Place attachment and brand loyalty: the moderating role of customer experience in the restaurant setting

Purpose – The purpose of this study is to ascertain the effects of place attachment on brand loyalty. This study further ascertains whether the effects of emotion-based attachment on brand loyalty are stronger for customers who have a positive experience with a restaurant brand. Additionally, the authors investigate whether emotion-based attachment mediates the relationships between identity-based attachments, place dependence and brand loyalty in the restaurant setting. Design/methodology/approach – The authors administered the questionnaire to customers (diners) of restaurants in Ghana, and they were completed via a paper and pencil/pen approach. The authors tested their hypotheses using structural equation modeling. Findings – The findings show that identity-based and emotion-based attachment enhances brand loyalty within a restaurant setting. The results also show that place dependence attachment promotes emotional bonding with restaurant brands. The study’s findings also show that place dependence attachment does not have a direct and positive significant effect on brand loyalty except when an emotional response is produced. Originality/value – Place attachment studies in a restaurant setting are rare. This study thus contributes to the place attachment literature in restaurants setting.


Introduction
The restaurant sector has experienced astronomical growth in many emerging economies because of changing social dynamics, including increased affluence in many societies (Baum, 2013). These trends could be said to have also heightened competition in the restaurant sector. According to Jang et al. (2015), in a highly competitive market such as the restaurant sector, many firms try to develop marketing and branding strategies to help them minimize customer switching and improve customer loyalty (So et al., 2013). The restaurant largely depends on its physical appearance to enhance customer experience. Customers, in the long run, find an emotional relationship with the restaurant setting. The site attributes or the place facilities must necessarily align to the individual specific needs (Suntikul and Jachna, 2016), hence become "attached to the place." Place attachment has been identified as an individual and an environment (Ujang and Zakariya, 2015).
Citing Gross and Brown (2008), Zhan et al. (2020) mentioned two-dimensional constructs of place attachment to include place dependence and place identity. While place dependence is construed to mean the reliance on the functions of the specific place (Tsai, 2012), place identity reflects the similarities between oneself and the place (Scannell and Gifford, 2010;Zhan et al., 2020). Place attachment could be said to be part of people's feelings. The experience of pleasant feelings makes it possible for people to like places. Schwarz and Clore (1996) opine that pleasant or unpleasant feelings about places influence the level of happiness, which could eventually lead to satisfaction or otherwise. In the tourism industry, some studies (e.g. Veasna et al., 2013) on place attachment have looked into place attachment as an antecedent to customer satisfaction (Zhan et al., 2020). Other studies (Hemsley-Brown and Alnawas, 2016;Tumanan and Lansangan, 2012;Patwardhan et al., 2019) have also shown the importance of attachment in building brand loyalty in the hospitality industry. Tumanan and Lansangan (2012) noted that physical environment and social environment impact customer attachment to a coffee shop using the place attachment theory. Similarly, some studies (such as Choi and Kandampully, 2018) have shown the impact of the physical environment on consumer emotions and, subsequently, their behavior. Thus, from the literature, attachment theory is essential in designing brand strategies for restaurant brands that seek to beat the competition and survive in a highly completive market (Tumanan and Lansangan, 2012;Parry and Hassan, 2019;Spooner, 2019).
Despite these, limited studies have investigated the extent to which place attachment explains brand loyalty in the restaurant setting. However, it is necessary to apply place attachment in the restaurant setting since atmosphere-dominant service firms such as restaurants can create a sense of place and foster customers' attachment to a specific setting (see Brocato et al., 2015;Kim and Stepchenkova, 2017). Additionally, although the existing literature suggests that emotion-based attachment impacts brand loyalty (Hemsley-Brown and , Maharani (2013) postulates that a consumer's attachment to a restaurant could be stronger when the consumer experiences positive feelings during dining. We also support the literature, which shows that customers' bad experiences with a brand can stimulate brand hate and subsequently brand avoidance (Hegner et al., 2017).
As indicated earlier, restaurant facilities, when aligned to the specific needs of the individual, are likely going to influence the individual to revisit the place. This suggests that affective images in the restaurant could prove to be effective in ensuring repeat visits by customers. Many restaurants provide similar services (Martey, 2014), as such must find ways to encourage diners to make repeat visits. The ability of the restaurant to offer a relaxed ambiance would ensure repeat visits, and these are essential to creating loyal customers, hence keeping the business afloat. Positive customer behavior can also be influenced by a good customer experience (Ching-Shu, 2011). In this sense, it can be argued that the customer's loyalty can further be improved because of a positive customer experience (Choi et al., 2013).
Brand experience moderates the relationship between emotion-based attachment and brand loyalty. Our current research addresses these gaps by employing place attachment theory to explain brand loyalty within the restaurant context. We also examine the mediating role of emotion-based attachment in the relationship between identity-based attachment and place dependence and brand loyalty in the restaurant setting.
The study results can assist managers in the luxury restaurant sector make successful and effective strategic decisions regarding brand management and customer relationship (CRM) management.

IHR
Theoretical background: place attachment According to Ujang and Zakariya (2015), place attachment is a link or association between a person and an environmental setting. Ujang and Zakariya (2015) mention that place attachment is associated with the affective aspects of environmental meaning. Scannell and Gifford (2010) suggest that place attachment refers to the connection (closeness, bonding) between individuals and their meaningful environments. Loureiro (2014) also adds that place attachment has been defined variously to refer to a person's closeness to an environment, emotional investment in a place and the degree to which a person belongs to an environmental setting. This means that place attachment could refer to the unique emotion or sentiment that an individual has toward a particular place. Previous works (e.g. Loureiro, 2014) support the conceptualization of place attachment involving social and physical components. According to Tumanan and Lansangan (2012), attachment is an outcome of both physical and social necessities. They further added that the physical aspects that affect consumer behavior involve the place detail and design elements. In their work on place attachment and proenvironmental behavior, Ramkissoon et al. (2013) consider place attachment a multifaceted concept comprising place dependence, place identity, place effect and place social bonding. Loureiro (2014) further adds that place attachment reflects identification with physical attributes. Consistent with this study, this paper considers elements of a place to include the natural environment and layout of the place. Place attachment is responsible for strong emotional and physiological reactions in people (Lewicka, 2011). This can be connected to the elements of attraction, regularity of visits and level of familiarity (Ujang and Zakariya, 2015).
In the attachment literature, place attachment is mostly operationalized from fourdimensional perspectives; place affects, place social bonding, place dependence and place identity (Tumanan and Lansangan, 2012;Scannell and Gifford, 2010). Since place affect and social bonding are emotion-based (Ramkissoon et al., 2013), we identify three dimensions of place attachment in this study; identity-based attachment, place dependence attachment and emotion-based attachment. We also identify different elements of these dimensions of place attachment. Brand identification and self-image expression are identified as elements of identity-based attachment, while a sense of natural environment and layout are identified as elements of place dependence attachment. Also, brand passion, brand pride and brand love are identified as elements of emotion-based brand attachment. Our argument in this paper is that these three dimensions of place attachment directly and positively affect brand loyalty.
Additionally, the interplay between these dimensions of place attachment has a positive effect on brand loyalty. Specifically, we argue that emotion-based attachment mediates the relationship between place dependence attachment and brand attachment and the relationship between identity-based attachment and brand loyalty. We also argue that the effect of emotion-based attachment on brand loyalty is stronger for customers who have a positive experience with a restaurant brand than those who do not.
consumers' preparedness to pay a premium price or the expenditure of a substantial amount to acquire the brand (Oliver, 1999). Behavioral brand loyalty is a deeply held commitment to rebuy a preferred product or service consistently in the future, despite situational influences and marketing efforts having the potential to cause switching behavior. Therefore, consumers who are attitudinally loyal are willing to recommend their preferred brands and are differentiated from behaviorally inclined loyal customers in the latter's inability to disseminate favorable word of mouth about the brands. The importance of a loyal customer cannot be underestimated as loyal, and committed customers become more profitable to firms than those who only purchase repeatedly (Oliver, 1999). Loyal customers pay less attention to extra information when making a purchase and are resistant to the marketing efforts of rival companies. However, Tanford (2013) argues that a superior indicator of loyalty should be seen when a customer chooses a particular brand over others. In Drennan et al. (2015) on brand loyalty, some of the benefits of brand loyalty to the restaurant include reducing marketing communication costs and reducing the need for sales promotions or advertising compared to brands with a low level of brand loyalty.

Hypotheses development
Identity-based attachment Brand identification -The concept of identification emerged from the social identity theory, which posits that self-concept involves personality, group classification, categorizations, association and self-identity (Tajfel and Turner, 1985). In the marketing literature, brand identity has been studied and defined as a distinctive set of brand associations that brand strategists seek to create or maintain (Centeno and Wang, 2017). Brand identification in this study means diners' perceived sense of belonging or oneness with a restaurant brand. According to So et al. (2017), consumers do not buy brands only to do well. They buy because they can be identified with the brand. Also noting that brand identity aids consumers in developing a stronger relationship with the brand. The impact of brand identification on consumers' emotions has been documented. Brand identification has been found to influence brand love (Albert and Merunka, 2013;Alnawas andAltarifi, 2016) in the hotel industry (Mody et al., 2019); brand passion (Batra et al., 2012) and brand pride (Helm et al., 2016). In the work of Rather (2018) and Rather and Hollebeek (2019), the authors noted that brand identification also significantly influences brand loyalty in the hospitality industry.
H1a-c. Brand identification positively affects brand passion, brand love and brand pride.
Self-image expression -Self-image reflects what individuals think about themselves and their ability to undertake an activity. Self-image describes the ideas and feelings that people have about themselves concerning others through a socially determined frame of reference. Consumers usually favor brands they believe possess symbolic images akin to their selfimage and try to achieve further self-image congruence (Hosany and Martin, 2012). According to self-congruity, the theory describes a customer's propensity to buy brands compatible with their self-image (Lee and Jeong, 2014). Self-brand incongruence creates brand hate, while self-brand congruity impacts and builds on brand love (Ranjbarian et al., 2013). Following these findings, we hypothesize that: H2a-c. Self-image expression positively affects brand passion, brand love and brand pride.

Place dependence attachment
The tourism and leisure literature describes place dependence as visitors' functional attachment to a place and their consciousness of the distinctive nature of the place, which adds to meeting the visiting aims of the visitors (Williams et al., 1992). In this study, we identify two dimensions of place dependence: a sense of natural environment and layout, as both help describe the functional utility (e.g. for work, food or recreation) (Han et al., 2019). Sense of natural environment and layout serve as unique attributes embedded in a place (thus, restaurant settings in this study's context).
Sense of natural environment -The physical environment is a critical component of place attachment (Scannell and Gifford, 2010). A natural environment means that every part of a restaurant brand is made of natural materials, and the atmosphere of the restaurant is natural. In their study, Orteg on-Cort azar and Royo-Vela (2019) noted that individuals' perceptions of a natural environment could create a favorable emotional response and elicit positive attitudes from diners. Similarly, Jang et al. (2015) found that green practice positively and significantly affects consumers' emotional attachment to the store. Again, they noted that consumers' attachment to a store subsequently results in their loyalty. Given these, we argue that the natural environment creates bonding between a diner and a restaurant brand. We hypothesize that: H3a-c. Sense of natural environment positively affects brand passion, brand love and brand pride.
H3d. Sense of natural environment positively affects brand loyalty.
Layout -Layout means the way objects are arranged in an environment (Bitner, 1992). In this study, a layout is defined as the attractiveness of a restaurant brand's interior and exterior arrangement. It includes a seating arrangement and the building design of a restaurant brand. Some studies have shown that layout influences consumers' behavior and emotions. Duong (2016) studied the impact of a store layout on shopping satisfaction and impulse buying. Duong found that layout and interior design do not affect customer satisfaction. However, an earlier study by Baker et al. (2002) noted that an awkward layout could decrease shopping efficiency and affect customers' emotions, making them angry and subsequently reducing repurchase. In line with Baker et al. (2002), we argue that and hypothesize that: H4a-c. Layout positively affects brand passion, brand love and brand pride.
H4d. Layout positively affects brand loyalty. . Sarkar (2014) adds that passion mainly results from motivation and encompasses various sources of arousal. Passion is an intense feeling for a brand to the extent that its absence will likely create mental distress (Sarkar, 2014). Therefore, brand passion is an emotional attachment and "reflects intense and aroused positive feelings toward a brand (place)" (Thomson et al., 2005, p. 80).

Emotion-based attachment and brand loyalty
Brand pride -Kuppelwieser et al. (2011) suggest that "pride in something is an inherent feeling and belongs to the 'elemental emotions.'" According to Cavanaugh et al. (2011), pride reflects "[a] feeling of satisfaction and fulfilments over one's achievements and capabilities" (Bellezza and Keinan, 2014). Consumers develop inherent feelings with a brand when they seemingly identify themselves with it. The brand's success provides an enormous emotional stimulus, and the consumer will outwardly display his or her positive emotions.
In their conceptualization, Carroll and Ahuvia (2006) describe brand love as the degree of passion, emotional attachment a satisfied consumer develops for a specific brand name. Similarly, Langner et al. (2016, p. 16) "regard brand love as a relationship between consumers and their brands that is pivotally characterized by a long-lasting, deep affection for the brand and anticipated separation distress." Therefore, brand love is used to describe a strong emotional link between the brand and the consumer (Carroll and Ahuvia, 2006). As a result, brand love integrates "passion and declarations of love for the brand, attachment to the brand and positive emotions in response to the brand" (Drennan et al., 2015).
There is an abundance of research that shows that the various emotional bonds consumers develop for a brand positively affect brand loyalty. For example, in their work, Kim and Stepchenkova (2018) noted that positive emotions have a significant effect on brand loyalty in the restaurant industry. On the other hand, Jang et al. (2015) note that coffee shop consumers' emotional attachment influences brand loyalty. Thus, deep emotions from consumers will most probably achieve loyalty. In line with the above studies, we, therefore, hypothesize that: H5a. Brand passion positively affects brand loyalty.
H5b. Brand love positively affects brand loyalty.
The mediating role of emotion-based attachment Emotion-based attachment involves the close emotional connection between a diner and a restaurant brand. Consumer emotion has been associated with loyalty and customer citizenship behaviors (Wang et al., 2016;Cheng et al., 2016). Mehrabian and Russell's (1974) developed a stimulus, organism, response (S-O-R) model to study the effect of environmental stimuli on people's emotions and consequently their behavioral responses. Their findings indicate that the physical environment impacts people's emotions, resulting in their behavioral responses. Some researchers have used this model to investigate consumers' behavior in a retail environment and restaurants (Peng et al., 2017). This suggests that the effect of place attachment on brand loyalty is mediated by emotion-based attachment. A study by Sheng et al. (2011) on fashion apparel retailing shows that physical environment and social environment elicit positive emotions from customers, affecting their behavioral intentions.
Similarly, Hemsley-Brown and Alnawas (2016) found that brand affection partially mediates the effect of staff behavior on brand loyalty. In the same vein, Peng et al. (2017) noted that individuals' positive and negative emotions mediate the relationship between restaurant stimuli such as service staff, food quality, atmospherics and customer loyalty. In line with these existing works, we argue that identity-based attachment and place dependence attachment create emotion-based attachment and, subsequently, brand loyalty. Following this line of argument, we hypothesize that: H6a. Emotion-based attachment mediates the relationship between brand identification and brand loyalty.
H6b. Emotion-based attachment mediates the relationship between self-image expression and brand loyalty.

IHR
H6c. Emotion-based attachment mediates the relationship between a sense of natural environment and brand loyalty.
H6d. Emotion-based attachment mediates the relationship layout and brand loyalty.
The moderating role of customer experience The literature explains customer experience to mean the consumer's "internal and subjective response" to direct/indirect contact with a firm ( Similar to the sample employed by Kim and Stepchenkova (2018), this study sampled customers from four family restaurant brands (thus, two indigenous and two global brands). The restaurants were located in Accra, the capital city of Ghana, and a cosmopolitan city in nature. The staff of these restaurant brands were contacted via management and asked to administer the questionnaire on behalf of the researchers to summarize the findings of the result. However, before administering the questionnaire, the research pretested the questionnaire with 30 customers of a different restaurant brand located in Accra to help reduce the ambiguity for the scale items (Malhotra et al., 2017). The questions were modified according to the suggestions provided by the 30 customers. After the modification process, the staff of the four restaurant brands was given 200 questionnaires each to distribute across their restaurant chain (800 in total). The restaurant staff was required to use an intercept approach, where the questionnaire was administered to the respondents while they were dining and collected when they were leaving. Hence, sample selection was the result of convenient sampling.

Place attachment and brand loyalty
A total of 614 surveys were collected; however, 592 were used for analysis because 22 were deleted for missing data. The male respondents were 45.1%, and the other 54.9% were females. Their ages were 18 years and above. The majority of the respondents either had or were studying for their degree at the university level (59.5%), spend an average of GHC11 (equivalent to US$2) to GHC 50 (equivalent to US$10) at the particular restaurant in a week and had been eating at that restaurant for at least 1-6 months (37.2%). Table 1 shows the detailed sample characteristics.

Measurement instrument
The items measuring the constructs were adapted from the existing literature. They were chosen based on their consistency with our conceptualization of the constructs. Measures of self-image expression were adapted from Kim et al. (2011). We used measures from Escalas and Bettman (2003) to assess brand identification, Ryu and Jang (2007) to assess layout, Thomson et al. (2005) (2001). All the items were measured on a five-point Likert scale; 1 5 strongly disagree; 2 5 disagree; 3 5 neutral; 4 5 agree; and 5 5 strongly agree.
A confirmatory factor analysis (CFA) was conducted to assess the convergent validity of our measures. We evaluated the convergent validity of the measures by assessing how well they loaded on their assigned construct. We deleted items-construct loadings less than 0.5, and as a result, the remaining observed standardized loadings were high and significant (see Appendix), providing support for convergent validity. The fit indices for the CFA model after the deletion of items are χ 2 /df is 2.348, CFI is 0.934, SRMR is 0.039, RMSEA is 0.048 and PClose is 0.867, an indication of good model fit (see Hu and Bentler, 1999 Table 2 and in 'italics' were larger than the shared variances between constructs (see Table 2), indicating satisfactory discriminant validity (Fornell and Larcker, 1981). Although AVE for brand love was less than that of brand pride, this was not an issue since the items seemed very similar, and also, as noted by Malhotra and Dash (2011), AVE is often too strict, and reliability can be established through CR alone. We examine the potential for common method variance (CMV) using Harman's one-test factor method (Podsakoff et al., 2003). All the substantive variables were then entered into an exploratory factor analysis (EFA). Results showed that a single factor did not emerge from factor analysis. Moreover, when all items were forced to load onto a single factor, it accounted for 36.24% of the variance. This was an indication that there were no serious concerns about CMV.

Control variables
We used gender, educational level, average monthly spending and length of time eating from this restaurant in a month. Studies have shown that demographic variables such as the age of consumers, education levels and income levels significantly influence restaurant patronage (see Hu et al., 2010). Also, Jin et al. (2013), in their studies, show that a demographic variable such as gender influences loyalty in the context of fine-dining restaurants and relationship quality and behavioral loyalty, respectively.

Results
To test our hypotheses, we conducted a structural equation model (SEM) using IBM Amos 21. We first used the Amos to examine the antecedents and consequences without testing the moderating and mediating effects. The model exhibited good fits: χ 2 /df is 1.855, CFI is 0.994, SRMR is 0.062, RMSEA is 0.038 and PClose is 0.913. The result is presented in Table 3.
Hypothesis 1 stated that brand identification is positively associated with brand passion, brand love, brand pride and brand loyalty. The effect of brand identification is found to be positive and significant on brand passion (β 5 0.38, t 5 10.29), brand love (β 5 0.17, t 5 4.78), brand pride (β 5 0.10, t 5 2.21) and brand loyalty (β 5 0.21, t 5 5.70). Consequently, H1a-H1d are supported. Diners who perceived a sense of belonging or oneness with a restaurant brand are more likely to build emotional bonds with that brand and build loyalty directly.
Hypothesis 2, on the other hand, predicted that self-image expression positively affects brand passion, brand love, brand pride and brand loyalty. Self-image expression is found to have positive and significant effects on brand passion (β 5 0.16, t 5 3.65), brand love (β 5 0.21, t 5 4.81) and brand pride (β 5 0.25, t 5 4.71), but has a negative and significant effect on brand loyalty. Thus, H2a, H2b and H2c are supported. Diners who believe a restaurant brand possesses symbolic images akin to their self-image are more likely to develop emotional bonds such as passion, pride and love for the brand.
Hypothesis 3 posited a positive relationship between a sense of natural environment and brand passion, brand love, brand pride and brand loyalty. Sense of natural environment had a significant positive relationship with brand passion (β 5 0.10, t 5 2.24), brand love (β 5 0.45, t 5 10.86) and brand pride (β 5 0.32, t 5 6.32) but not brand loyalty. The results, therefore, support H3a, H3b and H3c, but not H3d. Restaurants with a sense of natural environment are more likely to stimulate emotional responses from diners, but this does not directly result in restaurant brand loyalty.  Table 2. Mean, standard deviation and correlations IHR Hypothesis 4 indicates that layout positively affects brand passion, brand love, brand pride and brand loyalty. Layout is found to have positive and significant effects on brand passion (β 5 0.29, t 5 5.70), brand love (β 5 0.11, t 5 2.26) and brand pride (β 5 0.15, t 5 2.44); however, it is negatively associated with brand loyalty (β 5 À0.29, t 5 À6.03). This finding provides support for H3a, H3b and H3c. Restaurants with an attractive layout and natural environment are highly likely to stimulate emotional responses. However, if an emotional response is not stimulated, brand loyalty is less likely to be galvanized. Age À0.01 À0.62 Note(s): Standardized coefficients are report Two-tailed tests are used *p < 0.05, ***p < 0.001 Table 3. Test results: brand attachment and its impact on restaurant brand loyalty Place attachment and brand loyalty Hypothesis 5 suggested that the elements of emotion-based attachments are positively associated with brand loyalty. Thus, brand loyalty, brand passion and brand love showed a significant positive relationship (β 5 0.36, t 5 6.73) and (β 5 0.55, t 5 9.96) respectively, but the effect of brand pride was not significant. Brand passion and brand love drive loyalty in restaurant brands.
Hypothesis 6 proposed that emotional-based attachment mediates the relationship of the identity-based attachment (brand identification and self-image expression), place dependence attachment (sense of natural environment and layout) and brand loyalty. The bootstrapping procedure was used to examine the indirect effects. The results reveal a significantly positive effect of brand identification on brand loyalty (β 5 0.26, p ≤ 0.001), self-image expression and brand loyalty (β 5 0.19, p ≤ 0.001), sense of natural environment and brand loyalty (β 5 0.30, p ≤ 0.001) and layout and brand loyalty (β 5 0.18, p ≤ 0.001) via emotion-based attachment. We find that emotion-based attachment mediates the effects of brand identification, selfimage expression, natural environment and layout and brand loyalty. Among the antecedents, the natural environment did not directly affect brand loyalty and thus was fully mediated by emotion-based attachment. Brand identification, self-image expression and layout have significant direct effects on brand loyalty and are thus partially mediated by emotion-based attachment. The direct effects of self-image expression and layout on brand loyalty are negatively significant, indicating that when self-image expression and layout do not elicit some emotional bonds, its effect on loyalty is not significant.
We constructed a second SEM to test the moderation effects in Hypothesis 7 using brand passion and brand love (which is significantly related to brand loyalty) and customer experience (see Table 4). Brand pride was not included in this model since its direct relationship with brand loyalty was not significant. Thus, hypotheses 7a and 7b stated that customer experience moderates the effects of brand passion and brand love on restaurant brand loyalty, respectively. To test this, the main effects, brand passion and brand love, were first entered (model 1), next was the addition of customer experience (model 2), and then sequentially, the interaction effect was added (model 3). The results of the analysis show that each step significantly enhanced the R 2 value of the models. We found no significant Independent  Table 4. Test results: contingent effect of the mediating role of customer experience on brand attachment and loyalty IHR interaction between brand love and customer experience on restaurant brand loyalty; thus, hypothesis 7b was not supported. However, brand passion and customer experience interaction were significant and positive on restaurant brand loyalty (β 5 0.11, t 5 2.71); therefore, hypothesis 7a was supported. This result and the interaction plot in Figure 1 show support of hypothesis 7a, showing that customer experience strengthens the positive relationship between brand passion and brand loyalty (see Figure 2 and Table 5).

Discussion of findings
This study examines the effect of identity-based attachment and place dependence attachment on emotion-based attachment and, consequently, their effects on brand loyalty in restaurant settings. This study further ascertains whether the effect of emotion-based attachment on brand loyalty is stronger for customers who have a positive experience with a restaurant brand. Additionally, we investigate whether emotion-based attachment mediates the relationships between identity-based attachments, place dependence and brand loyalty in the restaurant setting.
The study's results supported the hypothesis that brand identification positively relates to emotion-based attachment (brand passion, brand love, brand pride). This suggests that the degree to which an individual defines oneself regarding the expected attributes of a restaurant (Hughes and Ahearne, 2010) positively influences emotion-based attachment. Again, findings also indicate that brand identification is positively related to brand loyalty. This finding is in congruence with the results of Kumar and Kaushik (2017) in their study on destination brand identification and concluded that brand loyalty is a positive outcome of brand identification. The findings of the study also reveal that self-image expression influences emotional-based attachment and brand loyalty. Diners here believe that some images they consider symbolic and analogous to their self-image would more likely build up Place attachment and brand loyalty some emotional attachment to the restaurant (Support Reference). Again when these diners believe that those symbolic images are consistent over time, they will likely make repeat visits to the restaurants. The study further reveals that the elements of emotion-based attachment (brand passion, brand love) bar brand pride are positively related to the brand loyalty of the diners.
From the analysis, we further found support for the hypothesis that place dependence attachment (sense of natural environment and layout) is positively related to emotion-based attachment. This means that the sense of the natural environment and the restaurant's layout arouse some level of emotional response from the diners. The sense of natural environment and the layout in service-related premises have a positive effect on consumers' psychological states, supporting the findings of Purani and Kumar (2018). Place dependence attachment is

Dimensions Hypothesis Result
Identity-based attachment Brand identification positively affects brand passion Accepted Brand identification positively affects brand love Accepted Brand identification positively affects brand pride Accepted Brand identification positively affects brand loyalty Accepted Self-image expression positively affects brand passion Accepted Self-image expression positively affects brand love Accepted Self-image expression positively affects brand pride Accepted Self-image expression positively affects brand loyalty Accepted Place dependence attachment Sense of natural environment positively affects brand passion Accepted Sense of natural environment positively affects brand love Accepted Sense of natural environment positively affects brand pride Accepted Sense of natural environment positively affects brand loyalty Rejected Layout positively affects brand passion Accepted Layout positively affects brand love Accepted Layout positively affects brand pride Accepted Layout positively affects brand loyalty Rejected Emotion-based attachment Brand passion positively affects brand loyalty Accepted Brand love positively affects brand loyalty Accepted Brand pride positively affects brand loyalty Rejected Figure 2.
The interaction between brand passion and customer experience on restaurant brand loyalty Table 5. Summary of hypothesis IHR also related to brand loyalty even though one of the elements (sense of natural environment) does not have a direct relationship. This can be explained by the full mediation effect of emotion-based attachment. Thus, if an emotional response is not stimulated, brand loyalty is less likely to be galvanized. This could be because the sense of aroused emotions is not steady for a more extended period, so diners do not see the essence of making repeat visits. The study's findings further supported the hypothesis that elements of the emotionbased attachments (brand passion and brand love) are positively related to brand loyalty bar brand pride. Therefore, feelings of satisfaction and fulfillment over the achievements of a restaurant alone are not enough to significantly influence brand loyalty. Nevertheless, brand love and brand passion, which describe a rich and affective recollection of love varying from warmth and affection to obsessive dependency and are viewed as the first dimension of brand relationship quality, significantly influence loyalty (Fournier, 1998). This indicates that loyalty to a restaurant brand is dependent on stronger bonding relationship quality such as passion and love. The findings reveal that higher customer experience strengthens the positive relationship between brand passion and brand loyalty, but not for brand love. This could be because love is a stronger emotional connection compared with brand passion.
From the findings, emotion-based attachment is vital in the influence of the natural environment on brand loyalty. Despite the environment being essential to the diner, the diner must necessarily be passionate and further develop a love for the restaurant brand to make repeat visits. The effects of brand identification, self-image expression and layout on brand loyalty are partially mediated by emotion-based attachment. The direct effects of self-image expression and layout on brand loyalty are negatively significant, and this suggests that when self-image expression and layout do not elicit some emotional bonds, its effect on loyalty is not significant.
The study further found the effect of brand passion on brand loyalty in the presence of customer experience. Customer experience is, however, absent in the interaction between brand love and brand loyalty. It suggests that customer experience is essential in determining how customers become loyal despite their passion for the restaurant brand. It is, therefore, necessary to continually improve the experience of the customers into a positive, memorable experience.

Theoretical contributions
This study proposes a conceptualization of identity-based attachments, place dependence and emotion-based attachment for understanding place attachment and how it results in brand loyalty.
According to Patwardhan et al. (2019 p. 15), in the tourism literature, "satisfaction is among the most studied mediating variable to between predictor variables and loyalty formation." Therefore, with emotion-based attachment as a mediator variable to diner's loyalty behavior in a restaurant context, this study offers new insight. To the best of our knowledge, this is the first study to test these relationships in the restaurant context of a developing country.
The use of these theories in the context contributes to the body of knowledge concerning tourism and hospitality literature, thus addressing the need for further expansion in the field by linking empirical research with established theories. The results connect identity-based attachment and place dependence attachment to emotion-based attachment, explaining the developmental processes by which place attachment arises. The framework, therefore, allows us to understand three closely related aspects of place attachment: place identity, place dependence and emotion-based attachment, which people establish because of their identification with a place and the arousal from the environmental cues.
The study provides the opportunity to understand the role of emotion-based attachment on the relationship between place dependence attachment and loyalty. The quality of the environment alone may not evoke a sense of repeat visits at the restaurant unless efforts are made to stimulate the emotional response of the diner. Further, the study makes it possible to appreciate the importance of customer experience in determining the strength of the brand passion on brand loyalty. Customer encounter is essential to improving customers' positive experiences, and this is likely going to provide adequate grounds for customers with a passion for the restaurant brand to make repeat visits consistently.

Managerial contribution
The findings of this study indicate that managers must strive to create a natural environment and an attractive environment to stimulate consumers' emotions in the restaurant. To ensure this is attainable, managers must monitor and understand how their natural environment and layout affect customer emotions to make improvements where needed. This will allow them to have some level of control in shaping consumer emotion-based attachment toward that brand. For consumers that bring their kids along, some recreational facilities should be created for the kids. This will enhance the fun and excitement involved in the patronage of the services. For the restaurants that are primarily located in the traditionally sensitive area, much of the layout should portray the traditions of the locality and a sense of natural environment, as these will help the consumers identify themselves more with the setting.
Restaurant managers should also invest in marketing activities that enhance place identity since a sense of identification from the consumer toward the place appears to be critical for establishing emotional bonds with that place. Generally, managers' activities should aim to convey a place image reflecting, participating in or create consumers' selfidentity and self-image expression to help increase their identification with the place.
Limitations and directions for future research We tested our model using data from the restaurant setting; however, there are several place settings where place attachment can be studied. In Ghana, traditional restaurants popularly known as chop bars are good places to investigate place attachment. The layout of these conventional restaurants is usually designed with a traditional Ghanaian home setting compared to the modern restaurants, and their environment is usually more natural than the contemporary restaurants. That is, these traditional restaurants are different from modern restaurants in terms of layout and natural environment. Therefore, we propose that future studies should test our model within this setting. Another setting where place attachment can be studied is educational institutions. Thus, we suggest that future studies should test our model within settings such as higher educational institutions.