unfavorable evaluation of tourists when engaged in certain behaviors (Ajzen, 1991;Schiffman & Kanuk, 2004, Lee, 2009aSparks, 2007). It has been underlined that attitude toward destination is characterized by three components: cognitive, affective, and behavioral (Unger & Wandersman, 1985;Vincent & Thompson, 2002). The cognitive component refers to the way in which the attitude is forming; the affective component reflects the psychological traits in terms of tourist preference; the behavioral component captures the tourist intention.
Given its relevance, attitude toward destination could be significantly predictive of tourist loyalty. In fact, it has been maintained that tourist attitude is an effective predictor of tourist's decision to travel to a certain destination (Ragheb & Tate, 1993;Jalilvand & Samiei, 2012).
Loyalty has been conceptualized by one of the three main approaches, such as behavioral, attitudinal, and composite (Jacoby & Chestnut, 1978). The behavioral approach is based on the analysis of the purchase process or purchase volume and by using repeat visit as a measurement indicator. This approach has been criticized for its inability of explaining the factors affecting customer loyalty (Yoon & Uysal, 2005). Many empirical studies demonstrated that behavioral intention, rather than actual behavior, is an effective indicator of loyalty (Horng et al., 2011;Kaplanidou & Gibson, 2010). The attitudinal approach estimates tourist revisit intention to a destination or recommendation to other potential tourists. Studies have established that a positive correlation exists between tourists' intention to recommend and image components of destination, including overall image (Bigné et al., 2001), affective image (Lee et al., 2005), and cognitive image (McDowall & Ma, 2010). The composite approach advances the integration of both behavioral and attitudinal approaches (Backman & Crompton, 1991;Iwaskaki & Havitz, 1998). In this regard, tourists who demonstrate behavioral loyalty toward specific destinations tend to have a positive attitude toward those destinations (Zhang et al., 2014). Studies found that a positive attitude toward destination leads to higher level of composite loyalty demonstrated by tourists (Bosque & Martín, 2008;Lee, 2009a) and affects future tourists' behavior (Lee, 2009b).

THE PROPOSED MODEL AND RESEARCH HYPOTHESES
The aim of this study is to investigate the causal relationship between a single push-motive (namely, curiosity), pull-motives, attitude toward the destination, and loyalty. The proposed model is represented in Figure  1. Pull motivations are measured as a second order factor assessed by four specific destination attributes, which are combined to propose an integrated and holistic dimension, as suggested by Leong et al. (2015).

Source: The authors
The study hypothesized that curiosity (push motive) positively influences the holistic destination attributes (pull motive)-more precisely, a combination of night life and entertainment (NL & E), accommodation quality (AQ), accessibility to the destination (AD), and relevance of cultural heritage (CH) sub-dimensions-attitude toward destination and loyalty. Moreover, both curiosity and pull motivations affect loyalty, and pull motivation influences attitude. Table 3 summarizes the hypotheses formulated and reports a selection of references supporting such hypotheses as derived from literature review and from the discussion held in previous paragraphs.  The method used in this research contemplates a pen-and-paper survey. A first group of subjects has been personally contacted in some of the main relevant Brazilian educational institutions by three field researchers from September 5, 2016 to September 12, 2016.
From the various Brazilian cities, the survey was administered in Rio de Janeiro for several reasons. First, it is the second-most populous municipality in Brazil and the sixth-most populous in the Americas. Secondly, in line with the economic development of the entire nation, Rio de Janeiro middle class has expanded considerably over the last ten years, determining an increase in gross domestic product (UNWTO Tourism Highlights 2015). It follows that the emerging middle class, characterized by high spending power, represents the driving force of outbound tourism demand and of economy, in general. Brazilian tourists' arrivals in Europe are continuously growing (European Travel Commission, 2015).
Moreover, the research has been conducted in different universities and cultural public institutions. The involved institutions are the Cultural Institute of the Italian Consulate in Rio de Janeiro and several Brazilian Universities, such as: UNISUAM (Centro Universitário Augusto Motta); Estácio de Sá; Federal University of Rio de Janeiro (UFRJ). Finally, the survey has been administered to the students and professors of SUESC School (Sociedade Unificada de Ensino Superior e Cultura), to the students and managers of Maestro Lorenzo Fernandes School and to the employees of Brazilian satellite television "Nossa Tv".
In total, the researchers contacted 290 visitors by adopting a convenience sampling approach: however, 17 questionnaires were incomplete and, therefore, eliminated from the study. It follows that 273 questionnaires were accepted for the final analysis, with a response rate of 93.7%.

Data analysis
All constructs in this study were measured with multiple items, as recommended by Churchill (1979) and Kline (2005). A preliminary list of measuring items was generated after an extensive review of the literature on the push-pull framework, including the conceptualizations of attitude and loyalty. The questionnaire was elaborated in Italian and then translated into Portuguese. A pre-test was conducted with 10 graduate students majoring in Economy at the UFRJ. Items identified as ambiguous were reformulated for more clarity. The final list of measurement items, presented in Table 2, was adapted from previous studies. For all constructs, a seven-point Likert scale was adopted, ranging from 1 ("strongly disagree") to 7 ("strongly agree").
To empirically validate the proposed research hypotheses, the technique of structural equation modeling (SEM) was employed using AMOS 22.0 and the maximum likelihood method of estimation. Structural equation modeling is commonly adopted in tourism marketing literature in general (Lee et al., 2004;Bosque & Martin, 2008;Nowacki, 2009) and specifically in tourism studies (Chi and Qu, 2008;Wang et al., 2016;Yoon & Uysal, 2005). This technique allows to statistically test multiple relationships among variables measured with multiple items. Differently from multiple regressions, it simultaneously allows estimating the relationship between multiple dependent and independent variables not observed (Gefen et al., 2000).
According to the procedure recommended by Anderson and Gerbing (1988) a two-stage testing has been adopted. In the first step, confirmatory factor analysis (CFA) has been used to estimate the measurement model and, in the second stage, concerning the assessment of the structural model, the hypothetical relationships among all the variables have been identified.
Regarding model fit, chi-square, Root Mean Square Error of Approximation (RMSEA), and Standardized Root Mean Square Residual (SRMR) have been measured to understand how the research model fits the data without comparisons with other models. Moreover, Comparative Fit Index (CFI) has been used to reveal how the research model fits the data comparing it to null model, which hypothesizes that all variables are uncorrelated. Scholars highlight that when the CFI index exceeds 0.9 and RMSEA and SRMR indices do not exceed 0.08, adequate fit has been achieved, while, when CFI exceeds 0.95 and RMSEA does not exceed 0.06, the proposed model is acceptable (Hair et al., 2010;Hooper et al., 2008).

Demographic characteristics of respondents
Concerning the demographic characteristics of the respondents, as shown in Table 4, the sample is composed of 141 females (51.8%) and 131 males (48.2%) and it is made up of more married (51.2%) than single (48.8%) travelers. The majority of the subjects is among 26 and 35 years old (33.3%) and among 19 and 25 years old (20.6%).
With reference to occupation, most respondents belong to the middle class: civil servants, in fact, represent 40.9% of the sample. The second most common category is that of students (22.8%), followed by businessmen, showing that the sample has an acceptable heterogeneity in terms of social classes. Besides, consumers in the sample have a high educational level, since half of the respondents hold a bachelor's degree (50.8%) and 38.3% have completed a postgraduate degree (see following Table 4).

Measurement validity
The measurement model derived from CFA reveals satisfactory levels for all fit indices with χ 2 /df equal to 1.869, SRMR equal to 0.061, CFI equal to 0.954, and RMSEA equal to 0.054 (p-close = 0.099).
Additionally, all the constructs demonstrate adequate psychometric properties of measurements and show high Composite Reliability coefficients above the cut-off point of 0.7; these results show a high level of reliability for each construct (Nunnally & Bernstein, 1994). Similarly, as indicated in Table 5, all average variance extracted (AVE) values for the multi-item scales are above the minimum levels of 0.5 (Hair et al., 2010), indicating an acceptable level of convergent validity for all the proposed constructs (Garbarino & Johnson, 1999). The AVE of each construct is greater than the variance shared between that construct and the other ones in the model, which demonstrates satisfactory discriminant validity. In fact, to confirm discriminant validity, the square roots of AVE have been calculated. Table 6 lists the correlation matrix for all first-order constructs. Diagonals are the square root of AVEs. In all cases, the square root of AVE for each construct is larger than the correlation of that construct with all the other constructs in the model, which indicates satisfactory discriminant validity (Fornell & Larcker, 1981).

Hypothesis testing and discussion
In the light of the satisfying results of the measurement model, to test the overall relationships among constructs, the structural model has been assessed.
First, the structural model shows adequate levels for all fit indices with χ 2 /df equal to 1.869, SRMR equal to 0.0617, CFI equal to 0.954 and RMSEA equal to 0.057 (p-close = 0.099). All indices reveal that we have a good structural model.
Second, the estimated results of the proposed research model highlight that five of six proposed hypotheses are supported (see Table 7). Research results related to H1, which states that individual's curiosity positively influences the individual's attitude toward destination, is significant (path coefficient = 0.32; p<0.001). Furthermore, individual's curiosity positively influences the individual's response to the holistic destination attributes (path coefficient = 0.358; p<0.001) and, additionally, individual's curiosity positively influences the individual's future visit intention and willingness to recommend (path coefficient = 0.392; p<0.001). Thus, H1, H2, and H3 strongly and highly demonstrate that curiosity has: (1) a crucial role in amplifying tourists' loyalty and consequentially word of mouth and revisiting intention; (2) a strong impact in shaping visitors' attitude toward the destination; (3) a robust influence on the motives by which a tourist is pulled to the destination. In terms of policy implications, these results underline that for tourism organizations it is fundamental to satisfy need of cognition and travelers' curiosity prior to the departure for the selected destination. In other terms, the purposive feeding of destination information represents a key-factor in developing and maintaining high levels of curiosity, for instance, through the Internet and well-designed and well-organized institutional destination web sites. Surprisingly, the holistic destination attributes (pull-motives) are not statistically significant in predicting individual's future visit intention and willingness to recommend (loyalty), being the p-value higher than 0.05. In any case, standardized coefficient would have been negative. Although not significant, the negative relationship indicates that the holistic destination attributes are not sufficient on a stand-alone basis, to create loyalty behavior. Thus, H4 is not supported. The result is in line with Yoon and Uysal (2005). In the research context of this study, such a result is likely connected to the current situation of European context. Besides terrorist attack, European institutions (European Parliament, European Commission, etc.) are now experiencing a moment of great difficulty because of the nationalist forces in many countries (e.g. Brexit, etc.): nowadays it is very difficult to program any kind of policy in any economic (and not only) field (namely, transportation, cultural heritage, tourism, etc.). In addition, a repeatedly emphasized bureaucracy prevails in the European Institutions. Consequently, in absence of a broader European tourism policy, visitors outside the European Union do not perceive Europe as a unique destination, reducing the impact of destination attributes on willingness to recommend and revisiting intention.
The last observation is in line with the H5, which is statistically supported: individual's response to the holistic destination attributes positively influences the individual's attitude toward destination (path coefficient = 0.159; p-value<0.05). In fact, no doubt that Europe (as a destination) has many features and attributes that are a result of its history, of the tourism policies that individual countries composing European Union have arranged in the past and keep on arranging todays. Thus, in the absence of a European Union tourism policy, the attributes of Europe as destination fail to affect loyalty or, at least, minimally impact on visitors' attitude toward Europe as destination.
Finally, individual's attitude toward destination positively produces a significant and positive effect on individual's future visit intention and willingness to recommend (path coefficient .324; p<.001), supporting H6. This finding is perfectly in line with previous tourism studies, which outline that attitude influences willingness to recommend (Lee et al., 2008), confirming the link between attitude and loyalty in destination management. This emphasizes the critical role of tourist attitude toward destination in mediating relationships among the destination attributes and tourist loyalty and between curiosity and tourist loyalty. Thus, it is still critical for European Union institutions or for the destination management organizations of each individual European Country to promote a positive attitude among travelers' potential referent groups, specifically, toward travel agents in Brazil. In other words, organizations or institutions should emphasize the uniqueness of Europe for history, cultural, and natural heritage, for representing a melting pot of people with their own traditions (Bertoli & Resciniti, 2013). These social and cultural environments could be considered as unique and appealing to the Brazilian travelers, who could be pulled to the destination to satisfy their curiosity on Roman and Greek civilization, on German organization, on their own Portuguese origin, etc.
The analysis of squared multiple correlation (SMC) values offers additional information allowing further discussions consistent with the previous assessment. In particular, the proposed model explains a substantial amount of the variance of loyalty as the squared multiple correlation (SMC) reveals (SMC = 0.313) and explains only a partially acceptable amount of the variance of attitude toward destination (SMC = 0.164). Finally, curiosity, the single push motive, explains only 12.8% (SMC = 0.128) of the variance in destination pull motives. These results are consistent with the multidimensional nature of motivation (Reiss, 2012), confirming that willingness to recommend and revisiting intention and attitude toward destination are the consequence of different motives through which curiosity plays a crucial role.
The measurement of destination attributes as a holistic phenomenon pursued by means of the second-order factor requires discussion here. As pointed out by the SMC values, the accessibility to the destination (AD) is the main component (SMC = 0.860) of the overall pull motive followed by accommodation quality (AQ) (SMC = 0.568). This finding suggests that even though Europe is famous for its culture, (both cultural and natural) heritage, cities, and the possibility to walk through historical city-centers, the ability of European institutions to facilitate the arrangement of direct and relaxed flights or to facilitate the way by which reaching and traveling within Europe is crucial, taking into account the dominant role of these two dominant pull forces. Europe pull motives perform probably just a role of mediation between curiosity and attitude toward destination. In this direction, policy makers should address strategic policy initiatives for valuing cultural heritage, for instance stimulating travelers visiting by offering bundling services in an ecosystems perspective of value cocreation (Barile & Polese, 2010;Barile et al., 2012;Barile et al., 2014;Pencarelli & Splendiani 2008;Tommasetti et al., 2017;Wieland et al., 2012).

CONCLUSION, LIMITS AND FUTURE RESEARCH
The purpose of this study was to investigate the causal relationships between curiosity, destination attributes, attitude toward destination, and loyalty in the context of potential Brazilian visitors willing to travel to Europe. The originality of the work should be seen in the mediation perspective of the model that, as we opportunely suggested, is still not explored in literature.
In this regard, we found that curiosity represents the starting point of the potential Brazilian tourist decisionmaking, directly, and strongly influencing visitor attitude towards destination, the individual's response to the holistic destination attributes, and the willingness to recommend or revisit Europe (see Figure 2). From a theoretical point of view, the identification of a specific push motivation, which has a role in modeling pull motivations, in shaping attitude and in influencing loyalty, represents a general advancement in tourist's motivation research, improving the existing understanding on the key role that psychological factors and destination attributes play in shaping tourist assessment and decision-making. Additionally, by examining the destination attributes measurement, the study provided a better understanding of the pull motives specifically attracting Brazilian tourists to Europe. In terms of policy and managerial implications, the study allows exploiting visitors' curiosity to travel to Europe, considering the direct effect on loyalty pull motives and attitude. In order to generate loyalty, it will be crucial for Europe to develop a common ground tourism policy, promote direct flight to Europe, making it easy to travel across Europe, and finally propose an adequate quality standard of accommodation across all the countries comprising the European Union. Moreover, given its strong influence, curiosity should be stimulated through promotional campaigns by implementing a purposive feeding of destination information from an ecosystems' perspective of value co-creation.

Source: The authors
The study examined Brazilian tourists only. Such a limitation offers opportunities for future researches in other emerging Countries (i.e. other Latin American countries, or Asian countries). Furthermore, the study should be verified in relation to other tourist destinations. In this context, what other destinations might include curiosity as a push motive and how do they compare to this destination?
The model was aimed at starting a debate on curiosity as push motivation to travel in the tourist emerging market and in order to reduce the potential complexity of the model, we intentionally omitted to explore curiosity taking into account all its components (e.g., perceptual, epistemic, diversive, etc.). Therefore, future research might introduce these constructs for a more refined version of the model developed here.