Receiving neighboring behavior reflects the amount of assistance obtained from neighbors when confronted with difficulties (3). Prior studies have highlighted the benefits of receiving neighboring behavior on mental health. For instance, Zu et al. suggested that assistance from neighbors will relieve work-family conflicts, thereby enhancing mental health (22). Giving neighboring behavior is a process in which residents spend time helping others to resolve problems in the community (4). Recently, the giving of neighboring behavior has been highlighted as an effective tool for cultivating community resources. Zhang et al. explored the relationship between the giving of neighboring behavior and social functioning (5). Research suggests that helping neighbors rewards the helpers with cognitive information processing abilities and enhances emotional regulation skills (5), both of which are essential for maintaining mental health.

Integrating the above statements, the authors assume that when receiving and giving neighboring behavior are highly congruent, residents both receive support from neighbors and obtain resources by helping their neighbors. Under this context, residents have sufficient resources, such as social capital and positive emotions, to enable them to maintain their mental health. Because of a low level of receiving neighboring behavior, such residents are less likely to receive social and psychological resources from the community when confronted with difficulties. They have no extra access to social support and psychological resources, which leads to an increased risk of mental illness. Due to the low level of giving neighboring behavior, working residents cannot cultivate good reputations and favorable social relationships. These working residents are usually unfamiliar with their neighbors and more likely to experience mental illness due to a lack of social interactions (3). As a result, when the congruence in receiving and giving neighboring behavior is at a low level, residents are usually not involved in the community and are isolated from their neighbors, which is not beneficial for them to maintain their mental health. Hence, the following is hypothesized:

As aforementioned, helping behavior from neighbors helps residents to release stress associated with their daily life and work. For example, Griggs et al. (1) highlighted the importance of community support for attenuating work-family conflict for low-income workers. Supporting this, Zu et al. (22) suggested that support from neighbors may help low-income residents to cope with work-family conflict, thereby enhancing their mental health and career satisfaction.

With regard to giving neighboring behavior, prior studies have highlighted both its advantageous and disadvantageous influences on psychological and behavioral responses. Giving neighboring behavior is beneficial because it provides helpers with enhanced social functioning, which is essential for maintaining mental health (5). Recently, several scholars have begun to focus on the negative aspects of helping behavior. Helping behavior entails individuals assigning some of their time to assist others in resolving and preventing difficulties (23). Time is a limited resource in any day (15). This means that helpers face a trade-off between completing their own tasks and helping others with their limited time. Thus, when residents consume time and other resources (e.g., cognitive and emotional resources) to help their neighbors, they are more likely to experience ego depletion (15), which is detrimental to their mental health (24).

In the aggregate, when receiving and giving neighboring behavior are imbalanced in either direction, residents will present different levels of mental health. In the condition of giving high levels of neighboring behavior while receiving low levels of neighboring behavior, residents both experience the advantages and disadvantages of helping their neighbors. In particular, residents suffer from the loss of time and other resources without benefitting from resource supplementation (i.e., receiving neighboring behavior). Under this condition, residents are more likely to suffer lower levels of mental health because of ego depletion. Under the condition of receiving high levels of neighboring behavior while giving low levels of neighboring behavior, residents receive support from their neighbors but extend little effort to help others, even though the cultivation of social support would aid them in maintaining their mental health. Hence, the following is hypothesized:

Zhang et al. suggested that the spillover processes through which community resources are delivered into the work and family domains are contingent upon personal traits (4). Moreover, Kinnunen et al. found that the association between work-family conflict and psychological well-being depends on personality (39). To better understand the boundary conditions under which receiving and giving neighboring behavior fit is beneficial for mental health, this study assumes labile self-esteem to be a moderator in the relationship between work-family interference and meaning in life, in light of its impacts on psychological well-being.

Labile self-esteem reflects an individual's tendency to experience fluctuations in their level of self-esteem (40). In contrast to self-esteem, people with a high level of labile self-esteem are more likely to shift the perception they hold of themselves, thereby exposing themselves to increased risks for depressive symptoms. Roberts and Monroe found that persons with high labile self-esteem exhibit a special sensitivity to stress (41). Moreover, Roberts and Kassel found that individuals with high self-esteem have a higher likelihood of experiencing life stress (42). Work-family interference is commonly regarded as a hindrance stressor, which consistently consumes working residents' job resources and results in decreased meaning in life (43, 44). Given the increased sensitivity to stress caused by a higher level of labile self-esteem, work-family interference exerts strong negative impacts on meaning in life for working residents with a high level of labile self-esteem. By contrast, working residents with a low level of labile self-esteem tend to cope with stress proactively and are less likely to be impacted by it. Thus, work-family interference is less detrimental to meaning in life for working residents with low levels of labile self-esteem. Hence, the following is hypothesized:

This research focuses on the impacts of receiving and giving neighboring behavior fit on mental health through work-family interference and meaning in life. Two criteria for samples were set: having full-time jobs and having lived in the current communities for over 1 year. Samples were chosen randomly from people living in urban communities in Harbin City, China. With the assistance of trained social workers, questionnaires were sent out in hard copies. To control for common method variance, this study adopted a lagged questionnaire survey and collected data at two time points. The first time point was on 1st November 2018. Neighboring behavior (i.e., receiving and giving neighboring behavior) and work-family interference (i.e., family interference with work and work interference with family) were surveyed at this time point. The second time point occurred on 16th December 2018, and both meaning in life and mental health were surveyed at this time point.

A total of 317 surveys were collected in the first wave, and 220 surveys in the second wave. The effective response rate was 69%. The results of drop-out analysis indicated that the dropped samples exhibited an insignificant difference from the retained samples in regard to neighboring behavior, work-family interference, meaning in life, and mental health. Respondents worked in diverse industries (e.g., governments, manufacturing, and internet companies). On average, they were 41.26 (± 9.60) years old and had 21.77 (± 11.26) years of work experience. 41.4% of the respondents were male, and 78.6% were married. Regarding their education level, 19.1% of the samples had senior school education and below, 38.6% had high school education, 27.3% had a college education, and 15.9% had a bachelor's degree.

All scales were originally developed in English and were translated into Chinese. To ensure their validity, the back-translation procedure suggested by Brislin was employed (45). A five-point Likert scale was used ranging from 1 = strongly disagree to 5 = strongly agree without special statements.

The distribution of demographic information is depicted in Table 1. Furthermore, we have calculated the means, standard deviations, and correlations between the focal variables. Theses results are shown in Table 2.

Polynomial regression with response surface analysis was adopted to test the hypotheses (53). This statistical method has been employed by researchers in the fields of psychology and management to explore how the combination of two independent variables is related to one dependent variable, particularly in the case of congruence and discrepancy measures (54). Dawson suggested that this approach is superior to traditional regression analysis because polynomial regression analysis can provide a three-dimensional view of the interactive influence of two predictors on one dependent variable (55).

The classical polynomial regression equation is Z = b₀ + b₁X + b₂Y + b₃X² + b₄XY + b₅Y² + e. In this equation, Z is the dependent variable (mental health), X represents giving neighboring behavior, and Y represents receiving neighboring behavior. In the response surface analysis, coefficients in the polynomial regression are used to examine the response surface pattern, which is depicted to provide a three-dimensional visual representation of the data for the interpretation of the polynomial regression results (56). The surface pattern was determined by the slope and curvature of the congruence line (X = Y) and the incongruence line (X = -Y). For Hypotheses 3 through 8, to assess the moderated mediation effect on the relationship between receiving and giving neighboring behavior fit and mental health, Edwards and Cable's (57) approach was followed first. A block variable that combined the five polynomial terms (X, Y, XY, X2, and Y2) was calculated based on their respective weights in the polynomial regression analysis (58). Then, path analysis was conducted to examine the moderated mediation model by using Mplus 7.4.

Following Edwards's (57) suggestions, X and Y were centered. To test Hypothesis 1, it was determined whether the slope along the congruence line (X = Y) was significantly positive. This would indicate that mental health increased as giving and receiving neighboring behavior matched at higher levels rather than lower levels. Regarding Hypothesis 2, it was determined whether the slope along the incongruence line (X = –Y) was significantly negative. Such a result would indicate that the working residents' mental health was better under the condition of lower giving neighboring behavior with higher receiving neighboring behavior rather than higher giving but lower receiving neighboring behavior.

Polynomial regression analysis was used to determine how congruence and incongruence in receiving and giving neighboring behavior affect mental health. This analysis was conducted in Mplus 7.4. The results of Model 2 (Table 3) indicate a negative association between giving neighboring behavior and mental health (β = −0.40, p < 0.01). However, the association between receiving neighboring behavior and mental health was significantly positive (β = 0.96, p < 0.01).

To test both H1 and H2, response surface analysis was used and the response surface pattern was examined based on the curvature and slopes of the congruence and incongruence lines. The results shown in Table 3 indicate that the slopes of the congruence line (x = y) (β = 0.56, SE = 0.08, p < 0.01) and the incongruence line (x = –y) (β = −1.35, SE = 0.38, p < 0.01) are significant. The positive slope of the congruence line indicates that mental health was higher when giving and receiving helping behavior were congruent at higher levels. This supports Hypothesis 1. The negative slope of the incongruence line indicates that mental health was higher in the condition when working residents received much but gave little neighboring behavior compared to those who gave much but received little neighboring behavior. This supports Hypothesis 2. To interpret the results holistically, coefficient estimates, standard errors, and covariances in the polynomial regression analysis were used. The overall response surface within the data range was plotted by adopting the method developed by Shanock et al. (59) (see Figure 2).

To assess the underlying spillover process that links community resources to mental health, this study used a block variable approach as specified by Edward and Cable (57) to test hypotheses H3–H6. The results are shown in Figure 3. Receiving and giving neighboring behavior fit was negatively correlated with both family interference with work (β = −0.36, p < 0.01) and work interference with family (β = −0.39, p < 0.01). This supports Hypothesis 3. Family interference with work (β = −0.19, p < 0.01) and work interference with family (β = −0.12, p < 0.01) were negatively associated with meaning in life. This supports Hypothesis 4. Meaning in life was positively associated with mental health (β = 0.77, p < 0.01), which supports Hypothesis 5.

To test hypothesis 6, a bootstrapping test was adopted, the results of which are shown in Table 4. The serial mediating effect was significant for both the family interference with work path (Effect = 0.11, SE = 0.03, 95%CI = [0.06, 0.18]) and the work interference with family path (Effect = 0.08, SE = 0.03, 95%CI = [0.03, 0.14]).

To test Hypotheses 7 and 8, two interactive items were added to the path analysis. The interactive item of labile self-esteem with family interference with work was significant (β = −0.19, p < 0.01). A simple slope test was further conducted to assess the moderating effect. The results indicate that the negative influence of family interference with work on meaning in life emerged for working residents with high labile self-esteem (Effect = −0.28, SE = 0.04, 95%CI = [−0.35, −0.21]). The moderating effect of labile self-esteem in the relationship between family interference with work and meaning in life is shown in Figure 4.

The interactive item of labile self-esteem with work interference with family was significant (β = −0.07, p < 0.05). A simple slope test was further conducted to assess the moderating effect. The results indicate that a negative influence of work interference with family on meaning in life affected working residents with high labile self-esteem (Effect = −0.17, SE = 0.04, 95%CI = [−0.25, −0.08]). The moderating effect of labile self-esteem in the relationship between work interference with family and meaning in life is shown in Figure 4. This supports Hypothesis 7.

Furthermore, the moderated mediation model was examined through bootstrapping tests. The results shown in Table 4 indicate that the indirect effect of giving and receiving neighboring behavior fit on mental health through the family interference with work path affected working residents with high labile self-esteem (Effect = 0.23, SE = 0.05, 95%CI = [0.13, 0.33]). Also, the indirect effect of giving and receiving neighboring behavior fit on mental health through the work interference with family path affected working residents with high labile self-esteem (Effect = 0.13, SE = 0.04, 95%CI = [0.06, 0.21]).

The current study has several limitations that point to the following directions for future research. First, this study could not establish firm causal relationships. Receiving and giving neighboring behavior were not manipulated. Therefore, the causal impact of receiving and giving neighboring behavior fit on mental health could not be inferred. Future research may adopt an experimental design and cross-lagged panel design to overcome this shortcoming.

Second, this study could not rule out common method variance [CMV; (66)]. Although a two-wave questionnaire survey design was adopted, CMV still led to potential bias in the results. The employed questionnaires were self-reported and more than one questionnaire was completed at the same time point (e.g., meaning in life and mental health). This may have introduced CMV bias into the results. Future research may adopt a multi-source survey design or objective data to avoid CMV. For example, giving neighboring behavior can be assessed by neighbors.

Third, this study was conducted within Chinese communities, which limits external validity. Collectivism culture is widespread in China (67), resulting in more widespread interactions between neighbors, especially among older working residents. Future research may replicate this study in different age groups and explore differences between Chinese and Western cultures.