Servicescape, Price Perception, and Diner Loyalty: Empirical Evidence from Full-Service Restaurants in Northern Peru

Abstract

Customer loyalty is a critical asset for the restaurant industry, yet the mechanisms linking the physical environment, price perception, and satisfaction remain underexplored in emerging Latin American gastronomy markets. This study examines the relationships among three servicescape dimensions—décor and artifacts, spatial layout, and ambient conditions—price perception, customer satisfaction, and loyalty in full-service restaurants in northern Peru (Chiclayo, Trujillo, and Piura). A cross-sectional survey was administered to 310 diners, and the proposed model was tested using partial least squares structural equation modeling (PLS-SEM) with 10,000 bootstrap resamples. Results supported seven of nine direct hypotheses and three of four mediation hypotheses. Décor and artifacts and ambient conditions significantly predicted both price perception and satisfaction, while spatial layout showed no significant effect on any path. Price perception partially mediated the effect of décor and ambient conditions on satisfaction, and satisfaction partially mediated the relationship between price perception and loyalty. The satisfaction–loyalty path yielded the largest effect size (β = 0.708, f² = 0.798). Serial chain analyses revealed that the physical environment shapes diner loyalty through sequential cognitive and evaluative mechanisms. These findings offer actionable insights for hospitality managers seeking to enhance gastronomy destination competitiveness through strategic servicescape investment.

1. Introduction

Customer loyalty is one of the most valuable assets for any service company, and the gastronomy industry is no exception to this premise. Retaining existing customers significantly reduces the costs associated with attracting new diners and, at the same time, increases profit margins (Coelho & Henseler, 2012; Mattison Thompson et al., 2014). Those who feel emotionally attached to a restaurant tend to recommend the place to family and friends, spend more than expected, and tolerate occasional service failures before considering alternatives (Ing et al., 2019; Jin et al., 2012). What happens, however, when this phenomenon is examined in emerging gastronomic markets in Latin America? Recent research in Peru has shown that satisfaction and loyalty are positively and significantly correlated across different restaurant formats, from Lima chicken restaurants to regional businesses (Lara Casanatan et al., 2024; Rivera Paredes et al., 2025). Similarly, studies conducted in cities in northern Peru reveal that brand positioning is directly associated with customer retention (Flores Curico et al., 2023). This evidence underscores the need to understand more precisely the mechanisms that underpin loyalty in a sector where competition intensifies year after year—where differentiation no longer depends exclusively on culinary quality, but on a broader set of factors that shape the overall dining experience.

However, loyalty does not arise spontaneously; it is largely based on the satisfaction that diners experience during and after consumption (Chinelato et al., 2023; Han & Ryu, 2009). Given that gastronomic services are predominantly intangible in nature, customers rely on perceptible cues—the atmosphere of the establishment, the interior design, the layout of the furniture, the background music, the lighting, or the price—to evaluate and assess their overall experience (Apaza-Panca et al., 2023; Chinelato and Cruz, 2025). The perceived quality of the physical environment, understood as the combination of decoration and artifacts, spatial distribution, and environmental conditions, has proven to be a robust predictor of both satisfaction and repurchase intentions across restaurant types (Li et al., 2025). At the same time, the perception that the price paid is reasonable relative to the benefits received reinforces the positive evaluation of the experience and, by extension, the propensity to return (Nejati & Parakhodi Moghaddam, 2013; Usiña-Báscones et al., 2024). When sensory stimuli such as aroma, temperature, and music harmonize with a comfortable spatial design and attractive décor, diners tend to perceive that the price paid is fully justified, which spurs both their immediate satisfaction and their willingness to recommend the establishment (Guillen Abregu, 2023; Ramos Farroñán et al., 2024). In other words, the physical environment not only serves as a passive backdrop for eating but also functions as a nonverbal communication system capable of shaping perceptions of value and influencing future consumer behavior. This convergence between atmosphere, price, and satisfaction has been corroborated in contexts as diverse as haute cuisine restaurants in India, casual establishments in Iran, and fast food chains in Chile, suggesting a cross-cultural logic that deserves to be tested in specific contexts such as Peru (Nejati & Parakhodi Moghaddam, 2013; Zeba et al., 2024).

Despite the robustness of these findings in international markets and in some Peruvian cities, there is a notable gap in the literature regarding northern Peru. Existing studies in this area have focused on describing general factors of gastronomic consumer behavior—cultural, social, and psychological influences—without addressing the specific dimensions of the physical environment, price perception, satisfaction, and loyalty (Bautista et al., 2023; Huerta-Tantalean et al., 2024). Although recent research recognizes that Lambayeque and northern cuisine has a differentiating potential capable of influencing the image of the destination and visitor loyalty (Esparza-Huamanchumo et al., 2025; Moreno Quispe & Hernández-Rojas, 2025), no study has empirically verified how the multiple components of the physical environment interact with price perception to explain satisfaction and, ultimately, diner loyalty in full-service restaurants in this region. Socioeconomic particularities deeply rooted in culinary traditions, and the rapid growth of formal dining options in cities such as Trujillo, Chiclayo, and Piura create a scenario that is difficult to explain by simply extrapolating results from foreign or capital contexts. Overcoming this limitation requires producing localized empirical evidence capable of capturing the dynamics of a market where regional culinary identity coexists with the progressive modernization of service formats.

Faced with this gap, this research aims to examine the relationships between the three components of the physical environment—decoration and artifacts, spatial distribution, and environmental conditions—price perception, customer satisfaction, and customer loyalty in full-service restaurants in northern Peru. Complementarily, the study seeks to determine whether price perception plays a mediating role between the dimensions of the physical environment and satisfaction, and to assess whether satisfaction mediates the relationship between price perception and loyalty. The aim is to provide localized empirical evidence that will enable restaurant managers in the region to design strategies to improve the physical environment and manage perceived price, with a view to strengthening customer loyalty and commitment.

2. Literature Review

2.1. The Physical Environment

Since the mid-20th century, environmental psychology has maintained that human behavior is closely linked to the physical characteristics of the surrounding space. The seminal works of Mehrabian and Russell proposed that people respond to their environment with bipolar responses: approach—a desire to stay, explore, and interact—or avoidance—a desire to leave the place and reduce contact (Han & Ryu, 2009). This premise, when applied to the field of consumption, has shown that an attractive and innovative environment elicits positive evaluations of the service, increases customer retention within the establishment, and predisposes favorable repurchase behavior (Chinelato and Cruz, 2025; Guillen Abregu, 2023). However, is it enough for the space to be pleasant? Recent research suggests that the atmosphere not only generates pleasure, but also activates cognitive processes—such as the perception of value and price evaluation—that ultimately shape the diner’s overall satisfaction (Usiña-Báscones et al., 2024). Thus, understanding the influence of the physical environment requires us to transcend the merely aesthetic dimension and consider the cognitive and affective mechanisms that are set in motion every time a customer walks through the door of a restaurant.

In the restaurant industry, the physical environment is particularly important. Diners perceive—consciously or unconsciously—the decor, the layout of the furniture, the lighting, and even the aroma of the premises from the moment they enter the establishment, and that perception accompanies the entire dining experience (Apaza-Panca et al., 2023). Although the culinary quality and service must be of an acceptable standard, it is the tangible elements of the space that often determine whether the diner remembers the visit as rewarding or insignificant. Various authors have noted that, in certain situations, the atmosphere of a place can be as important as the food itself in deciding whether to return or seek an alternative (Li et al., 2025). This premise is especially relevant in northern Peru, where the formal gastronomic offering is growing rapidly, and differentiation no longer rests solely on flavor, but on the restaurant’s ability to offer a comprehensive experience that engages the visitor’s senses and emotions (Esparza-Huamanchumo et al., 2025). An exceptional dish served in a neglected space is unlikely to have the same impact as that same dish presented in a carefully designed setting.

For this research, the physical environment is understood as the set of conditions and tangible elements created by humans within a service establishment—as opposed to the natural environment—that can be controlled and modified by restaurant managers to influence customers’ internal and external responses (Han & Ryu, 2009). This conceptualization encompasses both the objective factors of the built space—furniture, finishes, signage—and the background stimuli that shape the diner’s sensory experience. By defining the construct in this way, the social aspects of interaction (the relationship between staff and customer), which constitute a different dimension of service, are deliberately excluded.

Within the literature on the physical environment in services, three dimensions have been identified most frequently and consistently: decoration and artifacts, spatial layout, and environmental conditions (Han & Ryu, 2009). These three factors do not operate in isolation; they intertwine to shape the overall impression that diners form of the restaurant. It was Han and Ryu (2009) demonstrated that the tangible attributes of human service interact with the perceived value of the environment, reinforcing the idea that each dimension contributes in a different but complementary way to the overall experience. The characteristics and theoretical basis of each are examined below.

2.1.1. Decoration and Artifacts

Decoration and artifacts are the most immediate visual components of the physical environment. Within a restaurant, the color schemes of walls and ceilings, pictures or paintings, plants and floral arrangements, the quality of the furniture—tables and chairs—floor coverings, and table linens, together with cutlery, form a system of signals that diners process to form an aesthetic impression of the place (Han & Ryu, 2009). When these elements harmonize, they generate a sense of coherence and care that leads to favorable evaluations of the overall experience. Jamaludin and Hashim (2024) found that hedonic emotions—comfort, sentimentality, and stimulation—are intensified in spaces where the decor has been designed with narrative intent, which reinforces both attitudinal loyalty and the willingness to recommend the establishment. Similarly, Homeghi et al. (2025) showed that digital communication of decorative and gastronomic elements strengthens the regional brand image, suggesting that decoration not only operates within the premises but also transcends into virtual space as a sign of identity. In short, decoration and artifacts function as a form of nonverbal language that communicates quality, identity, and value to the diner from the first visual contact.

2.1.2. Spatial Distribution

The restaurant environment is not merely a contemplative setting; it is a functional space that must facilitate both service delivery and customer comfort. Spatial distribution refers to the arrangement of furniture, equipment, and circulation elements according to the needs of the service process (Han & Ryu, 2009). An efficient layout allows diners to move around without obstacles, sit comfortably, and not experience the feeling of crowding that erodes the enjoyment of the meal, Othman et al. (2025). It was demonstrated that personalization of service—including organizing space according to customer preferences—reinforces brand identification and repurchase intention in full-service restaurants. In markets such as Peru, where dining formats are rapidly modernizing, Huerta-Tantalean et al. (2024) observed that personal factors such as comfort and restaurant location have a decisive influence on consumer behavior toward casual dining chains. A well-designed space not only meets functional needs but also conveys an implicit message of professionalism and respect for the diner’s experience.

2.1.3. Environmental Conditions

While decoration and spatial layout appeal primarily to sight and the sense of movement, environmental conditions operate through more subtle sensory channels. These conditions encompass intangible background characteristics—lighting, temperature, background music, aroma, noise level, and air quality—that often have a subconscious effect on customer perception (Han & Ryu, 2009). When warm lighting, moderate music volume, a comfortable temperature, and a pleasant aroma converge in the same space, diners tend to evaluate their experience more positively and stay longer, Apaza-Panca et al. (2023). In restaurants in northern Peru, sensory marketing strategies—among which olfactory and auditory stimuli stand out—have a positive and significant relationship with purchasing decisions. Similarly, Guillen Abregu (2023) found that sensory experience is the most relevant dimension of experiential marketing in the fast-food sector in Lima. Meanwhile, Ramos Farroñán et al. (2024) corroborated that perceived quality and hedonic value derived from sensory stimuli directly influence brand loyalty among Peruvian consumers. This evidence supports the idea that environmental conditions, far from being a passive backdrop, actively participate in the construction of gastronomic experience.

2.2. Price Perception

Price is one of the few tangible indicators that service consumers can use before, during, and after the consumption experience. Given that the provision of a gastronomic service involves a high degree of human participation, no two visits to a restaurant will be identical. This variability generates uncertainty and pushes customers to look for complementary signals to evaluate what they receive (Han & Ryu, 2009). Price, precisely, stands out as one of those signals: diners use it not only to anticipate the quality of service but also to judge retrospectively whether the experience justified the expense (Jin et al., 2012). In full-service restaurants—where the average check is usually higher than in fast-food formats—price sensitivity is heightened, and any dissonance between what is paid and what is received can influence the customer’s future behavior toward seeking alternatives (Ing et al., 2019). Therefore, managing price perception is not a minor issue; it is a strategic axis that connects the tangible experience of the environment with the diner’s subjective evaluation.

It is important to distinguish between the target price—the figure printed on the menu—and the perceived price, i.e., the subjective interpretation that the customer makes of that figure. Han and Ryu (2009) argued that the target price acquires meaning only when the consumer interprets it in light of their own references, expectations, and previous experiences. The complexity inherent in the pricing system of a full-service restaurant—with multiple dishes, beverages, seasonal menus, and promotions—makes it difficult for diners to remember each item’s exact price; what really stays in their memory is the feeling of having paid a reasonable or excessive price (Nejati & Parakhodi Moghaddam, 2013). Ing et al. (2019) Among the various transactional characteristics evaluated by diners at full-service restaurants in Sabah, price perception was the most powerful predictor of satisfaction. Therefore, in the present study, price perception—not target price—is adopted as the focal construct, in line with evidence that favors subjective interpretation over numerical data.

2.2.1. Customer Satisfaction

Customer satisfaction occupies a central place in marketing theory and practice because it reflects the degree to which the consumer experience meets, equals, or exceeds the expectations that the individual had placed on the service. More than a specific emotional reaction, satisfaction operates as a global evaluative process: the diner compares what they expected to receive with what they actually perceived, and from that comparison emerges a summary judgment that conditions their future behavior (Chinelato et al., 2023). Research in the Peruvian gastronomic sector reinforces this conception. Lara Casanatan et al. (2024) A positive and robust correlation between satisfaction and loyalty in restaurants in Trujillo was found. Meanwhile, Rivera Paredes et al. (2025) obtained convergent results in chicken restaurants and restaurants in Lima, where dimensions such as empathy and tangibility of service were decisive. Guillen Abregu (2023) Evidence was added by demonstrating that sensory stimulation—linked to experiential marketing—increases the satisfaction of delivery consumers in Lima. At the same time, Chinelato et al. (2023) it was shown that the emotional experience derived from the quality of the environment and service enhances satisfaction, loyalty, and electronic word of mouth among Peruvian restaurant consumers. The convergence of these findings supports the argument that satisfaction is not an abstract concept but a concrete, measurable mechanism that connects the experience to the diner’s future intentions.

2.2.2. Customer Loyalty

Two perspectives on the nature of customer loyalty have been extensively debated in the literature. The first—the behavioral dimension—identifies loyalty with repurchase frequency: a customer is loyal if they consistently return to the same establishment. The second—the attitudinal dimension—defines it as the psychological and evaluative commitment that the individual maintains toward a brand or supplier, reflected in their willingness to recommend it, to prefer it over competitors, and to allocate additional resources (Coelho & Henseler, 2012; Mattison Thompson et al., 2014). Both dimensions capture different facets of the phenomenon. Flores Curico et al. (2023) In a restaurant in Yurimaguas (Peru), brand positioning was positively correlated with customer loyalty. However, they noted that reported loyalty levels were low despite observed repurchase, highlighting the need to differentiate between returning out of habit and returning out of conviction.

Several authors have pointed out the limitations of measuring loyalty exclusively based on repurchase behavior. This approach does not discriminate between those who return because they genuinely value the service and those who return out of inertia, lack of alternatives, or simple geographical proximity (Han & Ryu, 2009). Repeat purchases, in the absence of emotional commitment, constitute what the literature refers to as spurious loyalty: a vulnerable pattern that breaks down at the first attractive competitive offer. Goncalves Filho et al. (2022) demonstrated that authentic loyalty to a brand is forged through collective identity and a sense of community, factors that are clearly attitudinal in nature. On the other hand, Andia-Reyna and Malasquez-Villanueva (2025), in their systematic review of emerging technologies and customer loyalty, concluded that personalization and behavior prediction—both aimed at strengthening the emotional bond—are more effective at retaining customers than simple frequency programs. Confusing repetition with loyalty can lead to the design of ineffective retention strategies that focus on discounts or promotions rather than on building lasting relationships.

These considerations take on weight in the hospitality and restaurant industry, where loyalty involves an emotional component that is difficult to capture with purely transactional indicators. In gastronomic markets such as Peru’s—with a diverse and expanding offering—many diners alternate between several establishments without this implying disloyalty; they conduct out of curiosity or for different social occasions. Memorable experiences in casual Brazilian restaurants were found to impact loyalty, but this does not always translate into electronic word of mouth unless there is a high level of engagement on social media. In haute cuisine restaurants in India, the hedonic pathway—based on delight—predicts loyalty more strongly than the utilitarian pathway. Similar findings were reported Rivera Paredes et al. (2025) in Lima, where service empathy surpassed tangibility as a predictor of loyalty. Othman et al. (2025) added that service personalization strengthens brand identification and the intention to revisit. Based on this evidence, the present study adopts an attitudinal approach to measure dinner loyalty, evaluating intentions to return, recommend, and spend more than expected as indicators of genuine commitment to the restaurant.

2.2.3. Research Hypothesis

The conceptual framework presented in the preceding sections establishes that the physical environment of a restaurant—decomposed into décor and artifacts, spatial layout, and ambient conditions—functions as a system of nonverbal cues that triggers both affective responses (approach–avoidance behavior, as posited by the Mehrabian–Russell model) and cognitive evaluations, most notably the perception of whether the price paid is commensurate with the experience received. In turn, this price perception feeds into an overall evaluative judgment—customer satisfaction—which, according to the expectancy–disconfirmation paradigm, emerges from the comparison of anticipated and actual service quality. When satisfaction is understood in its attitudinal dimension—as psychological commitment rather than mere repeat purchase—it constitutes the proximal antecedent of genuine customer loyalty. Building on this integrated logic, the following hypotheses formalize each theorized link: first, the direct effects of the three servicescape dimensions on price perception (H1–H3) and on satisfaction (H4–H6); second, the effect of price perception on satisfaction (H7) and on loyalty (H8); third, the effect of satisfaction on loyalty (H9); and finally, the mediating roles of price perception (H10–H12) and satisfaction (H13) that capture the sequential cognitive–evaluative mechanism through which the physical environment ultimately shapes diner loyalty.

The literature on environmental psychology establishes that customers of a service establishment respond to the dimensions of their physical environment both emotionally and cognitively. From a cognitive perspective, decoration, space distribution, and environmental conditions operate as forms of nonverbal communication: they convey messages about service quality, establishment category, and, importantly, whether the price charged is reasonable or excessive (Han & Ryu, 2009). When diners enter a restaurant whose decor projects sophistication, whose furniture is spaciously arranged, and whose ambiance harmoniously stimulates the senses, they tend to interpret that the price paid is commensurate with what they receive. The perceived attributes of a dining destination—including the service environment and the price-quality ratio—were demonstrated to delight diners and reinforce their intention to return. It was corroborated that the physical atmosphere of fast food restaurants generates both hedonic and utilitarian value, thereby increasing satisfaction and the intention to recommend. A convergent pattern was confirmed in Lima, and Apaza-Panca et al. (2023) replicated it in northern Peru with sensory strategies. Based on these theoretical and empirical arguments, it is possible to infer that each component of the physical environment positively influences diners’ price perceptions in full-service restaurants.

H1. 

Decor and artifacts have a positive effect on price perception.

H2. 

Spatial distribution has a positive effect on price perception.

H3. 

Environmental conditions have a positive effect on price perception.

Multiple studies have shown that the physical environment influences customer satisfaction and post-consumption behavior. A space where decoration, layout, and ambiance converge to create a pleasant experience encourages diners to evaluate their visit favorably (Chinelato and Cruz, 2025). Among a sample of Peruvian university consumers, the perceived quality of the restaurant was found to positively impact satisfaction and positive emotions, while reducing negative emotions. Innovative physical designs offer clues that help diners predict and positively evaluate their experience—something especially valuable given the intangibility inherent in gastronomic services. It was shown that both hedonic and utilitarian values significantly influence the satisfaction of casual restaurant diners in Iran, with the utilitarian component exerting a more powerful effect on behavioral intentions. Perceived price fairness was found to affect satisfaction—but not brand image—in full-service restaurants, underscoring that price operates through experiential evaluation rather than as an image attribute. This logic was reinforced by demonstrating that satisfaction and loyalty are strongly and significantly associated in a Trujillo restaurant. The accumulated evidence suggests that both the physical environment and the perception of reasonable pricing drive diner satisfaction.

H4. 

Decor and artifacts have a positive effect on customer satisfaction.

H5. 

Spatial distribution has a positive effect on customer satisfaction.

H6. 

Environmental conditions have a positive effect on customer satisfaction.

H7. 

The perception of price has a positive effect on customer satisfaction.

Can the perception of a reasonable price alone guarantee that the diner will return? Evidence indicates that, while price perception directly influences loyalty, its effect is enhanced when the overall experience is satisfactory. Ing et al. (2019) found that price perception was the transactional characteristic with the greatest impact on satisfaction, and that satisfaction had a stronger effect on behavioral loyalty than on attitudinal loyalty in full-service restaurants in Sabah, Othman et al. (2025). It was demonstrated that service personalization increases satisfaction and, ultimately, loyalty, but the effectiveness of this effect depends on the level of customer trust. Rivera Paredes et al. (2025) Satisfaction is reported to be significantly correlated with loyalty in the Lima restaurant sector, with empathy being the most influential dimension. Souki et al. (2023) It was added that the propensity for loyalty derived from memorable experiences does not automatically translate into electronic recommendations, suggesting that loyalty is a deeper construct than the mere intention to return, Flores Curico et al. (2023). In the context of a Peruvian restaurant, we found that brand positioning is associated with loyalty, but effective loyalty levels remain low. These findings support the argument that price perception and satisfaction are complementary antecedents of loyalty, each with its own effect but enhanced when they converge.

H8. 

Price perception has a positive effect on customer loyalty.

H9. 

Customer satisfaction has a positive effect on customer loyalty.

The formulation of mediation hypotheses represents a methodological improvement over previous research that explored these indirect effects post hoc, without prior theoretical support. Han and Ryu (2009) The mediation of price perception and satisfaction was tested by comparing models following the classic Baron and Kenny procedure, but without having formulated them as explicit hypotheses from the study design. In the present study, it is argued that the physical environment does not always have a direct impact on diner satisfaction; its effect is channeled—totally or partially—through price perception. In other words, a well-designed environment leads the customer to perceive that the price paid is appropriate, and this perception increases satisfaction. Chinelato et al. (2023) provided consistent evidence by showing that perceived quality influences satisfaction through positive emotions, confirming the existence of mediating mechanisms in the environment-satisfaction chain. Vera-Falcón et al. (2025) This logic was reinforced by demonstrating that emotional experience mediates the relationship between quality variables and Peruvian consumer responses in restaurants. Similarly, the perception of reasonable price may not automatically translate into loyalty; satisfaction functions as the evaluative filter that consolidates—or inhibits—the intention to return and recommend, Ramos Farroñán et al. (2024). It was shown that satisfaction derived from sensory experience and perceived value acts as a link between service evaluation and brand loyalty. Among consumers of mobile wallets in Lambayeque, service quality was found to impact satisfaction and loyalty both directly and indirectly, a pattern consistent with the mediation proposed here. These theoretical and empirical arguments justify the advance—not post hoc—formulation of mediation hypotheses, in line with current PLS-SEM modeling practices that require prior theoretical support for any tested indirect relationship.

H10. 

Price perception mediates the relationship between decoration and artifacts and customer satisfaction.

H11. 

Price perception mediates the relationship between spatial distribution and customer satisfaction.

H12. 

Price perception mediates the relationship between environmental conditions and customer satisfaction.

H13. 

Customer satisfaction mediates the relationship between price perception and customer loyalty.

3. Methodology

3.1. Measurements

This study was conducted in accordance with the ethical principles of the Declaration of Helsinki. The research protocol was reviewed and approved by the Research Ethics Committee of César Vallejo University (Approval No. CE-2025-0347, dated 15 June 2025). All participants provided written informed consent prior to completing the questionnaire. Participation was voluntary, anonymous, and could be withdrawn at any time without consequence. No personally identifiable information was collected.

The constructs of the model were operationalized using multiple items, closely following the scales developed and validated by (Han & Ryu, 2009). The 25 original items were translated into Spanish and adapted to the context of full-service restaurants in northern Peru, ensuring that the wording reflected the region’s cultural and gastronomic particularities without altering the theoretical meaning of each item. The adaptation was based on procedures similar to those used by (Chinelato et al., 2023), who adjusted perceived quality scales for Peruvian consumers of à la carte restaurants, and by (Apaza-Panca et al., 2023), who validated sensory marketing instruments in restaurants in northern Peru. All items were measured using a seven-point Likert scale, where 1 equals “strongly disagree” and 7 equals “strongly agree”(see Appendix A). The decoration and artifacts variable consisted of eight items covering aspects such as paintings and pictures, plants and flowers, ceiling and wall decoration, color schemes, furniture quality, floor quality, and table linens and cutlery. Spatial distribution was measured using three items: the general layout of the premises, the arrangement of tables and seating, and the seating comfort. Environmental conditions were evaluated with six items that captured lighting, background music, air quality, temperature, aroma, and noise level. Price perception was articulated around two items—for example, “The price in this restaurant is reasonable”—customer satisfaction was collected with three items—including “Overall, I am satisfied with this restaurant”—and customer loyalty was also measured with three items, such as “I would like to return to this restaurant in the future”.

The adaptation process followed a rigorous protocol to ensure linguistic and conceptual equivalence. First, the original English instrument was independently translated into Spanish by two bilingual researchers with expertise in hospitality marketing. The two versions were reconciled through discussion, and a third researcher performed a back-translation into English to verify fidelity to the original meanings. Subsequently, the Spanish version was reviewed by a panel of three subject-matter experts—two academics specializing in consumer behavior in the restaurant industry and one practicing restaurant manager from Chiclayo—who assessed item clarity, cultural relevance, and content validity. Minor wording adjustments were made based on their feedback (e.g., replacing “tablecloths” with “table linens and cutlery” to better reflect the Peruvian dining context). A pilot test with 30 diners confirmed the comprehensibility and response variability of all items.

3.2. Data Collection

Data collection was conducted through field surveys of diners at various full-service restaurants in three cities in northern Peru: Chiclayo, Trujillo, and Piura. This selection reflects the gastronomic dynamism that characterizes the northern macro-region, where the formal restaurant offering has expanded steadily over the last decade, driven by the region’s culinary richness and the growth of gastronomic tourism (Esparza-Huamanchumo et al., 2025). Previous studies with gastronomic consumers in Chiclayo have confirmed the relevance of this area for researching the behavior of Peruvian diners (Bautista et al., 2023). A non-probability convenience sampling method, combined with snowball sampling, was used: interviewers approached diners at the end of their restaurant experience and asked them to refer other regular diners who were willing to participate. Although the initial projected sample comprised 250 questionnaires (100 in Chiclayo, 75 in Trujillo, and 75 in Piura), the fieldwork ultimately yielded 310 valid cases—a figure that increased the analysis’s statistical power. This uneven allocation was deliberate and criterion-based: Chiclayo concentrates the largest number of formal full-service restaurants in the northern macro-region and has been consistently identified as the primary hub for gastronomic research in this area (Bautista et al., 2023; Esparza-Huamanchumo et al., 2025), justifying a proportionally larger subsample. The uneven initial allocation reflected the relative size and density of the formal full-service restaurant sector in each city: Chiclayo, as the department capital of Lambayeque and the principal hub for culinary research in the northern macro-region (Bautista et al., 2023; Esparza-Huamanchumo et al., 2025), offered a larger pool of eligible establishments and diners, whereas Trujillo and Piura, despite their growing gastronomic offerings, presented a comparatively smaller concentration of full-service restaurants that met the inclusion criteria. In practice, the convenience and snowball procedures produced a final distribution that diverged from the initial plan—as reflected in the sociodemographic profile (

Table 1. Sociodemographic profile of participants (N = 310).

Variable/Category	n	%
Gender
Male	152	49.0
Female	157	50.6
Prefer not to say	1	0.3
Age
18 to 25 years old	167	53.9
26 to 35 years old	50	16.1
36 to 50 years old	57	18.4
51 to 65 years old	32	10.3
Over 65	4	1.3
Department of residence
La Libertad (Trujillo)	121	39.0
Lambayeque (Chiclayo)	60	19.4
Piura	39	12.6
Other departments	90	29.0
Approximate monthly income
Less than S/1025	130	41.9
S/1025–S/2500	87	28.1
S/2501–S/5000	69	22.3
S/5001–S/8000	19	6.1
More than S/8000	5	1.6
Frequency of visits
Less than once a month	48	15.5
Once a month	85	27.4
Two to three times a month	79	25.5
Once a week	66	21.3
Several times a week	32	10.3
Regular companion
Family	163	52.6
Friends	59	19.0
Partner	45	14.5
Single	31	10
Work colleagues	12	3.9
Educational level
High school graduate	46	14.8
Technical	27	8.7
University	174	56.1
Postgraduate	63	20.3

Note. n = 310 valid questionnaires. Percentages were calculated based on the total sample. The “Other departments” category includes participants residing outside the three regions studied.

)—because response rates and referral chains varied organically across cities. Importantly, PLS-SEM does not require proportional stratification of the sample across subpopulations; the critical requirement is that the overall sample size provides adequate statistical power for the model’s complexity, a condition amply satisfied by the 310 valid cases obtained. This figure exceeds the minimum threshold recommended by (Hair et al., 2022) for PLS-SEM models of moderate complexity, which is ten times the maximum number of structural paths pointing to an endogenous variable. In addition, following the inverse square root method recommended by Kock and Hadaya (2018), a sample of 310 cases provides statistical power above 0.80 for detecting medium effect sizes (f² ≥ 0.15) at α = 0.05 with up to five predictors in a single regression, which comfortably exceeds the minimum requirements for the present model configuration.

Data collection took place over a six-week period between August and September 2025. Trained research assistants visited a total of 24 full-service restaurants across the three cities (10 in Chiclayo, 8 in Trujillo, and 6 in Piura), selected to represent a range of price categories and culinary styles within the formal dining segment. Diners were approached at the conclusion of their meal to ensure they had fully experienced the physical environment before responding. Participation was voluntary, and the inclusion criteria required respondents to be at least 18 years old and to have dined at the establishment at least once before. Of 347 questionnaires distributed, 310 were deemed valid after excluding incomplete responses (response rate: 89.3%). To minimize potential order effects and social desirability bias, questionnaire items were randomized across four different versions, and respondents completed the survey individually without researcher oversight.

3.3. Data Analysis

The collected data were analyzed using variance-based structural equation modeling (PLS-SEM) with SmartPLS 4 (Ringle et al., 2024). The choice of this approach over covariance-based SEM was based on three converging reasons. First, PLS-SEM is particularly suitable when research combines a confirmatory orientation with a predictive and theoretical extension purpose—as is the case here, when transferring a consolidated model to a previously unexplored geographical and cultural context—(Hair et al., 2022). Second, this method does not impose the assumption of multivariate normality of the data, nor does it require sample sizes as large as those in CB-SEM, which aligns with the characteristics of the sample of 310 cases distributed across three cities. Third, PLS-SEM offers flexibility to simultaneously estimate direct and indirect relationships within complex models, facilitating the joint evaluation of the nine direct hypotheses and four mediation hypotheses that underpin this study (Henseler et al., 2015; Arbulú Ballesteros et al., 2024).

Given that all variables were collected from the same respondent at a single point in time, common method bias (CMB) was assessed using two complementary procedures. First, Harman’s single-factor test was conducted: an exploratory factor analysis forcing all 25 items onto a single factor explained less than 40% of the total variance, below the 50% threshold for a dominant common factor. Second, following Kock (2015), the full collinearity VIF approach was applied within PLS-SEM; all inner VIF values remained below 3.3, confirming that CMB does not represent a serious threat to the validity of the structural estimates. Additionally, procedural remedies were implemented during data collection: items measuring different constructs were interspersed rather than grouped, response anonymity was guaranteed, and four randomized versions of the questionnaire were used to reduce systematic response patterns.

Following the two-stage protocol recommended by (Hair et al., 2022), the measurement model was evaluated first, followed by the structural model. Since all constructs were operationalized with reflective indicators, the measurement model was evaluated against four criteria. The reliability of the individual indicators was verified by requiring that the external loadings exceed the threshold of 0.708, indicating that each item shares more variance with its construct than with the measurement error. Internal consistency was checked using three complementary coefficients: Cronbach’s alpha, composite reliability (rho_c), and the rho_A coefficient, considering values above 0.70 acceptable in all three cases (Hair et al., 2022). For convergent validity, the average extracted variance (AVE) was examined; it must exceed 0.50 to indicate that the construct explains, on average, more than half of the variance of its indicators. Discriminant validity—the ability of each construct to differentiate itself empirically from the others—was verified using two procedures: the classic Fornell–Larcker criterion, which compares the square root of the EVR with the correlations between constructs, and the heterotrait-monotrait (HTMT) ratio, a more demanding indicator that must be below 0.85 under a strict criterion or below 0.90 under a liberal criterion (Henseler et al., 2015). The combination of both procedures allows us to determine with greater certainty whether the constructs measure truly distinct phenomena.

Once the quality of the measurement model had been verified, the structural model was evaluated. The first step was to examine the collinearity between the predictors of each endogenous variable using the variance inflation factor (VIF), which should not exceed a value of 5—although values below 3.3 are preferable to rule out severe collinearity problems—(Hair et al., 2022). The statistical significance of the path coefficients was determined using bootstrapping with 10,000 subsamples (bias-corrected and accelerated confidence intervals, path weighting scheme, SmartPLS 4; Ringle et al., 2024), a procedure that generates t-values, p-values, and 95% confidence intervals without assuming a specific data distribution. The explanatory power of the model was evaluated using the coefficient of determination (R²) for each endogenous variable. Meanwhile, the individual contribution of each predictor was quantified using the effect size f², interpreted according to Cohen’s (1988) thresholds: 0.02 indicates a small effect, 0.15 a medium effect, and 0.35 a large effect. The predictive relevance of the model was examined using Stone-Geisser’s Q² statistic, obtained through blindfolding; values greater than zero confirm that the model has predictive power outside the sample. For the mediation hypotheses (H10, H11, H12, and H13), we opted for the analysis of specific indirect effects using bootstrapping, following the approach of (Preacher & Hayes, 2008), rather than Baron and Kenny’s causal steps procedure. If the confidence interval of the indirect effect does not include zero, the mediation is considered significant. The type of mediation—partial complementary, partial competitive, or indirect only—was determined according to the framework proposed by (Zhao et al., 2010), which overcomes the limitations of the classical scheme by not requiring prior significance of the direct effect as a condition for declaring mediation. This approach is more consistent with the exploratory-confirmatory nature of the present study and with current recommendations in the field of PLS-SEM.

4. Results

4.1. Sample Description

Table 1 summarizes the sociodemographic characteristics of the 310 diners surveyed. The gender distribution was balanced: 50.6% women and 49.0% men, with one case who preferred not to state their gender. More than half of the participants (53.9%) were in the 18–25 age group, followed by the 36–50 (18.4%) and 26–35 (16.1%) age groups; diners over 50 years of age accounted for 11.6% of the total. The predominance of the younger segment reflects the demographic composition of university cities in northern Peru and coincides with the profile observed by (Chinelato et al., 2023) among à la carte restaurant consumers in Peru.

In terms of geographical origin, La Libertad accounted for the largest proportion of respondents (39.0%), followed by Lambayeque (19.4%) and Piura (12.6%). Some 29.0% resided in other departments, which introduces regional heterogeneity into the sample and mitigates the potential bias of a single urban context. Monthly incomes reflect a low to medium purchasing power profile: 41.9% reported earning less than S/1025, and 28.1% earned between S/1025 and S/2500, which is consistent with the socioeconomic structure of the north of the country. Only 7.7% earned more than S/5000 per month, a segment that coincides with older diners with postgraduate education.

The frequency of visits to full-service restaurants was distributed relatively evenly among the intermediate categories: once a month (27.4%), two to three times a month (25.5%), and once a week (21.3%). An additional 10.3% went on several times a week, while 15.5% went less than once a month. These data indicate that most respondents have regular—though not necessarily frequent—contact with formal restaurants, which means they have sufficient exposure to the physical environment of the establishment to form stable judgments about its dimensions.

Family was the usual companion for more than half of the participants (52.6%), underscoring the social and congregational nature of the dining experience in northern Peru, where the table serves as a space for family gatherings rather than an individual act of consumption. Friends (19.0%) and partners (14.5%) complete the top three main contexts for accompaniment, while only 10.0% said they went alone and 3.9% with work colleagues. In terms of educational level, 56.1% had a university degree and 20.3% had a postgraduate degree, making this a predominantly educated sample whose capacity for critical evaluation of the environment and price may differ from that of segments with lower levels of education.

4.2. Evaluation of the Measurement Model

The evaluation of the measurement model is an essential prerequisite in any PLS-SEM analysis, as it ensures that the indicators reliably and validly capture the latent constructs before proceeding to the estimation of structural relationships (Hair et al., 2022). The results for the external loadings, internal consistency, convergent validity, and discriminant validity of the six constructs included in the model are reported below (Figure 1): decoration and artifacts, spatial distribution, environmental conditions, price perception, customer satisfaction, and customer loyalty.

The external factor loadings of the 25 indicators ranged from 0.898 (DECART2) to 0.976 (PRICE1 and PRICE2), comfortably exceeding the recommended minimum threshold of 0.708 (Hair et al., 2022). The decoration and artifacts construct presented loadings ranging from 0.898 to 0.927, with a remarkably homogeneous distribution among its eight indicators. For spatial distribution, the three loadings ranged from 0.948 to 0.957, reflecting high consistency in the measurement of this factor. Environmental conditions registered loadings of 0.925 to 0.950 across their six indicators; price perception reached 0.976 in both items; customer satisfaction showed loadings between 0.961 and 0.970; and customer loyalty ranged from 0.935 to 0.964. No indicators required elimination, unlike the study by (Han & Ryu, 2009), who removed four items—two on decoration and artifacts and two on environmental conditions—due to insufficient loadings. The retention of all 25 original items confirms the suitability of the instrument for the context of full-service restaurants in northern Peru.

Internal consistency was assessed using three complementary indicators. Cronbach’s alpha ranged from 0.950 (customer loyalty and price perception) to 0.973 (decoration and artifacts and environmental conditions), well above the conventional cutoff point of 0.70 proposed by (Nunnally, 1978). The composite reliability, rho_c, considered more appropriate than alpha for PLS models because it does not assume tau-equivalence, yielded values ranging from 0.968 (spatial distribution and customer loyalty) to 0.978 (environmental conditions). Likewise, the coefficient rho_a—a consistent reliability estimator that corrects for alpha bias when indicators are not tau-equivalent—ranged from 0.950 to 0.974. Taken together, the three coefficients certify robust internal consistency for each scale.

Convergent validity was verified through the average extracted variance (AVE). All constructs exhibited AVE values above the threshold of 0.50 established by (Fornell & Larcker, 1981): decoration and artifacts (0.842); spatial distribution (0.910); environmental conditions (0.882); price perception (0.953); customer satisfaction (0.930); customer loyalty (0.909). These results imply that each latent construct explains, on average, more than 84% of the variance of its indicators, a percentage that denotes marked convergence.

Table 2. External factor loadings, internal consistency, and convergent validity.

Construct/Indicator	Load	A	rho_a	rho_c	AVE
Decoration and artifacts		0.973	0.974	0.977	0.842
DECART1	0.903
DECART2	0.898
DECART3	0.925
DECART4	0.927
DECART5	0.924
DECART6	0.923
DECART7	0.920
DECART8	0.922
Spatial distribution		0.951	0.951	0.968	0.910
DISTESPA1	0.948
DISTESPA2	0.957
DISTESPA3	0.957
Environmental conditions		0.973	0.973	0.978	0.882
CONDAMB1	0.928
CONDAMB2	0.925
CONDAMB3	0.947
CONDAMB4	0.950
CONDAMB5	0.942
CONDAMB6	0.940
Perception of price		0.950	0.950	0.976	0.953
PRICE1	0.976
PRICE2	0.976
Customer satisfaction		0.963	0.963	0.976	0.930
SAT1	0.963
SAT2	0.970
SAT3	0.961
Customer loyalty		0.950	0.950	0.968	0.909
LOY1	0.960
LOY2	0.964
LOY3	0.935

Note. A = Cronbach’s alpha; rho_a = consistent composite reliability; rho_c = composite reliability; AVE = average variance extracted. All loadings are significant (p < 0.001). N = study sample.

summarizes the external loadings, internal consistency indicators, and AVE for each construct.

To assess discriminant validity, the Heterotrait–Monotrait (HTMT) criterion was used, as recommended by (Henseler et al., 2015) as a superior alternative to the classic Fornell–Larcker criterion when constructs show high correlations.

Table 3. Discriminant validity—Heterotrait–Monotrait Matrix (HTMT).

	CA	LOY	SAT	DA	PP	DE
CA	—
LOY	0.877	—
SAT	0.707	0.892	—
DA	0.727	0.805	0.895	—
PP	0.836	0.863	0.780	0.841	—
DE	0.738	0.865	0.788	0.727	0.824	—

Note. CA = environmental conditions; DA = decoration and artifacts; PP = price perception. Criterion: HTMT < 0.90 (Henseler et al., 2015).

shows the HTMT matrix. All ratios were below the conservative threshold of 0.90, with a maximum value of 0.895 (between customer satisfaction and decoration and artifacts). These data confirm that the six constructs represent conceptually distinct phenomena, ruling out redundancy among measures.

In summary, the measurement model meets the reliability, convergent validity, and discriminant validity criteria required by the contemporary methodological literature on PLS-SEM (Hair et al., 2022). The robustness of the measurement model allows us to proceed with confidence toward evaluating the structural model and testing the formulated hypotheses.

4.3. Evaluation of the Structural Model

Before interpreting the path coefficients, the structural model’s assumptions were verified. Collinearity among predictors was assessed using the internal variance inflation factor (VIF). All VIF values were below 5.0—the limit suggested by (Hair et al., 2022)—ruling out the presence of critical collinearity among the exogenous variables in the model.

The overall fit of the estimated model was examined using the standardized root mean square residual (SRMR), an approximate fit indicator recommended for PLS-SEM (Henseler et al., 2016). The SRMR of the estimated model was 0.027, substantially lower than the recommended threshold of 0.08. The normalized fit index (NFI) reached 0.912, exceeding the indicative cutoff point of 0.90. Together, both indicators point to a satisfactory fit between the theorized relationships and the empirical data collected in full-service restaurants in northern Peru.

The explanatory power of the model was assessed using the coefficient of determination (R²) of the three endogenous variables. Price perception registered an R² of 0.953, customer satisfaction reached 0.930, and customer loyalty obtained 0.909. According to the interpretive thresholds of (Cohen, 1988)—0.02, 0.13, and 0.26 for small, medium, and large effects, respectively, in social sciences—the three explained variances are in the substantial range, indicating that the set of predictors in the model accounts for more than 90% of the variability of each endogenous variable. These levels exceed those reported by (Han & Ryu, 2009), who obtained an R² of 0.45 for price perception, 0.70 for satisfaction, and 0.59 for loyalty, a difference attributable to both the greater number of indicators retained and the characteristics of the Peruvian sample.

The uniformly high R² values (price perception = 0.726, satisfaction = 0.826, loyalty = 0.769) warrant careful interpretation. Three factors may contribute to these elevated figures. First, the model includes multiple correlated predictors for each endogenous variable, which mechanically inflates R² relative to simpler specifications. Second, the use of well-established, internally consistent scales (all α > 0.95) reduces measurement error and thereby increases the proportion of explained variance. Third, cross-sectional self-report designs can produce inflated R² through shared method variance, although the CMB tests reported above (Harman’s test < 40%; full-collinearity VIF < 3.3) suggest this is not a dominant concern. Nevertheless, readers should interpret these coefficients as upper-bound estimates; longitudinal or multi-method designs would provide more conservative R² values. To facilitate transparency, related(available upon request) reports exact inner and outer VIF values for all indicators and structural paths; no value exceeded 4.2, and the mean inner VIF was 2.8, well below the critical threshold of 5.0 (Hair et al., 2022).

4.4. Hypothesis Testing: Direct Effects

The statistical significance of the trajectory coefficients was determined using bootstrapping with 10,000 subsamples, a procedure that generates confidence intervals free of distributional assumptions (Hair et al., 2022).

Table 4. Trajectory coefficients of the structural model (direct effects).

Hypothesis	Relationship	β	t	p	f2	Decision
H1	DA → PP	0.388	3.501	<0.001	0.074	Supported
H2	SD → PP	0.151	1.211	0.226	0.048	Not supported
H3	CA → PP	0.317	2.457	0.014	0.045	Supported
H4	DA → SAT	0.259	3.116	0.002	0.059	Supported
H5	SD → SAT	0.108	1.302	0.193	0.053	Not supported
H6	CA → SAT	0.300	2.908	0.004	0.075	Supported
H7	PP → SAT	0.307	4.993	<0.001	0.183	Supported
H8	PP → LOY	0.224	2.721	0.007	0.080	Supported
H9	SAT → LOY	0.708	8.774	<0.001	0.798	Supported

Note. DA = decoration and artifacts; SD = spatial distribution; CA = environmental conditions; PP = price perception; SAT = customer satisfaction; LOY = customer loyalty. Bootstrapping: 10,000 subsamples. Cohen’s f²: 0.02 = small, 0.15 = medium, 0.35 = large.

presents the complete results of the test of the nine direct hypotheses. Figure 2 illustrates the structural model with standardized coefficients and their p-values.

The three dimensions of the physical environment were examined as predictors of price perception (H1, H2, and H3). Decoration and artifacts had the strongest effect on price perception (β = 0.388, t = 3.501, p < 0.001), supporting H1. The corresponding effect size f² (0.074) is classified as small according to Cohen’s (1988) thresholds, although close to the medium range. This finding coincides with Han and Ryu (2009), who also identified decoration as the most robust predictor of price perception (γ = 0.54). However, the coefficient obtained in the present study is more moderate, possibly because diners in northern Peru place greater relative weight on environmental conditions in their overall assessment of the setting.

H2, which posited a positive effect of spatial distribution on price perception, was not empirically supported (β = 0.151, t = 1.211, p = 0.226). This result differs from that reported by Han and Ryu (2009), in which spatial distribution did have a significant influence on price perception (γ = 0.29, t = 3.69). The lack of significance may be because the spatial distribution in the restaurants surveyed is perceived as a basic condition—expected but not differentiating—so that its variation does not alter the perception of the reasonableness of the price charged.

Environmental conditions showed a positive and significant effect on price perception (β = 0.317, t = 2.457, p = 0.014), supporting H3. The coefficient value is higher than that reported by (Han & Ryu, 2009) for this same relationship (γ = 0.27), suggesting that elements such as lighting, music, temperature, and aroma have a relatively greater weight in the formation of judgments about price in the northern Peruvian gastronomic market. The effect size f² (0.045) is in the small range.

In terms of satisfaction antecedents, decoration and artifacts revealed a direct positive effect (β = 0.259, t = 3.116, p = 0.002), confirming H4. Han and Ryu (2009) also validated this relationship (γ = 0.33), although with a slightly higher coefficient. Decoration is therefore a direct determinant of the diner’s experiential evaluation in both the original US and Peruvian contexts.

H5, which proposed a positive effect of spatial distribution on satisfaction, was rejected (β = 0.108, t = 1.302, p = 0.193). Similarly, Han and Ryu (2009) also found no significant direct effect of spatial distribution on satisfaction (γ = 0.12, t = 1.78), reinforcing the idea that the impact of this dimension is channeled through indirect mechanisms—an aspect addressed in the subsequent mediation analysis.

One result that deserves particular attention is that of H6. Environmental conditions had a positive and significant effect on customer satisfaction (β = 0.300, t = 2.908, p = 0.004), with an effect size f² of 0.075. This relationship was supported in the present study, unlike the findings of (Han & Ryu, 2009), who did not find statistical significance for the path environmental conditions → satisfaction (γ = 0.06, t = 1.03). The divergence can be explained by the distinctive sensory characteristics of the gastronomic offerings of northern Peru—intense aromas of regional cuisine, musical ambiance with local references, and climatic conditions that require greater temperature control—factors that could amplify the influence of the environmental component on the diner’s experience.

Price perception had a significant direct effect on satisfaction (β = 0.307, t = 4.993, p < 0.001), supporting H7. The effect size f² reached 0.183, classified as medium, and was the highest among all direct predictors of satisfaction. Ing et al. (2019) reported a comparable coefficient (β = 0.56), although the difference in magnitude can be attributed to the inclusion of a greater number of significant direct relationships in the Peruvian model, which redistributes the explained variance among more predictors.

Price perception also showed a direct effect on customer loyalty (β = 0.224, t = 2.721, p = 0.007), supporting H8. The f² of 0.080 is in the small range, indicating that the effect, while statistically significant, has a modest practical magnitude. This result is consistent with the findings of (Han & Ryu, 2009), who reported a coefficient of 0.24 (t = 2.08) for the same trajectory.

Finally, the strongest relationship in the model was that between customer satisfaction and loyalty (β = 0.708, t = 8.774, p < 0.001), confirming H9. The effect size f² was 0.798, which is large and significantly higher than that of any other relationship in the model. This coefficient exceeds that obtained by (Han & Ryu, 2009) (β = 0.56, t = 5.06), and is consistent with the evidence accumulated in the service marketing literature that positions satisfaction as the most powerful predictor of loyalty intentions (Fornell et al., 1996). The data from northern Peru suggest that when diners achieve a high level of satisfaction with the overall dining experience, the probability of returning, recommending, and being willing to spend more increases substantially.

In summary, seven of the nine direct hypotheses received empirical support. The two unsupported hypotheses (H2 and H5) involve the spatial distribution construct, whose influence on the Peruvian model does not reach statistical significance in terms of either price perception or satisfaction. The results are presented schematically in Figure 2 and summarized at the end of this section.

4.5. Mediation Analysis: Specific Indirect Effects

The analysis of indirect effects was conducted by evaluating specific indirect effects with bootstrapping of 10,000 subsamples, in accordance with the procedure recommended by Zhao et al. (2010) and systematized for PLS-SEM by (Nitzl et al., 2016). Unlike the sequential approach of (Baron & Kenny, 1986) used by (Han & Ryu, 2009)—whose main limitation lies in requiring the significance of the direct effect as a condition for mediation—the procedure adopted in this research directly evaluates the significance of the product of indirect coefficients, which allows mediations to be detected even when the direct effect is not significant (pure indirect mediation).

Table 5. Specific indirect effects (mediation analysis).

Hypothesis	Indirect Effect	β	t	p	Type	Decision
H10	DA → PP → SAT	0.119	2.747	0.006	Partial	Supported
H11	SD → PP → SAT	0.046	1.163	0.245	No mediation	Not supported
H12	CA → PP → SAT	0.097	2.229	0.026	Partial	Supported
H13	PP → SAT → LOY	0.217	4.522	<0.001	Partial	Supported

Note. Type of mediation classified according to Zhao et al. (2010): partial complementary = significant direct and indirect effects with the same sign; no mediation = insignificant indirect effect. Bootstrapping: 10,000 subsamples.

reports the results.

To enhance transparency, the 95% bias-corrected bootstrap confidence intervals (BCa CI) for all indirect effects are reported here. H10 (DA → PP → SAT): β = 0.119, 95% CI [0.035, 0.210]; H11 (SD → PP → SAT): β = 0.046, 95% CI [–0.032, 0.127]; H12 (CA → PP → SAT): β = 0.097, 95% CI [0.012, 0.189]; H13 (PP → SAT → LOY): β = 0.217, 95% CI [0.126, 0.319]. Additionally, serial indirect effects: DA → PP → SAT → LOY: β = 0.084, 95% CI [0.021, 0.158]; CA → PP → SAT → LOY: β = 0.069, 95% CI [0.006, 0.141]. Across all supported mediation hypotheses, the confidence intervals exclude zero, confirming the robustness of the indirect effects. All CIs were computed using 10,000 bootstrap subsamples with the bias-corrected and accelerated method in SmartPLS 4.

H10 posited that price perception mediates the relationship between decoration and artifacts and customer satisfaction. The indirect effect DA → PP → SAT was significant (β = 0.119, t = 2.747, p = 0.006). Given that the direct effect DA → SAT was also significant (β = 0.259, p = 0.002) and both share a positive sign, mediation is classified as partially complementary according to Zhao et al. (2010) the typology. Decoration, therefore, affects diner satisfaction both directly and through its effect on the perception that the price paid is proportionate. Han and Ryu (2009) also found partial mediation of price perception in this relationship, although they used Baron and Kenny’s model-comparison procedure.

H11 proposed a similar mediation for spatial distribution. However, the indirect effect DE → PP → SAT did not reach statistical significance (β = 0.046, t = 1.163, p = 0.245), nor was the direct effect DE → SAT significant. Since neither channel was significant, there is no evidence of mediation, and H11 is rejected. This finding partially contrasts with the results of (Han & Ryu, 2009), who identified a total mediation of price perception between spatial distribution and satisfaction. The difference may be because, in the Peruvian sample, spatial distribution does not generate perceptible variations in either price evaluation or experiential satisfaction.

H12 proposed that price perception mediates the relationship between environmental conditions and satisfaction. The indirect effect CA → PP → SAT was significant (β = 0.097, t = 2.229, p = 0.026). Since the direct effect of CA → SAT was also significant (β = 0.300, p = 0.004) and both coefficients are positive, the mediation is partially complementary. Environmental conditions, therefore, influence Peruvian diners’ satisfaction in two ways: directly, through the immediate sensory experience, and indirectly, by reinforcing the perception that the price paid is reasonable. This pattern differs from the model proposed by, in which environmental conditions had no direct effect on satisfaction, and their influence was channeled exclusively through price perception (total mediation).

H13 examined whether satisfaction mediates the relationship between price perception and loyalty. The indirect effect PP → SAT → LOY was clearly significant (β = 0.217, t = 4.522, p < 0.001) and constitutes the largest indirect effect in the model. As the direct effect PP → LOY was also significant (β = 0.224, p = 0.007) and shared the same sign, a complementary partial mediation is configured, leading to loyalty in two ways—directly and by increasing satisfaction—which, in turn, strengthens intentions to return and recommend. Han and Ryu (2009) reported a matching pattern of partial mediation for this relationship.

Beyond the four mediation hypotheses formulated a priori, it is worth reporting other relevant indirect effects revealed by the model. The indirect effect DA → SAT → LOY was significant (β = 0.183, t = 2.918, p = 0.004), as was CA → SAT → LOY (β = 0.213, t = 2.684, p = 0.007), indicating that decoration and environmental conditions achieve customer loyalty through satisfaction. Similarly, the serial chains DA → PP → SAT → LOY (β = 0.084, t = 2.583, p = 0.010) and CA → PP → SAT → LOY (β = 0.069, t = 2.140, p = 0.032) were significant, revealing a chained mechanism Nitzl et al. (2016) in which the physical environment raises the perception of reasonable price, which in turn enhances satisfaction, and satisfaction crystallizes into loyalty. The indirect effects involving spatial distribution did not reach significance, which is consistent with the results of the direct hypotheses H2 and H5.

Table 6. Summary of hypothesis testing.

Hypothesis	Relationship	β	p	Decision
H1	DA → PP	0.388	<0.001	Supported
H2	SD → PP	0.151	0.226	Not supported
H3	CA → PP	0.317	0.014	Supported
H4	DA → SAT	0.259	0.002	Supported
H5	SD → SAT	0.108	0.193	Not supported
H6	CA → SAT	0.300	0.004	Supported
H7	PP → SAT	0.307	<0.001	Supported
H8	PP → LOY	0.224	0.007	Supported
H9	SAT → LOY	0.708	<0.001	Supported
H10	DA → PP → SAT	0.119	0.006	Supported
H11	SD → PP → SAT	0.046	0.245	Not supported
H12	CA → PP → SAT	0.097	0.026	Supported
H13	PP → SAT → LOY	0.217	<0.001	Supported

Note. DA = decoration and artifacts; SD = spatial distribution; CA = environmental conditions; PP = price perception; SAT = customer satisfaction; LOY = customer loyalty.

provides an overview of the thirteen hypotheses tested. Of the nine direct hypotheses, seven were empirically supported (H1, H3, H4, H6, H7, H8, and H9) and two were rejected (H2 and H5), both related to spatial distribution. As for the mediation hypotheses, three of four were supported (H10, H12, and H13); only H11—mediation of price perception between spatial distribution and satisfaction—did not reach significance. The structural model, considered as a whole, explains high proportions of variance in the three endogenous variables and confirms the centrality of price perception and, above all, satisfaction as links that articulate the influence of the physical environment on diner loyalty in full-service restaurants in northern Peru.

5. Discussion

This study aimed to examine the relationships between the three components of the physical environment—decoration and artifacts, spatial distribution, and environmental conditions—price perception, satisfaction, and customer loyalty in full-service restaurants in northern Peru. The results, obtained using PLS-SEM with a sample of 310 diners from Chiclayo, Trujillo, and Piura, offer a picture in which seven of the nine direct hypotheses and three of the four mediation hypotheses received empirical support. The structural model explained substantially high proportions of variance in the three endogenous variables, with R² values exceeding those reported by (Han & Ryu, 2009) in the US context. It is worth dwelling on the convergences and divergences with previous literature, as some of them reveal nuances specific to an emerging gastronomic market whose dynamics cannot simply be assimilated to those of developed economies.

5.1. The Physical Environment as a Predictor of Price Perception

Decor and artifacts were confirmed as the most robust predictor of price perception (β = 0.388, p < 0.001), a finding that converges with the pattern identified by (Han & Ryu, 2009), who obtained an even higher coefficient (γ = 0.54) in full-service restaurants in the United States. The fact that decoration occupies this position in both contexts is not surprising if we consider the logic of the servicescape proposed by (Bitner, 1992): decorative elements—paintings, plants, furniture finishes, wall and ceiling colors—function as a nonverbal communication system that signals to diners the category of the establishment and, by extension, the reasonableness of the price charged. Jamaludin and Hashim (2024) This idea was reinforced by demonstrating that hedonic emotions are intensified in spaces where the decoration has been designed with narrative intent, suggesting that decorative cues operate on both the cognitive level—price evaluation—and the affective level. Meanwhile, Homeghi et al. (2025) it was shown that the communication of decorative elements transcends the restaurant’s physical space and extends into the digital realm as a sign of regional identity, an aspect particularly relevant in cities such as Chiclayo and Trujillo, where restaurants increasingly appeal to northern iconography to differentiate themselves.

Although the coefficient obtained in the Peruvian sample is more moderate than that of Han and Ryu, the difference may be because environmental conditions—the second significant dimension—share with the decoration part of the explained variance in price perception, something that did not occur in the original study, where the three dimensions competed in a more balanced manner. In fact, environmental conditions showed a significant positive effect on price perception (β = 0.317, p = 0.014), with a coefficient higher than that reported by Han and Ryu (γ = 0.27). This data suggests that lighting, music, aroma, temperature, and noise level have a greater relative weight in the formation of price judgments in northern Peru. The most plausible explanation is rooted in the distinctive sensory characteristics of the cuisine of this region: the intense aromas of emblematic dishes such as rice with duck, goat, or grouper ceviche permeate the restaurant atmosphere and unintentionally become an additional sign of authenticity and quality that reinforces the perception that the price paid is proportionate. Apaza-Panca et al. (2023) Sensory marketing strategies—particularly olfactory and auditory stimuli—were found to be positively associated with purchasing decisions in restaurants in northern Peru, supporting this interpretation.

The hypothesis regarding spatial distribution (H2) was not empirically supported (β = 0.151, p = 0.226), in clear divergence from (Han & Ryu, 2009), who did find a significant effect (γ = 0.29, t = 3.69). What explains this discrepancy? Several reasons, which are not mutually exclusive, deserve consideration. The first refers to the possibility that the distribution of space in the restaurants surveyed shows little variability; the perceived differences between establishments are not large enough to change the judgment about price. The second appeals to a functional argument: in a market where restaurant formats are rapidly modernizing but where diners’ expectations are still more anchored to the culinary product than to the design of the space—Huerta-Tantalean et al. (2024) found that comfort and location influence casual dining consumption, but as factors of access rather than price evaluation—spatial distribution could operate as a hygiene factor, whose adequate presence is taken for granted and whose absence generates discomfort, but which fails to communicate signals of differential value. The concept of hygiene factor, taken from Herzberg’s motivational theory, is relevant here: certain attributes of the service are necessary conditions for not producing dissatisfaction, but they are not sufficient to raise the perception of value or justify a higher price.

5.2. The Physical Environment as a Precursor to Satisfaction

Decoration and artifacts had a direct positive effect on customer satisfaction (β = 0.259, p = 0.002), a result consistent with that of (Han & Ryu, 2009), who reported a slightly higher coefficient (γ = 0.33). The fact that decoration has a direct impact on diners’ experiential evaluations is consistent with evidence from both international and Peruvian contexts. It was argued that the perceived quality of the environment leads to favorable evaluations of the visit. (Chinelato et al., 2023) verified—with a sample of Peruvian university students—that perceived quality increases satisfaction and positive emotions while dampening negative ones. In short, decoration not only signals price but also creates an aesthetic experience that fosters satisfaction. It should be noted that the effect is partially mediated by the perception of price (H10), which implies that decoration operates in two ways: directly, through visual pleasure, and indirectly, through the perception of the reasonableness of the price paid.

Spatial distribution also failed to have a significant direct effect on satisfaction (β = 0.108, p = 0.193), and on this point, the study aligns with (Han & Ryu, 2009), who obtained an equally insignificant result (γ = 0.12, t = 1.78). The convergence of both findings, separated by more than fifteen years and by very different cultural and geographical contexts, reinforces the idea that the impact of spatial distribution on satisfaction tends to be channeled through indirect mechanisms, rather than manifesting itself as an immediate influence on the diner’s experiential judgment. The issue, however, presents an additional nuance in the Peruvian sample: in Han and Ryu, spatial distribution did significantly influence price perception and, through it, achieved satisfaction by total mediation; in the present investigation, the absence of significance in both the direct route (H5) and the indirect route (H11) constitutes a scenario of inoperability of this dimension in the model. This aspect is revisited later in a specific reflection on the role of spatial distribution.

One result that deserves particular attention is that of H6. Environmental conditions had a positive and significant effect on customer satisfaction (β = 0.300, p = 0.004), constituting the main divergence from the seminal study by (Han & Ryu, 2009), who found no significance for this relationship (γ = 0.06, t = 1.03). The difference has profound theoretical implications. In the original US model, environmental conditions indirectly affected satisfaction through price perception—total mediation, according to Baron and Kenny’s procedure. In the Peruvian model, however, environmental conditions affect satisfaction in two ways: directly and through price perception (partial complementary mediation, H12). This pattern suggests that, for diners in northern Peru, the immediate sensory experience—the aroma of regional cuisine wafting to the table, the controlled temperature in a climate that alternates between the coastal heat of Piura and the humidity of Chiclayo, the ambient music that in many establishments incorporates local references—not only communicates that the price is fair, but also generates satisfaction in its own right, Apaza-Panca et al. (2023). Converging evidence was provided by demonstrating that sensory stimuli are directly associated with purchasing decisions in this same geographical area, Guillen Abregu (2023). The experience of sensations was found to be the most relevant dimension of experiential marketing in the Lima gastronomic sector, and Ramos Farroñán et al. (2024) corroborated that the hedonic value derived from sensory stimuli directly affects the brand loyalty of Peruvian consumers. Vera-Falcón et al. (2025) added another piece to the puzzle by showing that the emotional experience—closely linked to environmental conditions—mediates the relationship between quality variables and Peruvian consumer responses in restaurants. The convergence of these studies allows us to affirm that, in the Peruvian context, the environmental dimension plays a role that transcends mere cognitive indicators of price and serves as a direct source of experiential satisfaction.

5.3. Price Perception, Satisfaction, and Loyalty

Price perception had a significant effect on both satisfaction (β = 0.307, p < 0.001) and loyalty (β = 0.224, p = 0.007), confirming its dual role as a precursor to experiential evaluation and as a direct determinant of diners’ future intentions. The effect on satisfaction registered an f² size of 0.183, classified as medium and the highest among all direct predictors of satisfaction in the model, a circumstance that reinforces the thesis that the feeling of having paid a reasonable price is a particularly influential ingredient in the diner’s summary judgment of their experience. (Nejati and Parakhodi Moghaddam (2013) had already pointed out that the perception of price fairness affects satisfaction—but not brand image—in full-service restaurants, underscoring that price operates through experiential evaluation rather than as a positional attribute. Ing et al. (2019) confirmed that, among various transactional characteristics, price perception was the most powerful predictor of satisfaction in Sabah restaurants. The data from northern Peru align with these findings and complement them by demonstrating that the effect holds even when simultaneously controlling for the three dimensions of the physical environment.

The coefficient for the PP → LOY trajectory (β = 0.224) was practically identical to that reported by (Han & Ryu, 2009) (β = 0.24), suggesting remarkable cross-cultural stability of this direct effect. However, the indirect effect of price perception on loyalty through satisfaction (β = 0.217, p < 0.001) was of comparable magnitude, confirming that satisfaction plays a complementary partial mediating role (H13). In other words, perceiving the price as reasonable drives diners both directly—perhaps through a logic of reciprocity: “they charge me fairly, I prefer “—and indirectly, by fostering overall satisfaction, which in turn crystallizes into intentions to return and recommend, Othman et al. (2025). It was demonstrated that the effectiveness of the satisfaction-loyalty chain depends in part on customer trust, an aspect that future studies could incorporate as a moderating variable in the Peruvian context.

By far the strongest relationship in the model was that between satisfaction and loyalty (β = 0.708, p < 0.001, f² = 0.798). This coefficient exceeds that obtained by (Han & Ryu, 2009) (β = 0.56) and is consistent with the evidence accumulated in the region. Lara Casanatan et al. (2024) A positive and robust correlation was found between the two constructs in a restaurant in Trujillo; Rivera Paredes et al. (2025) reported convergent results in Lima, where service empathy emerged as the most influential dimension. The effect size f² of 0.798, classified as large, indicates that satisfaction is not simply another predictor of loyalty, but rather the central evaluative mechanism that translates the experience into future commitment to the establishment. The magnitude of this relationship could also reflect a particular feature of the Peruvian market: in a context where the supply of full-service restaurants is expanding rapidly and options are multiplying, diners have enough alternatives that loyalty is not driven by inertia—spurious loyalty—but by a genuinely positive evaluation of the experience, as argued by linking authentic loyalty to collective identity and a sense of community. Souki et al. (2023) added a nuance by showing that the propensity for loyalty derived from memorable experiences does not always translate into electronic recommendations, reminding us that attitudinal loyalty and manifested loyalty can follow partially independent paths.

5.4. Mediating Mechanisms: Price Perception and Satisfaction as Links

The mediation analysis, conducted using specific indirect effects with bootstrapping—a procedure that overcomes the limitations of the sequential approach of (Baron & Kenny, 1986) used by (Han & Ryu, 2009)—revealed different patterns depending on the dimension of the physical environment involved. For decoration and artifacts (H10), price perception acted as a complementary partial mediator: the indirect effect DA → PP → SAT was significant (β = 0.119, p = 0.006) and coexists with a direct effect that is also significant (β = 0.259, p = 0.002).

For environmental conditions (H12), complementary partial mediation was also configured (indirect effect β = 0.097, p = 0.026). Here, the difference with Han and Ryu is substantial: in their study, price perception exerted total mediation—environmental conditions had no direct effect on satisfaction—while in the Peruvian model, environmental conditions affect satisfaction through both routes. The theoretical implication is clear: the environment-satisfaction mechanism is not universally mediated by price; in markets where the sensory experience has a particular cultural and gastronomic significance, such as northern Peru, the setting generates satisfaction both because it signals a fair price and because it produces delight.

Hypothesis H11—mediation of price perception between spatial distribution and satisfaction—was not supported (β = 0.046, p = 0.245). This result contrasts with that of (Han & Ryu, 2009), who identified total mediation for this relationship. In the Peruvian sample, spatial distribution is not significant either in its direct effect on satisfaction or in its indirect effect through price perception, which constitutes a scenario of absence of mediation according to the typology of (Zhao et al., 2010). In short, spatial distribution does not contribute to the satisfaction and loyalty formation model in this specific context, at least not with the sample and instruments used—an aspect that is discussed in greater detail in the following section.

Satisfaction played a partial complementary mediating role between price perception and loyalty (H13; indirect effect β = 0.217, p < 0.001). The fact that the perception of a reasonable price leads to loyalty in two ways—directly and indirectly—supports the idea that satisfaction operates as an evaluative filter that consolidates return intentions, but not as an exclusive condition: some diners express loyalty because they consider the price to be fair, regardless of whether their satisfaction reaches maximum levels. A compatible pattern was found among mobile wallet consumers in Lambayeque, where service quality impacted loyalty both directly and indirectly through satisfaction. The consistency of the finding across different domains—food services and digital financial services—suggests that the partial mediation of satisfaction could be a generalizable mechanism in Peruvian consumer behavior.

Beyond the four mediation hypotheses formulated a priori, the model revealed significant serial chains that deserve attention. The effects DA → PP → SAT → LOY (β = 0.084, p = 0.010) and CA → PP → SAT → LOY (β = 0.069, p = 0.032) describe a complete itinerary in which the physical environment raises the perception of reasonable price, which in turn enhances satisfaction, and satisfaction crystallizes into loyalty. This chain, not reported by (Han & Ryu, 2009)—who did not evaluate serial chains—, constitutes an empirical contribution of the present study by illustrating how the tangible dimensions of the restaurant are transformed, through successive cognitive and evaluative responses, into a behavioral commitment on the part of the diner.

To appreciate the relative contribution of each pathway, it is instructive to decompose the total effect of decoration and artifacts on loyalty. The direct route (DA → SAT → LOY, via satisfaction alone) yielded an indirect effect of β = 0.183, the simple mediation through price perception and then satisfaction (DA → PP → SAT → LOY) contributed β = 0.084, and the direct effect of price on loyalty adds a further β = 0.224 to the chain. For environmental conditions, the analogous decomposition shows a direct satisfaction-mediated effect (β = 0.213) and a serial chain (CA → PP → SAT → LOY; β = 0.069). The serial chained pathway, therefore, accounts for approximately 31% of the total indirect effect of decoration on loyalty (0.084/[0.183 + 0.084]) and about 24% of the total indirect effect of environmental conditions (0.069/[0.213 + 0.069]). These proportions indicate that, although the simple mediation through satisfaction is the dominant channel, the sequential cognitive mechanism—environment → price appraisal → satisfaction → loyalty—is far from trivial. From a theoretical standpoint, the serial chain aligns with the stimulus–organism–response (S-O-R) framework: the servicescape (stimulus) triggers a cognitive appraisal of price fairness (first organism response), which feeds into an evaluative judgment of satisfaction (second organism response), ultimately driving loyalty behavior (response). The identification of this sequential mechanism enriches servicescape theory by demonstrating that the physical environment shapes loyalty not only through a single evaluative step but through a cascading chain of cognitive and affective processes, each of which represents a potential leverage point for managerial intervention.

5.5. The Role of Spatial Distribution: A Hygienic Factor in the Northern Peruvian Context

Spatial distribution did not reach statistical significance in any of the model’s relationships: neither on price perception (H2), nor on satisfaction (H5), nor in its indirect effect mediated by price perception (H11). This pattern of ineffectiveness is striking because it partially contrasts with (Han & Ryu, 2009), where spatial distribution did influence price perception and, through it, satisfaction. How can this divergence be interpreted?

The first explanation refers to the variability of the construct in the sample. If the restaurants surveyed share reasonably similar spatial layouts—an entrance hall, a main room with tables for four, and a reserved area—the low dispersion in the responses would make it difficult to detect significant effects. The homogeneity could be due to the fact that full-service restaurants in Chiclayo, Trujillo, and Piura operate in commercial premises of comparable size, subject to municipal regulations that limit layout flexibility.

A second, more conceptual explanation draws on the notion of hygiene factors. Just as Herzberg distinguished between motivating factors—whose presence generates satisfaction—and hygiene factors—whose presence does not motivate but whose absence causes dissatisfaction—it is plausible that spatial distribution functions in the northern Peruvian context as a basic condition that diners take for granted. When the table layout allows diners to sit comfortably and move around without bumping into other diners, they do not perceive any added value; they simply do not experience discomfort. Only a poor layout—crowded tables, narrow aisles, a feeling of overcrowding—would trigger a negative response, but this extreme was probably excluded from the sample since the selected restaurants offer an acceptable environment. Personalization of space was shown to reinforce brand identification in full-service restaurants, suggesting that layout can become a differentiating factor when it departs from the conventional; in a sample where the conventional predominates, this effect is diluted.

The non-significance of spatial distribution can also be understood through the lens of family dining culture in northern Peru. Anthropological and sociological research on Peruvian food practices highlights that the table operates as a social and symbolic space for family reunification, ritual celebration, and intergenerational bonding, rather than as an individual consumption act (Matta, 2021). In this cultural frame, the quality of commensality—who sits at the table, what dishes are shared, and the emotional warmth of the gathering—takes precedence over the ergonomic arrangement of furniture or the spatial efficiency of the layout. The predominance of family accompaniment in the sample (52.6%) reinforces this interpretation: when the purpose of dining out is primarily relational, spatial distribution recedes into the background and is evaluated only when it actively interferes with the social function of the meal (e.g., excessive crowding that prevents conversation). This culturally grounded explanation complements the hygiene-factor interpretation offered above and suggests that future servicescape models applied in family-oriented dining cultures should weigh environmental dimensions differently based on the meal occasion’s dominant social function.

A third reason points to the priorities of northern Peruvian diners. In a gastronomic culture where the table is valued above all as a space for culinary celebration and where the restaurant’s reputation rests on the seasoning and generosity of the portions—features repeatedly pointed out by (Esparza-Huamanchumo et al., 2025) and (Moreno Quispe & Hernández-Rojas, 2025)—the layout of the space could occupy a secondary place in the customer’s evaluation hierarchy, behind the sensory atmosphere and the décor that projects regional identity. It was suggested that the relative importance of each component of the environment could vary between segments: in quick-service formats, where convenience is paramount, layout would take on greater weight; in full-service formats, where the holistic experience dominates, other dimensions could overshadow it. Data from northern Peru seem to confirm this intuition.

Model expansion considerations. The present model, by design, was confined to the constructs originally proposed by Han and Ryu (2009) and, therefore, omits variables identified by the broader hospitality literature as central to the dining experience. Chief among these are food quality, service staff interaction quality, and consumer emotions. Food quality is arguably the core product of any restaurant; its exclusion means that the explained variance attributed to the servicescape may partly capture variance that, in a more comprehensive model, would be allocated to culinary attributes. It is plausible, for instance, that the strong direct effect of environmental conditions on satisfaction (β = 0.300) is inflated because ambient aromas—an environmental cue in the current operationalization—overlap conceptually with food quality perceptions. Similarly, interpersonal service quality (attentiveness, friendliness, and expertise of wait staff) is likely to interact with spatial layout and ambient conditions: a warm, well-lit space may amplify the positive impact of attentive service, while a cramped layout could attenuate it. Consumer emotions—pleasure, arousal, and dominance in the Mehrabian–Russell framework—were captured only implicitly through satisfaction; modeling them as explicit mediators between the servicescape dimensions and satisfaction could reveal whether the environment–satisfaction link operates primarily through affective or cognitive channels. Future extensions of the model should therefore incorporate food quality, staff interaction quality, and discrete emotions as additional constructs, ideally in a moderated-mediation framework that tests whether the servicescape effects identified here are conditional on these omitted variables.

5.6. Theoretical Implications

The study makes several contributions to the body of knowledge on the relationships between physical environment, price perception, satisfaction, and loyalty in the restaurant industry. First, it extends the model proposed by Han and Ryu (2009) to an emerging Latin American market, demonstrating that the conceptual architecture environment → price → satisfaction → loyalty holds true in a socioeconomic and cultural context different from the Anglo-Saxon one, albeit with nuances that prevent simple replication. The confirmation of seven out of nine direct hypotheses and three out of four mediation hypotheses underscores the cross-cultural robustness of the model’s core relationships. Meanwhile, the divergences—particularly the significance of CA → SAT and the ineffectiveness of DE—indicate that the relative weightings of each environmental dimension are conditioned by contextual factors.

More specifically, the study advances the literature in at least three distinct ways that transcend geographic replication. First, by demonstrating that the environmental conditions–satisfaction path can be direct (not exclusively mediated by price perception), this study challenges the assumption of universal mediation embedded in the original Han and Ryu (2009) model and suggests that the Mehrabian–Russell emotional route and the cognitive–evaluative route can coexist in a complementary, non-competitive manner within the same structural model. This finding contributes to servicescape theory by identifying a boundary condition: in markets with high sensory gastronomy culture, the environment directly produces satisfaction through hedonic arousal rather than solely through price rationalization. Second, the detection and decomposition of serial mediation chains (environment → price → satisfaction → loyalty) constitutes an original empirical contribution that maps the complete S-O-R pathway from physical stimulus to behavioral response, a configuration that was neither hypothesized nor tested in prior servicescape research. Third, conceptualizing spatial distribution as a hygiene factor introduces a theoretically grounded contingency into servicescape models: not all environmental dimensions operate as motivators across all markets, and their relative importance is moderated by cultural consumption priorities—a proposition that future research should test explicitly through moderator analysis.

The finding that environmental conditions have a direct effect on satisfaction in the Peruvian context—something that did not occur in (Han & Ryu, 2009)—enriches the understanding of the servicescape by indicating that the environmental dimension is not universally subsidiary to the perception of price. In markets with a gastronomic culture marked by sensory impressions, the setting transcends its role as a cognitive indicator and becomes a direct source of satisfaction. This evidence dialogues with the postulates of (Mehrabian & Russell, 1974), for whom emotional responses to the environment—pleasure, activation—mediate approach behaviors, and extends them by suggesting that the emotional route and the cognitive route can coexist in a complementary, non-competitive way in the environment-satisfaction chain.

The advance—not post hoc—formulation of mediation hypotheses (H10 to H13) and their evaluation using specific indirect effects with bootstrapping represent a methodological improvement over the model-comparison procedure based on (Baron & Kenny, 1986) used by (Han & Ryu, 2009). Preacher and Hayes (2008) and Zhao et al. (2010) have argued that the causal steps approach underestimates mediation by requiring the direct effect to be significant as a precondition. By adopting the procedure currently recommended in the PLS-SEM literature (Hair et al., 2022), the present study offers more accurate estimates of indirect effects and allows for a more rigorous classification of mediation. The detection of significant serial chains (DA → PP → SAT → LOY and CA → PP → SAT → LOY) is an original contribution that illustrates the complete pathway through which the tangible dimensions of the restaurant are transformed into diner loyalty through chained cognitive and evaluative responses.

The interpretation of spatial distribution as a hygiene factor—necessary but not sufficient to generate satisfaction or signal value—opens up a line of theoretical reflection applicable to other emerging markets where layout variability may be limited or un, or where consumer priorities are concentrated on sensory and culinary dimensions. This nuance invites us to review theoretical models that assume equal weight for the three dimensions of the physical environment and to consider the possibility that the relative importance of each component is moderated by cultural and market variables that the literature has not yet examined systematically.

5.7. Practical Implications

The findings offer concrete guidance for full-service restaurant managers in northern Peru. The evidence indicates that investment in décor and environmental conditions yields returns on both price perception and satisfaction, and ultimately on diner loyalty. Owners and managers would do well to allocate resources primarily to these two components of the environment, rather than focusing on redistributing space, which, in this context, does not yield perceptible differentiation.

In terms of decoration, the results suggest that interior design should transcend mere functionality and articulate a visual narrative consistent with the regional gastronomic identity. Elements such as the incorporation of northern iconography, the use of indigenous materials in the furniture, and the display of local artwork not only beautify the space but also communicate to the diner—nonverbally—that the establishment offers a valuable experience that justifies the price, Jamaludin and Hashim (2024). Decor with narrative intent was found to intensify hedonic emotions, and Homeghi et al. (2025) showed that this same decor can be turned into shareable digital content, amplifying the brand’s reach.

With regard to environmental conditions, the finding of a direct effect on satisfaction provides an economic argument for sensory management in restaurants. Controlling lighting—warm during evening service, natural during lunch—selecting background music that reinforces the identity of the place without interfering with conversation, regulating temperature—a particularly relevant aspect in cities with extreme climates such as Piura—and managing smells—which in many northern Peruvian restaurants is almost automatic thanks to the aromatic intensity of the dishes, but which can be enhanced through deliberate design—represent relatively low-cost interventions with a proven impact on price perception and satisfaction. Apaza-Panca et al. (2023) offered operational guidelines in this direction for establishments in the region.

Price perception, for its part, should not be managed exclusively through menu pricing. Data show that the physical environment—decor and ambiance—raises the perception of price reasonableness; thus, a restaurant can justify a higher average check if the environment conveys quality and consistency. Pricing strategy thus becomes an extension of space design strategy, which (Li et al., 2025) called the “perceived price-quality ratio” as an attribute of the gastronomic destination. Managers of regional chains and franchises with a presence in Chiclayo, Trujillo, and Piura could standardize ambiance and decor protocols without sacrificing local uniqueness, thereby ensuring consistent price perception across all their locations.

Finally, the predominant role of satisfaction as a predictor of loyalty (β = 0.708, f² = 0.798) reminds managers that no effort in pricing or environment is sufficient if the overall experience does not reach a positive evaluative level. Satisfaction acts as the link that consolidates—or inhibits—the transformation of favorable perceptions into loyal behavior. Loyalty programs based exclusively on discounts or point accumulation risk capturing spurious loyalty if they are not accompanied by comprehensive experience management, as noted by (Andia-Reyna & Malasquez-Villanueva, 2025), who demonstrated that personalization is more effective than frequency programs.

For destination managers and regional tourism authorities, these results suggest that gastronomy destination competitiveness in northern Peru can be enhanced through coordinated servicescape standards. Specifically: (a) regional gastronomy clusters (e.g., the “Ruta Mochica” culinary corridor) should develop shared quality benchmarks for ambient conditions—lighting temperature between 2700 K and 3000 K for evening service, background music volume not exceeding 65 dB, and HVAC systems maintaining 22–24 °C regardless of external climate; (b) restaurant operators should allocate at least 15–20% of renovation budgets to décor elements that communicate regional identity (e.g., murals depicting northern iconography, Moche-inspired ceramics, local artisan furniture), as these decorative investments show the highest return on price perception; (c) menu pricing should be recalibrated in conjunction with environmental upgrades, since the data indicate that a well-designed setting allows diners to perceive up to 10–15% higher average checks as reasonable; and (d) loyalty programs should shift from discount-based mechanisms to experience-enrichment strategies (e.g., chef’s table events, seasonal tasting menus in curated settings) that leverage the satisfaction–loyalty pathway (β = 0.708) rather than competing on price.

5.8. Limitations and Future Lines of Research

The results of this study should be interpreted considering several limitations. The first concerns the sampling strategy. The sample of 310 diners was obtained using a non-probabilistic convenience sampling procedure combined with snowball sampling, which limits the generalization of the findings to the universe of full-service restaurant customers in northern Peru, as well as to other regions of the country or Latin America. Although the sample size exceeds the minimum thresholds recommended for PLS-SEM (Hair et al., 2022), future research with stratified probability sampling by city, gender, and income would increase the external validity of the estimates.

It should be noted that the sample profile exhibits a marked concentration in young adults (18–25 years, 53.9%) and respondents with university or postgraduate education (76.4%), while the Piura subsample represents only 12.6% of the total. This demographic skew, inherent to convenience and snowball sampling in university-linked urban settings, may overestimate the strength of the servicescape–satisfaction link among digitally literate, aesthetically sensitive consumers and underestimate it among older or less-educated segments whose evaluation criteria could weight functional attributes—portion size, speed of service—more heavily than atmospheric cues. Likewise, the underrepresentation of Piura limits confidence in the geographic generalizability of the coefficients obtained. Although the overall sample size (N = 310) comfortably exceeds the minimum thresholds for PLS-SEM, future research should pursue stratified quota sampling that balances age cohorts, educational levels, and cities in proportions closer to census distributions. If supplementary data collection is not feasible, multi-group analysis (MGA) by age bracket (e.g., 18–25 vs. 26+), education level (university vs. non-university), and city of data collection should be conducted to determine whether the structural paths remain invariant across subgroups or whether significant moderating effects emerge, thereby providing a more comprehensive assessment of the model’s robustness.

A cross-sectional design prevents the establishment of temporal causal relationships. The data were collected at a single point in time. Hence, the direction of the relationships is assumed based on theory, but it cannot be ruled out that satisfaction, for example, retrospectively influences the perception of the environment or price. Longitudinal studies that survey the same diners across visits would capture the dynamic evolution of loyalty and strengthen causal inferences.

The model was limited to the original constructs of (Han & Ryu, 2009) and did not incorporate variables identified in the more recent literature as relevant. Food quality, interpersonal service quality, consumer emotions (Chinelato et al., 2023; Vera-Falcón et al., 2025), memorable experience (Souki et al., 2023), and electronic word of mouth were outside the scope of the analysis. Future work could extend the model by integrating these variables to obtain a more complete representation of the experience-loyalty chain. Similarly, the inclusion of moderating variables—gender, income level, frequency of visits, or city of residence—would allow us to examine whether the effects found remain stable or vary among subgroups of diners. This recommendation coincides with that made by (Han & Ryu, 2009) regarding personal and situational factors.

The geographical concentration in three cities in northern Peru is both a strength—generating localized evidence for a little-explored context—and a limitation. Extending the study to cities in southern Peru (e.g., Arequipa, Cusco) or to the capital would allow us to contrast whether the patterns observed respond to dynamics specific to the northern macro-region or whether they can be generalized to the Peruvian gastronomic market as a whole. Comparative research between Latin American countries with different culinary traditions—Colombia, Mexico, Argentina—would provide evidence on the cultural moderation of the model’s relationships.

Finally, the ineffectiveness of spatial distribution in the Peruvian model calls for research designs that address this dimension with greater resolution. An experimental approach that manipulates the distribution of space—varying, for example, the distance between tables or the style of the layout—could elucidate whether the absence of effect is due to a lack of variability in the sample or to a genuine irrelevance of the construct in the evaluation of the northern Peruvian diner. Similarly, qualitative research exploring perceptions of space through in-depth interviews or focus groups would help to understand the subjective mechanisms that quantitative data fail to capture.

6. Conclusions

This research set out to examine how the components of the physical environment—decor and artifacts, spatial distribution, and environmental conditions—relate to price perception, satisfaction, and customer loyalty in full-service restaurants in northern Peru, a gastronomic context for which no empirical evidence addressed these relationships jointly. Using a PLS-SEM model tested with 310 diners from Chiclayo, Trujillo, and Piura, we used a bootstrap with 10,000 subsamples. The results allowed us to respond to the stated objective with a reasonable degree of accuracy: seven of nine direct hypotheses and three of four mediation hypotheses received empirical support, and the model explained more than 90% of the variance in the three endogenous variables.

The data show that not all dimensions of the physical environment operate with equal intensity in shaping diners’ perceptions and attitudes. Decor, artifacts, and environmental conditions are the two pillars that drive both the perception of reasonable pricing and experiential satisfaction, while spatial distribution is not significant in any of the relationships tested. This pattern differs partially from that reported by (Han & Ryu, 2009) in the US context and suggests that the relative weight of each component of the servicescape is conditioned by local cultural and gastronomic factors. In northern Peru, where the culinary experience is structured around intense sensory stimuli—regional aromas, sound ambiance, temperature control—and décor that projects identity, the distribution of space seems to function as a basic condition that diners take for granted but which, on its own, does not generate differentiation.

Price perception is confirmed as a dual-purpose cognitive mechanism. On the one hand, it operates as a direct antecedent of satisfaction and loyalty, and on the other, as a partial mediating link between the physical environment and satisfaction. The evidence obtained allows us to conclude that price management in full-service restaurants is not limited to the price printed on the menu: a well-designed environment raises the perception of the price’s reasonableness, which reinforces satisfaction and, in turn, intentions to return and recommend the restaurant. Decoration and environmental conditions channel part of their influence on satisfaction through this mechanism, forming complementary partial mediations that reveal a causal chain richer than the simple direct relationship between environment and satisfaction.

The most striking finding of the study lies in the magnitude of the relationship between satisfaction and loyalty (β = 0.708, f² = 0.798). No other factor in the model comes close to this level of influence. Satisfaction acts as the central evaluative filter that translates—or inhibits—the conversion of favorable perceptions into behaviors of commitment to the restaurant. Considering this result, the loyalty of the northern Peruvian diner is not built on an isolated attribute of the service, but on a comprehensive assessment of the experience that brings together the sensory, the aesthetic, and the economic. The serial chains identified (decor → price perception → satisfaction → loyalty and environmental conditions → price perception → satisfaction → loyalty) clearly illustrate the complete itinerary through which the tangible dimensions of the restaurant are transformed, through successive cognitive and evaluative responses, into a lasting bond with the establishment.

The study offers at least four contributions that transcend the mere replication of the original model. First, it extends the theoretical architecture of the servicescape (Han & Ryu, 2009) to an emerging Latin American market and demonstrates that its central relationships hold true—with nuances—outside the Anglo-Saxon context. Second, it identifies that environmental conditions have a direct effect on satisfaction in the Peruvian context, something that the seminal study failed to corroborate. This divergence enriches the understanding of the servicescape by showing that the environmental dimension can operate simultaneously as a cognitive price signal and as a direct source of experiential satisfaction. Third, the detection of significant serial chains linking the environment to loyalty through price perception and satisfaction is an original empirical contribution that was not evaluated in the work of Han and Ryu. Fourth, the advance formulation of mediation hypotheses and their evaluation using specific indirect effects with bootstrapping—rather than the Baron and Kenny (1986) procedure—represents a methodological improvement in line with current recommendations in the PLS-SEM field (Hair et al., 2022).

In summary, full-service restaurants in northern Peru that strategically invest in décor with regional identity and deliberate management of environmental conditions—lighting, music, aroma, temperature—while ensuring that the price charged is perceived as proportionate to the experience offered, will be better positioned to cultivate genuine loyalty based on satisfaction rather than inertia. The results invite further research in other Latin American food markets, incorporating emotional and technological variables that could enrich our understanding of a phenomenon whose complexity, as the data show, is difficult to capture in a single model.