How Review Valence Shapes Visit Intention: Affective Commitment and Destination Reputation

: In the era of social media, online reviews have become a crucial factor influencing the exposure of tourist destinations and the decision-making of potential tourists, exerting a profound impact on the sustainable development of these destinations. However, the influence of review valence on visit intention, especially the role of affective commitment and reputation (ability vs. responsibility), remains unclear. Drawing on emotion as a social information theory, this paper aims to elucidate the direct impact of different review valences on tourists’ visit intentions, as well as mediating mechanisms and boundary conditions. Three experiments indicate that positive (vs. negative) reviews can activate stronger affective commitment and visit intention, with affective commitment also playing a mediating role. Additionally, destination reputation significantly moderates the after-effects of review valences. More specifically, a responsibility reputation (compared with an ability reputation) weakens the effect of negative valence on affective commitment and visit intention. This study provides valuable theoretical insights into how emotional elements in online reviews influence the emotions and attitudes of potential tourists. Particularly for tourism managers, review valence and responsibility reputation hold practical significance in


Introduction
People love traveling but are often hesitant when deciding on a destination due to the uncertainty and risks involved in travel [1-4].Nowadays, the emergence of online reviews has become a crucial source of information for potential tourists when making their final decisions [5-7].According to Fang [8], 65% of tourists search the internet before choosing a destination, and 69% of their decisions are influenced by online reviews.Thus, the impact of online reviews on potential tourists' attitudes is evident in reality.Online reviews have become a significant underlying factor for the sustainable development of tourist destinations.However, the theoretical mechanism underlying this effect has not been clearly elucidated [9-12].In particular, although review valence is a prominent feature of online reviews, its impact on the visit intentions of potential tourists has not received sufficient attention from researchers.
In marketing research, review valence has been found to have a significant impact on perceived usefulness and behaviorial intentions [13,14], however, the research results are inconsistent as to whether positive or negative reviews have a greater impact on consumer responses [15].According to negativity bias [16,17], many studies indicated that negative reviews are more useful than positive reviews and have a greater impact on consumers' purchase decisions [3,13,14].Nonetheless, some other studies have found that regarding hedonic consumption (e.g., tourism, music), consumers perceive negative reviews to be less useful than positive reviews [18-20].Whether the negativity bias exists in tourism products has not been thoroughly tested, especially in the context of online reviews [3].It is worth exploring how potential tourists will respond to online reviews of different valences.
In previous studies [3,13,14,[21][22][23], most researchers focused on the impact of review valence on the audience from a cognitive perspective.For example, they examined the effect of review valence on perceived usefulness [24] and perceived helpfulness [25].This kind of research perspective, however, ignores that review valence is essentially an emotional expression of the reviewer's consumption.Compared with cognitive impact, review valence is more likely to have an emotional impact on its audience first [26].Especially in a hedonic tourism context, potential tourists' emotions often play a major role in destination selection and the decision-making process due to tourism products' intangible and experiential nature [6,7].Therefore, observing the impact of review valence on potential tourists from an affective perspective may be a valuable way to understand the function of review valence.
In addition, previous studies have found that when potential tourists have doubts about a destination they want to visit, destination reputation can be an important factor in their final decision [27][28][29].Most existing studies divided destination reputation into good reputation and average reputation, and indicated that a good destination reputation could help to reduce perceived risk and levels of uncertainty, but an average reputation will signal a weak destination image [2,30-32].However, this kind of research regards reputation as homogenous and fails to recognize the inherent heterogeneity of reputation.In reality, destination reputation contains two significantly different components.One is related to providing high-quality tourism products and services, and the other is related to implementing social responsibility [33].To the best of our knowledge, how these two different reputations influence the effect of review valence on visit intention has not yet attracted the attention of researchers, and thus, it needs further exploration.
To fill these gaps, the current paper proposed and tested a conceptual model which investigates how review valence impacts visit intention through the lens of emotion as a social information theory.It also demonstrates the mediating role of affective commitment and the moderating role of destination reputation.The contributions of this paper can be summarized into three points.First, in this paper, review valence is regarded as a kind of social information in which reviewers express their emotions [34].Positive valence imparts positive energy to such social information and pushes potential tourists to improve their willingness to travel, which proves that negativity bias is not applicable in a hedonic tourism context.It also puts forward a theoretical point regarding the function of online travel reviews.Second, this paper focuses on the emotional impact of review valence on potential tourists, rather than just looking at the effect of review valence from a cognitive perspective, so as to ascertain whether affective commitment is an emotional path that can link review valence and visit intention; this provides a theoretical construction of the relationship between them.Third, this paper explains the moderating effect of destination reputation on the after-effects of review valence, which clarifies the different effects of ability reputation and responsibility reputation on the process of review valence.Moreover, it acts on visit intention and enriches the theoretical connection between review valence and visit intention.

Review Valence and Visit Intention
Online reviews are created by Internet users based on their personal experiences [7,35].An online review is a kind of self-expression of emotion in the age of social media.
It may contain positive emotions or negative emotions; in other words, online reviews vary in valence.By definition, valence concerns the positive or negative information about an object or situation [36].Review valence is a significant indicator for measuring reviewers' satisfaction with their review objects [37].Mudambi and Schuff [35] argue that review valence refers to the positive or negative reviews given by the consumer about a product or service.Review valence can be positive, negative, or neutral, but the influence of neutral reviews on potential consumers is significantly lower than that of positive and negative reviews [35,38], so this paper focuses on positive and negative reviews.
Previous studies put forward the view of "negativity bias", which is used to show that people value negative reviews when making decisions based on online reviews [13,14].However, in a tourism context, when a potential tourist searches for online reviews of a destination, they may have aspirations and expectations for that destination [39].In this case, although negative reviews may cause potential tourists to worry about risks and uncertainties, they may prefer positive reviews to confirm their desires and expectations of the destination, as anticipation is evocative and tourists generally enjoy vacations as hedonic experiences [40,41].This tendency for self-confirmation will make potential tourists actively look for evidence to affirm their initial attitude towards the tourism destination and make them look forward to positive information [42,43].Wei, Miao, and Huang [23] found that positive reviews that are consistent with consumers' predictive preferences can verify their interest during the pre-decision-making process when selecting a particular hotel, which facilities the purchase decision-making process afterwards.Park and Nicolau [3] indicated that travelers who seek and read an online review will find positive reviews more attractive than negative ones.Using emotion as a social information theory [34], positive reviews may tend to generate more satisfying social information for potential tourists; this kind of social information may satisfy their expectations of a certain destination, and thus, it enhances their visit intention.As mentioned above, there is support for our proposition of Hypothesis 1: H1.Positive (vs.negative) reviews activate stronger visit intention.

Review Valence and Affective Commitment
Much of the research on affective commitment appears in the organizational research literature, and it often goes hand in hand with the concept of continuance commitment [44].In an organizational context, affective commitment is defined as emotional attachment to, identification with, and involvement in an organization [45].Another opposing concept, continuance commitment, is defined as the willingness to stay in an organization [46].In recent years, affective commitment has been increasingly used in brand research, and it is defined as an emotional response that originates from the psychological identification and attachment of customers to a brand [47].In the context of tourism, the destination is often considered a brand [29], so we define affective commitment as an emotional attachment to, involvement in, and identification with a destination in this paper.
Previous studies have often looked at the impact of review valence on potential consumers from a cognitive perspective; for example, most studies have suggested that review valence can directly affect the perceived usefulness of online reviews and the consumer's purchasing decisions [3,13,14,21-23].In the context of tourism, Wei, Miao, and Huang [23] also investigated the potential tourists' perceived usefulness of usergenerated reviews and found that the heuristic clues in online reviews would enhance the usefulness of reviews.However, with continued research, a few studies have begun to note the importance of examining the impact of review valence on potential tourists from an emotional perspective [26].For example, Pan, Shu, Kitterlin-Lynch et al. [48]  indicated that positive information in online reviews has a positive impact on potential tourists' affective attitudes and emotional behaviors.Therefore, review valence may have an influence on potential tourists' affective commitment.
Travel review valence is essentially an emotional expression of the tourists' experiences.Using the theory of emotion as social information [34], this kind of emotional expression is also a kind of social information about a certain destination, and this social information will activate an emotional response from potential tourists [49].
Positive reviews may convey positive information about a destination to potential tourists, such as good service and travel experience.This information may reinforce potential tourists' expectations of certain destination and gradually form their affective commitment.Negative reviews, on the other hand, provide negative information about a destination.Negative information may conflict with potential tourists' expectations for certain destination, weaken their emotional tendency towards the destination, and thus reduce their affective commitment.As such, we propose the following hypothesis: H2.Positive (vs.negative) reviews activate stronger affective commitment.

The Mediating Role of Affective Commitment
When potential tourists become interested in a certain destination and find online reviews of the destination through the Internet, the social information about the destination contained in the review valence will have an effect on their expectations of the destination and activate their affective commitment toward the destination.This kind of affective commitment comprises the expectation and identification of potential tourists with the destinations they are interested in; in essence, it is a kind of positive emotion.According to broaden-and-build theory [50], when a person's positive emotions are activated, their momentary thought-action repertoire can be broadened, and their behavior may be guided to develop in a positive manner [49].Therefore, the activated affective commitment may improve their interest and expectations of this destination, and further influence their visit intention.
Moreover, using the theory of emotion as social information, emotion contains rich social information, which will further affect individual behavior after the information is accepted as valid by readers [49].As an emotional expression, travel review valence contains a wealth of information related to the tourist destination, and this kind of information can exert its social impact and influence the behavior of potential tourists when it is read by them.That is, after being read by potential tourists, review valance may stimulate potential tourists' affective commitment and further affect their willingness to travel.Accordingly, we propose that:

The Moderating Role of Destination Reputation
Reputation is an individual's evaluation of the persistence of attributes of an entity over a long period of time [51].In the domain of tourism, destination reputation has always been a hot research topic [29].Destination reputation is defined as the combination of opinions, views, knowledge, and attitudes of different stakeholders, including tourists, managers, and the community at tourism destinations [52].Existing work mainly examines destination reputation from this angle.For example, Su, Lian, and Huang [29] divided destination reputation into good reputation and average reputation.This kind of research regards reputation as homogenous and fails to recognize the inherent heterogeneity of reputation.
From the perspective of destination practice, we observe that destination reputation contains two significantly different components.One concerns the high quality of tourism activities, products, or services, whereas the other concerns social responsibility (i.e., does it help the residents of a destination or provide social donations).The research of Johnson, Mao, Lefebvre et al. [33] indicated that there are two types of brand reputation, as follows: ability reputation signals quality and innovation, and socially responsible reputation indicates that a brand is committed to philanthropic activities and giving back to the community.This theory of reputation is consistent with our realistic observations of destination reputation.Thus, drawing upon this research, this paper divides destination reputation into ability reputation and social responsibility reputation (hereafter referred to as responsibility reputation), in accordance with its inherent heterogeneity.
Destination reputation is a tourist's evaluation of the existence of attributes of a destination over a long period of time.Existing research has shown that destination reputation is essentially a signal of social recognition and an expression of collective emotion, which would affect tourists' emotional processes [53,54]; however, different types of reputation may have different impacts [33].When a destination has an ability reputation, it can convey positive information concerning tourism products or services to potential tourists, which will strengthen the affective commitment of potential tourists.Such a reputation may, however, confuse tourists who read negative reviews.Due to the contrast between tourists' long-term impressions of a destination and abrupt comments about it, doubt may occur.This doubt may further reduce potential tourists' affective commitment to this destination.As such, there will be a significant difference between the affective commitment of positive and negative reviews when the destination has an ability reputation.
In contrast, when the destination has a responsibility reputation, it can convey positive information concerning a destination's social responsibility to potential tourists.The perception of an organization's social responsibility has been shown to significantly increase the organization's sense of warmth [55], potentially creating a halo effect for the organization [56].When a potential tourist is affected by the halo effect, due to the responsibility reputation of a destination, their emotion towards the destination may be dominated by this halo effect.Therefore, we infer that when potential tourists realize that a destination has a favorable responsibility reputation, their affective commitment to the destination will be greatly increased, and it is not affected by the review valence.Hence, there may be no significant difference to affective commitment caused by positive and negative reviews.Thus, we propose the following hypotheses: H4.Destination reputation moderates the effect of review valence on affective commitment.H4a.When the destination has an ability reputation, positive (vs.negative) reviews have a stronger effect on affective commitment.H4b.When the destination has a responsibility reputation, there is no significant difference in terms of the effect of review valence on affective commitment.
The discussion above shows that review valence is a short-term individual emotion expression for destination experience, whereas destination reputation is a long-term collective emotion expression for destination experience.Using emotion as a social information theory [34], review valence and destination reputation both provide social information about destinations.The interactions between review valence and destination reputation inevitably have different impacts on potential tourists' attitudes compared with a single piece of information.More specifically, an ability reputation will strengthen the influence of positive reviews on visit intention, as both reflect positive travel experiences and can corroborate each other, whereas an ability reputation may strengthen the negative influence of negative reviews on visit intention, as there is a conflict of information between the two.As such, there will be a significant difference between visit intentions, regardless of positive and negative reviews, when the destination has an ability reputation.
On the other hand, the responsibility reputation of a destination may make potential tourists feel that the destination is friendly and benevolent [57], since they may think that implementing socially responsible activities means that the destination is prosocial and focused on others [33].This view indicates that potential tourists will form a highly trustworthy impression of the destination, greatly reducing their risk perception of the destination, and improving their travel intentions substantially.This trustworthy impression may also dominate potential tourists' attitudes towards the destination, leading them to ignore the negative aspects of online reviews.Ergo, there may be no significant difference between visit intentions caused by positive and negative reviews.Therefore, the paper proposes that:

H5. Destination reputation moderates the effect of review valence on visit intention.
H5a.When the destination has an ability reputation, positive (vs.negative) reviews have a stronger effect on visit intention.H5b.When the destination has a responsibility reputation, there is no significant difference in the effect of review valence on visit intention.
Based on the above five hypotheses, this paper proposes a conceptual model, as follows (Figure 1).To test these hypotheses, we conducted three experimental studies.Study 1 first determined the stimuli materials for online review valence (positive vs. negative) through a pretest, then, it examined the effects of different online reviews on visit intention and tested the mediating role of affective commitment.Study 2 initially determined the stimuli materials for destination reputation (ability vs. responsibility) through a pretest, then, it verified the moderating effect of destination reputation by simultaneously manipulating review valence and destination reputation.Study 3, building on Study 2, further validated the moderating effect of destination reputation using real tourists, thus enhancing the external validity of the research.

Pretest
A pretest was designed to prepare the stimuli of the review valence for the main experiment.It recruited 98 Chinese undergraduate students (Mage = 20.76,46.9% female) as participants, and they were randomly and equally assigned into two groups (review valence: positive vs. negative).Participants in different groups were asked to read different versions of online reviews (positive and negative).Following the approach of Doh and Hwang [58], and Sparks and Browning [59], positive review stimuli contained 5 positive reviews, 1 negative review, and 1 neutral review (in random order), whereas the negative review stimuli included 5 negative reviews, 1 positive review, and 1 neutral review (also in random order).Review length, review time, subjective and objective expressions, and reviewer usernames were controlled to exclude the influence of other factors on the results.In order to avoid influences from existing tourism destination brands, the pretest used a virtual tourism destination named A [29].After reading the stimulus material, participants were asked to judge the review valence by answering the 7-point question of "I think the above online reviews of tourism destination A are positive" [20], and they were asked to provide demographic information.
Everyone successfully completed this pretest.Then, we checked the manipulation effectiveness of review valence by using an independent sample t-test.The group exposed to the stimuli of positive reviews reported significantly higher scores than the group exposed to negative reviews (Mpositive = 6.24,Mnegative = 2.35; t = 25.69,p < 0.001).This result suggested that participants could distinguish between positive reviews and negative reviews based on the given stimuli; the manipulation of the review valence was successful.

Research Design and Procedure
A one-factor between-subjects design (review valence: positive vs. negative) was conducted with 180 participants, who were recruited from an online data platform called Credamo (https://www.credamo.com,accessed on 1 September 2023); they were paid in cash for completing this experiment.Participants were randomly and equally assigned to one of the two different scenarios.Before starting the experiment, each participant was informed that they were guaranteed confidentiality in their answers and could withdraw from the experiment at any time.Participants were instructed to read the following statement: "Imagine that you are planning a trip during the vacation and want to go to tourism destination A. You want to read online reviews from other travelers to help you make your final decision."Next, participants in different groups read two different versions of online reviews of destination A and completed a questionnaire measuring manipulation checks, affective commitment, visit intention, and scenario authenticity.
The method to conduct a manipulation check was the same as the pretest.The measurement scale of affective commitment was adapted from Iglesias, Markovic, and Rialp [47].It included the following three 7-point items (α = 0.72): I enjoy being a tourist at this destination; I have positive feelings about this destination; I feel attached to this destination.The visit intention was measured with the following five 7-point items (α = 0.77): I can imagine spending my holiday in this destination; I already thought about spending my holiday in this destination; I intend to visit this destination in the near future; I would seek more information about this destination; I would prefer to visit this destination rather than others.These were based on the studies by Matzler, Strobl, Stokburger-Sauer et al. [60], and Seo, Choi, and Shin [61].Scenario authenticity was measured by asking participants to indicate Yes or No as to whether the provided scenario "could happen in real life" [62].
Three participants did not complete the experiment, and we finally received 177 data sets (49.2% female) with a response rate of 98.3%.Among them, 88 were in the positive review valence group, and the other 89 were in the negative review valence group.The demographic information of these participants is displayed in Table 1.The scenario authenticity test showed that more than 95% (95.5%) of the participants believed that the scenarios were realistic for them.Manipulation checks showed that participants could correctly distinguish between the review valences in the stimuli (Mpositive = 6.06 vs. Mnegative = 2.56, t = 25.45,p < 0.001), which indicated that the manipulation of the review valence was successful.
Independent sample t-tests were used to test H1 and H2.Results indicated that there was a significant difference between visit intention and affective commitment in the two groups.More specifically, the positive group's visit intention was significantly higher than that of the negative group (Mpositive = 5.53 vs. Mnegative = 4.56, t = 11.24,p < 0.001), and the positive group's affective commitment was significantly higher than that of the negative group (Mpositive = 5.78 vs. Mnegative = 4.37, t = 15.71,p < 0.001), the results therefore supported H1 and H2 (see Figure 2).The mediating role of affective commitment was tested by using Model 4 of PROCESS Macro by Hayes [63].We used review valence as the independent variable, affective commitment as the mediating variable, and visit intention as the dependent variable.Subsequently, a mediation analysis was conducted through bootstrapping tests with 5000 replications and 95% confidence intervals.The results indicated that affective commitment served as a mediator between review valence and visit intention (95% CI: 0.271, 0.740).Thus, H3 was supported, and Table 2 provides detailed information about mediation.

Pretest
This pretest aimed to check the manipulation effectiveness of destination reputation.Eighty-seven MBA students (Mage = 31.78,47.1% female) from a Chinese university voluntarily participated in the pretest.They were randomly assigned to one of the two scenarios (destination reputation: ability vs. responsibility).Two versions of destination reputation were created, which focused on either the destination's ability or social responsibility.More specifically, regarding the ability condition, destination A was described as being well-known for providing high-quality tourism activities and advanced experience services.Participants also read that destination A recently invested 350,000 CNY in new amusement facilities to enhance travel experience.Regarding the social responsibility condition, participants were informed that destination A was wellknown for its tendency to give back to society, and had recently donated 350,000 CNY to help residents around destination A fight the COVID-19 pandemic.After reading the stimuli, all the participants were asked to rate the destination reputation using the following 7-point question: What do you think of the reputation of destination A (Johnson, Mao, Lefebvre et al., 2019 [33]), where 1 represents a reputation for ability and 7 represents a reputation for social responsibility.Finally, they completed items concerning demographic information.
The results showed that of the two groups, destination reputation showed a significant difference (Mability = 2.49, Mresponsibility = 6.23, t = 24.54,p < 0.001).That is, the group under the ability destination reputation condition exhibited a significantly lower score than the group under the responsibility destination reputation condition.Thus, the manipulation of destination reputation was successful.

Research Design and Procedure
The main experiment utilized a 2 (destination reputation: ability vs. responsibility) × 2 (review valence: positive vs. negative), factorial, between-subjects design.Three-hundred participants recruited from an online data platform (Credamo) completed this experiment in exchange for cash payment.They were randomly assigned to four different scenarios, according to the factorial condition, with 75 people in each.The destination reputation manipulation was the same as the above pretest, and review valence manipulation was the same as in Study 1.
Participants were asked to imagine themselves in the scenario, and then they completed a questionnaire.In this questionnaire about scenario authenticity, review valence, affective commitment (α = 0.72), and visit intention (α = 0.85) were measured, which is the same as in Study 1, except for the addition of one item to check the manipulation effectiveness of destination reputation; this item was the same as the above pretest.
Ten participants did not complete the experiment, and we finally received 290 data sets (50.3% female) with a response rate of 96.7%.More specifically, ability reputation with positive reviews (n = 74), ability reputation with negative reviews (n = 72), responsibility reputation with positive reviews (n = 70), and responsibility reputation with negative reviews (n = 74) were measured.The demographic information of these participants is displayed in Table 3.The scenario authenticity test showed that more than 95% (95.2%) of the participants believed that the scenarios were realistic for them, and that they could easily imagine the scenarios existing in real life.The manipulation check showed that the participants could correctly differentiate between destination reputations (Mability = 2.30, Mresponsibility = 6.15; t = 37.75, p < 0.001), and there were significant differences between their judgments of review valences (Mpositive = 5.67, Mnegative = 2.96; t = 23.04,p < 0.001).Therefore, the manipulation of destination reputation and review valence were successful.
To test the moderating effect of destination reputation on the relationship between review valence and affective commitment, we conducted a 2 × 2 analysis of variance (ANOVA) by using review valence and destination reputation as independent variables and affective commitment as the dependent variable.Results showed a significant interaction for affective commitment (F(1, 289) = 5.05, p < 0.001, η 2 = 0.23), and the hypothesis of H4 was thus supported.Further analysis revealed that in the scenario with an ability reputation, positive (vs.negative) reviews had a stronger impact on affective commitment (Mpositive = 5.64 vs. Mnegative = 4.60, t = 10.20,p < 0.001), while under the responsibility reputation condition, no significant difference was found (Mpositive = 5.73 vs. Mnegative = 5.61, t = 1.34, p > 0.05) (see Figure 3).Therefore, H4a and H4b were confirmed.Furthermore, a 2 × 2 ANOVA with review valence and destination reputation as independent variables, and visit intention as the dependent variable, was utilized to test H5, H5a, and H5b.Consistent with the previous hypothesis, the results indicated a significant interaction between review valence and destination reputation on the visit intention (F(1, 289) = 9.09, p < 0.05, η 2 = 0.36); H5 was thus confirmed.Subsequent analysis illustrated that when a destination had an ability reputation, positive (vs.negative) reviews had a stronger impact on tourists' intentions to visit the destination (Mpositive = 5.75 vs. Mnegative = 4.34, t = 15.16,p < 0.001).However, when a destination had a responsibility reputation, there was no significant difference (Mpositive = 5.73 vs. Mnegative = 5.70, t = 0.378, p = 0.706) (see Figure 4); such results therefore supported H5a and H5b.

Study 3: Replication of the Results of Study 2 with Real Tourists
In order to overcome the limitations of using online experimental participants in Study 2, Study 3 replicated Study 2 with actual tourist samples.

Research Design and Procedure
In this study, an intercept survey approach was used to sample the tourists who visited Yuelu Mountain in Changsha, Hunan Province, China.Participation in the study was voluntary, and their answers were guaranteed to be kept confidential.Due to the tourism recovery stage, the number of tourists on Yuelu Mountain had not returned to normal levels yet, and we finally received 145 valid questionnaires.The demographic information of these participants is displayed in Table 4. Ability reputation with positive reviews (n = 37), ability reputation with negative reviews (n = 36), responsibility reputation with positive reviews (n = 35), and responsibility reputation with negative reviews (n = 37) were measured.We estimated statistical power based on our sample size by using G*Power 3.1.It was found that in the case of two-way ANOVA (effect size = 0.4, α = 0.05, Numerator df = 1, Number of groups = 4), the power value of 145 samples was 0.97, which exceeded the threshold of 0.80.This indicates that the sample size is sufficient for this study [64].
The stimuli and procedures of Study 3 were basically the same as those of Study 2. First, the participants were asked to read the stimuli with different versions of destination reputation and online reviews.Second, they were asked to complete the measure of manipulation effectiveness, affective commitment (α = 0.78), and visit intention (α = 0.89); these measurement scales were the same as that of Study 2. Finally, demographic information was collected.

Data Analysis and Results
First, we tested the scenario authenticity, the manipulative effectiveness of the destination reputation, and review valence.The results showed that the scenario authenticity test showed that more than 94% (94.5%) of the participants believed that the scenarios were realistic for them, the participants could correctly differentiate between destination reputations (Mability = 2.66, Mresponsibility = 6.15; t = 26.65,p < 0.001), and there were significant differences between their judgments of review valence (Mpositive = 5.67, Mnegative = 2.96; t = 16.09,p < 0.001).Therefore, the manipulation of destination reputation and review valence were successful again.
Then, a 2 × 2 ANOVA showed that the interaction between review valence and destination reputation had a significant influence on affective commitment (F(1,144) = 2.785, p < 0.05, η 2 = 0.315).The results of a subsequent analysis of an independent sample t-test demonstrated that in a scenario with an ability destination reputation, positive (vs.negative) reviews had a stronger impact on affective commitment (Mpositive = 5.53 vs. Mnegative = 4.39; t = 8.03, p < 0.001).However, under the responsibility destination reputation condition, no significant difference was found between the impacts of review valence on affective commitment (Mpositive = 5.57 vs. Mnegative = 5.55; t = 0.13, p = 0.896).Hence, destination reputation plays a moderating role in the relationship between review valence and affective commitment again, which suggests that our hypotheses, H4, H4a, and H4b, can also be supported among real tourists.
Additionally, data from real tourists also showed that the interactions between review valence and destination reputation had a significant impact on visit intention (F(1,144) = 3.432, p < 0.01, η 2 = 0.358).Furthermore, the independent sample t-test indicated that when a destination had an ability reputation, positive (vs.negative) reviews had a stronger impact on tourists' intentions to visit the destination (Mpositive = 5.65 vs. Mnegative = 4.30, t = 9.51, p < 0.001).However, when a destination had a responsibility reputation, there was no significant difference regarding visit intention (Mpositive = 5.61 vs. Mnegative = 5.59, t = 1.04, p = 0.918).As such, the hypotheses of H5, H5a and H5b, as confirmed in Study 2, were repeated in this new sample.

Conclusion
Drawing on the theory of emotion as social information [34], the current paper proposed and tested a conceptual model to investigate how review valence influences affective commitment and visit intention.Five hypotheses were proposed, and three studies were designed to test them.Study 1 supported the notion that, compared with negative reviews, positive reviews could activate stronger visit intention and affective commitment.Moreover, Study 1 illustrated that affective commitment was a mediator between review valence and visit intention.Study 2 revealed that destination reputation significantly moderated the after-effects of review valence, and that responsibility (vs.ability) reputation would weaken the effect of negative valence on affective commitment and visit intention.Study 3 replicated the conclusions of Study 2 by using a sample of real tourists, which improves the generalization of the conclusions in this paper.
This study provides valuable theoretical insights into how emotional elements in online reviews influence potential tourists' emotions and attitudes, especially regarding tourism managers, review valence, and responsibility reputation, which are of practical significance in the sustainable development of tourist destinations.

Theoretical Contributions
This study provides insight into online review preference for tourism products.The previous marketing literature has explored the impact of review valence on the consumer purchasing decision-making process.The majority of these studies suggest that negative reviews are often more influential than positive reviews, and this effect is known as negativity bias [3,13,14,21-23].However, using the theory of emotion as social information, this paper suggests that review valence is a kind of social information in which reviewers express their emotions, and it confirms that positive valence enhances the magnitude of such social information and pushes potential tourists to improve their willingness to travel.In this way, the current paper proves that negativity bias is not generally applicable to all tourism contexts, and it also puts forward a theoretical point for the functioning of online travel reviews.
Second, according to the emotion as social information theory [34], there are two ways that social information from emotions influence others, as follows: inferential processes and affective paths [65].Previous studies have illuminated the inferential process of the effect of review valence on others through usefulness [24] or helpfulness [25], but they have ignored the affective path of the effect of the review valence on others.This paper focused on the emotional impact of review valence on potential tourists, rather than just looking at the effect of review valence from a cognitive perspective [24,25], so as to understand affective commitment as a kind of emotional path linking review valence and visit intention.This resulted in a theoretical construction of the relationship between them and expanded the explanatory power of emotion as social information theory in tourism research.
Third, this paper introduced a research perspective to study destination reputation which realizes that destination reputation is not homogeneous.Hence, due to its inherent heterogeneity, combined with the reality of the destination reputation and the related literature [33,53,54], destination reputation is divided into ability reputation and responsibility reputation.This classification is quite different from the existing research, which divides it into good and average, according to the degree of destination reputation; this is a valuable innovation and furthers research on the destination reputation.Furthermore, this paper clarifies the different effects of ability reputation and responsibility reputation on the process of review valence acting on visit intention, which enriches the theoretical connotations between review valence and visit intention through the new classification of destination reputation.

Managerial Implications
The sustainability of tourist destinations hinges upon various factors [66], with visit intention serving as a critical gauge of their sustainable development trajectory.Findings from this investigation underscore the pivotal role of positive reviews in substantially augmenting the visitation intention of potential tourists.Consequently, this underscores the theoretical significance of positive reviews as a fundamental determinant in fostering the sustainable development of tourist destinations.The results of the present study offer significant managerial insights that can guide the strategic utilization of online reviews to foster sustainable development in the tourism industry.The study demonstrates that the influence of positive reviews on affective commitment and visit intention outweighs that of negative reviews.Therefore, marketers operating in hedonic tourism destinations should allocate resources to amplify positive reviews rather than focusing on addressing negative feedback.Additionally, stakeholders involved in tourism destination marketing and online travel platforms should prioritize the dissemination of positive reviews to prospective tourists.This can be achieved through collaborations with reputable travel websites to enhance the visibility of positive reviews for tourism destinations.For example, collaborations with tourism social platforms can be established to initiate tourism experience sharing campaigns, with incentives offered to encourage travelers to spread positive tourism-related reviews.This allows positive tourism-related reviews to be recognized by a larger audience of potential travelers.By strategically emphasizing positive reviews, marketers can effectively shape tourists' perceptions and enhance destination appeal, thus contributing to the sustainable growth of the tourism industry.
As indicated by the findings, the moderating effect of destination reputation on the impact of review valence underscores the nuanced nature of reputation dynamics.More specifically, the study suggests that while ability reputation can enhance affective commitment and visit intention when coupled with positive reviews, its efficacy may be limited when mitigating the negative effects of unfavorable reviews.However, responsibility reputation emerges as a compensatory mechanism, capable of attenuating the adverse consequences associated with negative reviews.This finding underscores the importance of cultivating a multifaceted reputation strategy, wherein both ability and responsibility dimensions are carefully cultivated.In the contemporary era, which is marked by the pervasive influence of social media, tourists exhibit heightened discernment and are increasingly inclined to voice their grievances publicly.In this context, a robust responsibility reputation can serve as a shield against the deleterious impact of negative reviews, thus mitigating their potential harm to the destination's image.Consequently, while emphasizing the importance of ability reputation, tourism destinations should prioritize the cultivation of responsibility reputation.In doing so, they can establish a resilient reputation framework that not only fosters positive perceptions among tourists, but also safeguards against reputational risks arising from negative feedback in the digital realm.For example, when promoting a tourist destination, it should incorporate the destination's social responsibility initiatives and charitable activities as part of its strengths.Additionally, in the management and operation of the tourist destination, social responsibility activities should be integrated into daily operations, with developed corresponding strategies and investment plans.

Research Limitations and Future Research Directions
There are also some research limitations in this paper, which can provide some enlightening clues for future research.This paper investigates the influence of review valence (positive or negative) on potential tourists' responses.Although there is a negative (positive) review in the stimuli material of the positive (negative) valence, the numbers between the two are completely unequal, whereas in reality, potential tourists may read an equal number of positive and negative reviews about a certain destination on a social platform.Ergo, future research could consider how potential tourists respond when faced with equal numbers of positive and negative reviews by discussing the framing effect of review valence.In addition to valence, other online review features are also worth studying, such as review length [22], number of reviews [67], review platform types [5], or review readability [68].Finally, although this paper uses a new perspective to divide destination reputation into ability reputation and responsibility reputation, it only discusses its moderating role in terms of the after-effects of review valence.Future studies are encouraged to explore the different effects of ability reputation and responsibility reputation on variables such as revisiting intention, sharing intention, and tourist wellbeing.

Figure 2 .
Figure 2. The impact of review valence on visit intention, affective commitment, and resident quality of life.

Figure 3 .
Figure 3. Moderating effect of destination reputation on the relationship between review valence and affective commitment.

Figure 4 .
Figure 4. Moderating effect of destination reputation on the relationship between review valence and visit intention.

Table 1 .
Demographic information of Study 1.

Table 2 .
The mediation analysis results.

Table 3 .
Demographic information of Study 2.

Table 4 .
Demographic information of Study 3.