Abstract
Government information services play a critical role in helping farmers respond to meteorological disasters, yet the underlying psychological mechanisms remain insufficiently understood. By integrating the Information-Cognition-Behavior (ICB) framework, this study examines how government information services influence adaptive behavior among large-scale grain producers, with risk perception and government trust serving as dual mediating mechanisms and social capital acting as a moderator. Using survey data from 414 large-scale grain producers in Jiangxi Province, China, and a simultaneous equations model estimated using three-stage least squares (3SLS), the results show that government information services significantly increase both risk perception and government trust, which in turn contribute to farmers’ adaptive behavior. The mediating effect of government trust appears to be somewhat stronger than that of risk perception. Social capital positively moderates the relationships between government information services and cognitive responses, and these effects are more evident among producers with higher education and income levels. These findings provide additional evidence on the mechanisms through which government information services influence farmers’ adaptation to meteorological disasters and offer practical implications for strengthening rural information systems, institutional trust, and agricultural resilience under climate change.
1 Introduction
Against the backdrop of accelerating global climate change, the frequency and intensity of extreme meteorological events have risen sharply, inflicting substantial losses on agricultural production and rural livelihoods (Wu et al., 2022). Climate variability has emerged as a major barrier to achieving food security, poverty reduction, and sustainable agricultural development, with rural regions—characterized by fragile economic bases, limited resources, and high dependence on rain-fed farming—being disproportionately vulnerable (Qi and Yuan, 2017; Muema, 2018). In 2022 alone, Jiangxi Province in China experienced 26 natural disaster events, affecting 11.1 million people and causing direct economic losses of 28.21 billion RMB (Jiangxi Provincial Emergency Management Department, 2023). Among agricultural producers, large-scale grain producers deserve particular attention because they play a central role in safeguarding national food security while simultaneously facing greater exposure to climate-related production risks due to their larger operational scale and higher market dependence (Luo et al., 2016; Yang et al., 2012). Compared with smallholder farmers, large-scale producers generally possess stronger production specialization, greater capital investment, and more frequent interactions with governmental institutions, making their adaptive decisions more sensitive to government information services and institutional support mechanisms.
Farmers’ adaptive behaviors—such as adjusting planting schedules, improving irrigation and drainage infrastructure, adopting stress-resistant varieties, and optimizing agricultural input use—are widely recognized as essential strategies for reducing climate vulnerability and strengthening agricultural resilience (Truelove et al., 2015; Madhuri and Sharma, 2020). For large-scale grain producers, whose production decisions involve higher capital investment, greater market exposure, and broader cultivated areas, the consequences of meteorological disasters are often amplified, making effective adaptation particularly important (Xi, 2022). Existing studies have shown that farmers’ adaptive behavior is influenced by multiple factors, including household characteristics, resource endowments, climate experience, risk perception, and access to information (Deressa et al., 2009; Zhu and Zhou, 2011; Tian and Chen, 2014; Huang et al., 2017; Burnham and Ma, 2017; Wei et al., 2020). However, most prior research has focused either on smallholder farmers or on the direct effects of information access and economic incentives, while paying relatively limited attention to the underlying cognitive and institutional mechanisms through which government information services shape adaptive behavior among large-scale agricultural producers (Zhao and Zhou, 2012).
Despite the recognized importance of government information services, the mechanisms through which such services influence farmers’ adaptive behavior remain insufficiently explained. Existing studies have primarily focused on the direct effects of information access, agricultural extension services, technology promotion, or economic incentives on adaptation behavior (Deressa et al., 2009; Chai and Zhao, 2013; Burnham and Ma, 2017; Wang et al., 2023), while relatively limited attention has been paid to the cognitive processes that translate external information into behavioral responses. In reality, farmers do not respond mechanically to government information. Previous studies have shown that farmers’ responses to climate information are shaped by subjective interpretations, cognitive evaluations, and perceptions of institutional credibility (Truelove et al., 2015; Arbuckle et al., 2015). Instead, they first interpret and evaluate the information they receive, forming perceptions about meteorological risks as well as judgments regarding the credibility and competence of governmental institutions. These cognitive evaluations may subsequently shape their willingness to adopt adaptive measures.
To better explain this process, this study integrates the Information-Cognition -Behavior (ICB) framework to examine how government information services influence adaptive behavior among large-scale grain producers. Rather than proposing a completely new standalone theory, the ICB framework synthesizes the “information-cognition-behavior” logic embedded in Protection Motivation Theory (Rogers, 1975) and the Theory of Planned Behavior (Ajzen, 1991), emphasizing that external informational stimuli influence behavioral responses through cognitive evaluations. Within this framework, government information services function as external informational inputs, while risk perception and government trust represent two parallel cognitive mechanisms linking information exposure to adaptive behavior. In the context of rural China, where government agencies play a dominant role in meteorological warning dissemination, agricultural extension, and disaster governance, farmers often rely heavily on governmental institutions when evaluating the credibility and usefulness of climate-related information. Consequently, government trust constitutes an especially important institutional cognition that may shape farmers’ willingness to translate information into adaptive action (A et al., 2022; Ma and Xia, 2003). Accordingly, this study examines how government information services indirectly influence meteorological disaster adaptive behavior among large-scale grain producers through the dual mediating roles of risk perception and government trust, while also investigating the moderating role of social capital. In addition, considering that farmers with different socioeconomic characteristics may process information and respond to climate risks differently, this study further explores heterogeneity across education and income groups.
This study makes three main contributions. First, it advances the literature by jointly examining risk perception and government trust as dual cognitive mediators linking government information services to adaptive behavior among large-scale grain producers in China. Second, the application of the 3SLS system estimation addresses endogeneity concerns that have been largely overlooked in prior single-equation studies of farmer adaptation. Third, by focusing on large-scale grain producers as a distinct and policy-relevant population, the study addresses an important empirical gap and generates findings with direct implications for agricultural disaster governance in China.
2 Literature review and theoretical framework
2.1 Farmers’ adaptive behavior toward meteorological disasters
Farmers’ adaptive behavior toward meteorological disasters refers to the deliberate actions undertaken by agricultural producers to reduce climate-related vulnerability and maintain agricultural productivity under conditions of increasing climatic uncertainty (Smit and Wandel, 2006). As climate change intensifies the frequency and severity of extreme weather events, understanding the determinants of farmers’ adaptive behavior has become an important research priority in the fields of climate adaptation and agricultural sustainability.
Existing studies have identified a wide range of determinants of farmers’ adaptive behavior, including household socioeconomic characteristics, resource endowments, access to credit, and information availability (Maddison, 2007; Deressa et al., 2009; Luo et al., 2016). More recent research highlights the importance of agricultural extension services and climate information provision in reducing uncertainty and enhancing farmers’ adaptive capacity (Aker, 2011; Hansen et al., 2011; Feng et al., 2016).
More recently, scholars have paid increasing attention to the role of government information services in promoting agricultural adaptation. Through meteorological forecasts, disaster warnings, agricultural extension programs, and policy dissemination, governments can reduce information asymmetry and improve farmers’ capacity to respond to climate risks (Wang et al., 2023). However, existing studies have largely focused on the direct effects of information provision, while paying relatively limited attention to the mechanisms through which government information services influence adaptive behavior.
Although previous studies have demonstrated the importance of information access, relatively little attention has been paid to the cognitive and institutional processes through which government information services influence adaptation decisions. In particular, risk perception and government trust may represent two important yet insufficiently explored pathways linking information services to adaptive behavior. Moreover, most empirical evidence is derived from smallholder farmers, while large-scale grain producers remain comparatively understudied despite their growing importance in agricultural modernization and food security. To address these gaps, this study adopts the Information-Cognition-Behavior (ICB) framework to examine how government information services influence adaptive behavior among large-scale grain producers through the dual mediating roles of risk perception and government trust.
2.2 Information-cognition-behavior framework
Understanding how external information translates into behavioral responses has long been a central concern in behavioral research. Protection Motivation Theory (PMT) argues that individuals adopt protective actions when they perceive environmental threats as severe and likely (Rogers, 1975), while the Theory of Planned Behavior (TPB) emphasizes the role of cognitive evaluations in shaping behavioral decisions (Ajzen, 1991). Although originating from different theoretical traditions, both theories share a common premise: external stimuli influence behavior through cognitive processes.
Building on these insights, this study adopts the Information-Cognition-Behavior (ICB) framework as an integrative analytical lens. Rather than proposing a new standalone theory, the ICB framework synthesizes the core logic of PMT and TPB, emphasizing that external information influences behavioral outcomes through individuals’ cognitive evaluations. In the context of meteorological disaster adaptation, government information services represent the primary informational input, while farmers’ cognitive responses determine whether such information is translated into adaptive action. Within this framework, risk perception and government trust are conceptualized as two parallel cognitive mechanisms (Lu et al., 2025). Risk perception reflects farmers’ assessment of meteorological threats, whereas government trust reflects their evaluation of the credibility and competence of the institutions providing disaster-related information and support (Wen and Wu, 2022). In rural China, where government agencies play a dominant role in meteorological warning dissemination and disaster governance, both mechanisms are expected to influence adaptive behavior. The following subsections develop the specific hypotheses regarding these cognitive and behavioral relationships.
2.3 The mediating role of risk perception
Risk perception refers to farmers’ subjective assessment of the likelihood and severity of losses caused by meteorological disasters. It is a key psychological driver of adaptive behavior: farmers who perceive climate risks as serious and probable are more likely to take protective action (Rogers, 1975).
Government information services can raise farmers’ risk perception through multiple channels. Disaster early warnings and weather forecasts make climate threats more salient and concrete. Technical training programs improve farmers’ understanding of how meteorological events affect crop production. Together, these services help farmers recognize risks they might otherwise underestimate or ignore. Under conditions of bounded rationality (Simon, 1959), farmers rely heavily on external information to form risk judgments, and government-provided information plays a central role in shaping these judgments (Li et al., 2021; Song et al., 2022). By reducing information gaps and improving risk literacy, government information services lower the costs of adopting adaptive strategies and increase the perceived urgency of doing so (Conley and Udry, 2010). We therefore propose:
H1: Risk perception mediates the relationship between government information services and adaptive behavior.
H1a: Government information services positively influence large-scale growers' risk perception.
H1b: Higher risk perception promotes adaptive behaviors among large-scale growers.
2.4 The mediating role of government trust
Government trust refers to farmers’ confidence in the government’s ability and willingness to fulfill its public responsibilities, particularly in disaster prevention and response. It reflects a cognitive and affective evaluation of institutional competence and integrity (Xu et al., 2023). It is worth noting that risk perception reflects farmers’ assessment of meteorological threats, whereas government trust reflects their evaluation of the institutions responsible for providing disaster-related information and support. While both serve as cognitive mediators within the Information-Cognition-Behavior framework, they represent distinct psychological mechanisms through which information shapes behavior.
Government information services can strengthen trust by demonstrating the government’s capacity to deliver timely, accurate, and useful guidance. When farmers receive actionable disaster warnings, effective technical support, and relevant policy updates, they develop a more positive evaluation of government competence (Lu et al., 2014; Gai et al., 2021). Targeted services for large-scale growers—such as customized agronomic advice and precision early warnings—further signal that the government is attentive to their specific needs, reinforcing perceived legitimacy (Fu et al., 2025). Once established, trust acts as a motivational resource: farmers who trust the government are more willing to follow official recommendations and adopt government-endorsed adaptive practices (Zhang et al., 2023). This mechanism has been documented across a range of pro-environmental behaviors, including ecological compensation compliance (Xu et al., 2021), pesticide reduction (Qi, 2024), and organic fertilizer adoption (Tao et al., 2022). We therefore propose:
H2: Government trust mediates the relationship between government information services and adaptive behavior.
H2a: Government information services positively influence large-scale growers' government trust.
H2b: Higher government trust promotes adaptive behaviors among large-scale growers.
2.5 The moderating role of social capital
Social capital refers to the social resources available to large-scale grain growers through their networks of relationships, including ties with neighbors, local officials, and community organizations (Gai et al., 2021). Unlike government trust, which reflects an individual’s evaluation of institutional credibility, social capital reflects access to information, support, and collective resources through social interactions. It shapes how effectively farmers can access, interpret, and act on information.
Farmers with stronger social networks are generally better positioned to obtain supplementary information, exchange experiences, and verify the credibility of disaster-related communications (McClurg, 2003; Carreras and Bowler, 2019). As a result, government information services may become more salient and persuasive among growers with higher levels of social capital. Frequent interactions with fellow villagers and local officials can reinforce the transmission of meteorological warnings and policy information, thereby strengthening farmers’ awareness of climate-related risks. At the same time, social connections with governmental actors may enhance the perceived credibility of official information and reinforce positive evaluations of government competence (Xu, 2017; Baxter et al., 2008). Consequently, social capital is expected to strengthen the effects of government information services on both risk perception and government trust. We therefore propose:
H3a: Social capital positively moderates the relationship between government information services and risk perception.
H3b: Social capital positively moderates the relationship between government information services and government trust.
The full theoretical framework, incorporating dual mediation and moderated pathways, is illustrated in Figure 1.
Figure 1
3 Data, variables, and methods
3.1 Data sources
The data used in this study were collected through a household survey of large-scale grain growers conducted in Jiangxi Province, China, between July and September 2019. Although the survey was conducted in 2019, the study focuses on the underlying behavioral mechanisms linking government information services, cognitive responses, and adaptive behavior, which are relatively stable and not tied to a specific policy event or short-term shock. Jiangxi is one of China’s major rice-producing regions and is frequently affected by meteorological disasters, making it a suitable setting for studying climate adaptation behavior. This study focuses on large-scale grain growers because their production decisions have important implications for food security and agricultural resilience, while their larger operational scale may increase vulnerability to climate-related risks. A stratified sampling strategy was adopted, covering representative rice cultivation zones across 11 prefectures in northern and central Jiangxi. Trained enumerators administered face-to-face questionnaires with household heads managing cultivation plots of at least 50 mu (approximately 3.33 hectares). A total of 536 questionnaires were distributed, of which 414 were valid, yielding a response rate of 77.23%.
3.2 Ethical statement
This study was conducted in accordance with the ethical principles of the Declaration of Helsinki. All participants were fully informed of the study’s purpose, the voluntary nature of their participation, and the anonymity of their responses before the interview began. Verbal informed consent was obtained from all respondents. No identifiable personal information was collected or retained.
3.3 Variables
3.3.1 Independent variable
Government information services (GIS) are the key independent variable, measured using four items adapted from Li et al. (2017): (1) meteorological disaster early warnings, (2) disaster response information dissemination, (3) agricultural technology extension, and (4) capacity-building programs. The four items were aggregated into a composite GIS index by calculating the mean score, with higher values indicating greater exposure to government information services.
3.3.2 Dependent variable
The dependent variable is farmers’ adaptive behavior toward meteorological disasters, measured using five items following Tong et al. (2018). The five behaviors reflect the most commonly adopted adaptation strategies among large-scale growers: conservation tillage, infrastructure improvement, precision agrochemical application, adjustment of farming schedules, and increased procurement of fertilizers and pesticides. Exploratory factor analysis confirmed that all five items load on a single factor (KMO = 0.812), and the scale demonstrates acceptable internal consistency (Cronbach’s α = 0.799), supporting the reliability of the composite index. Variable definitions and coding are shown in (see Appendix Table S1).
3.3.3 Mediating variable
Risk perception is measured as farmers’ subjective assessment of the likelihood of three meteorological events: spring overcast and low-light conditions, late spring cold snaps, and flooding. This operationalization follows Fischhoff et al. (1978) and Zhang et al. (2021).
Government trust is measured using a single-item scale assessing farmers’ overall confidence in the government’s meteorological disaster governance. Specifically, respondents were asked to rate their degree of trust in the government’s performance in managing meteorological disasters on a five-point Likert scale ranging from 1 (strongly distrust) to 5 (strongly trust), following Xu et al. (2023). Variable definitions and coding are shown in (see Appendix Table S1).
3.3.4 Moderating variable
Social capital (SC) is operationalized through three dimensions: quality of relationships with fellow villagers, frequency of communication with government staff, and participation in village activities, following Gai et al. (2021). Variable definitions and coding are shown in (see Appendix Table S1).
3.3.5 Control variables
Following previous studies on farmers’ adaptation behavior (Deressa et al., 2009; Huang et al., 2017) and drawing on the Sustainable Livelihood Framework (DFID, 1999), this study controls for a range of individual, household, and production characteristics, including age, education, income, local resident, household size, health status, farming experience, and farm size.
3.4 Econometric model
This study employs a system of simultaneous equations to examine the direct and indirect effects of government information services on adaptive behavior. Considering the potential endogeneity arising from the interrelationships among risk perception, government trust, and adaptive behavior, the three-stage least squares (3SLS) estimator is adopted. Compared with single-equation estimation, 3SLS allows the simultaneous estimation of multiple interrelated equations while accounting for cross-equation error correlations. The simultaneous equations system takes the following form as specified in Equation 1:
Where AB denotes adaptive behavior, GIS denotes government information services, RP denotes risk perception, GT denotes government trust, X denotes a vector of control variables, and ε₁-ε₃ are error terms. (2) and (3) model risk perception and government trust as functions of government information services, capturing the indirect pathways through which GIS affects adaptive behavior. To test the moderating role of social capital, interaction terms between government information services and each dimension of social capital were subsequently introduced into Equations 2 and 3. The moderation effects were estimated separately for intra-village relational networks, governmental liaison frequency, and participation in village activities.
The 3SLS estimator proceeds in three stages. In the first stage, each endogenous variable is regressed on all exogenous variables and instruments to obtain predicted values. In the second stage, the predicted values replace the observed endogenous regressors in each structural equation, equivalent to equation-by-equation two-stage least squares (2SLS). In the third stage, generalized least squares (GLS) estimation is applied to the full system, accounting for contemporaneous correlations across equation disturbances and improving estimation efficiency relative to separate 2SLS estimation. The appropriateness of 3SLS over single-equation methods is empirically confirmed by the cross-equation disturbance term correlation tests reported in Section 4.1, which show that all pairwise correlations among equation residuals are statistically significant at the 1% level.
To address the potential endogeneity of risk perception, participation in agricultural skills training programs is introduced as an instrumental variable (IV). Training participation is associated with enhanced meteorological hazard literacy, a cognitive precursor to risk perception, satisfying the relevance condition (first-stage F = 8.36, p = 0.004). For the moderation analysis, all continuous variables entering interaction terms were mean-centered prior to estimation to reduce multicollinearity.
4 Results
4.1 Baseline estimation results
4.1.1 Disturbance term correlation test
Table 1 reports the correlations among the disturbance terms of the three equations. All pairwise correlations are positive and statistically significant at the 1% level (0.149 to 0.181), confirming the presence of contemporaneous correlation across equations. This justifies the use of the 3SLS estimator rather than separate OLS or 2SLS regressions.
Table 1
| Variable | Government trust | Risk perception | Adaptive behavior |
|---|---|---|---|
| Government trust | 1 | ||
| Risk perception | 0.149*** | 1 | |
| Adaptive behavior | 0.181*** | 0.171*** | 1 |
Correlation of disturbance terms.
*** p < 0.01, ** p < 0.05, * p < 0.1.
4.1.2 Baseline 3SLS estimation results
Table 2 presents the full 3SLS system estimates. Both risk perception and government trust are positively associated with adaptive behavior (β = 0.1906 and 0.1764, respectively, p < 0.01). Government information services are positively associated with both mediators (p < 0.01). The findings provide initial evidence for the dual cognitive pathways linking information services to meteorological disaster adaptation.
Table 2
| Variable | Adaptive behavior | Risk perception | Government trust |
|---|---|---|---|
| Government information services | 0.1449** (2.51) | 0.2700*** (3.23) | |
| Risk perception | 0.1906*** (3.19) | ||
| Government trust | 0.1764*** (3.05) | ||
| Income | 0.0779 (1.00) | 0.0330(0.41) | 0.0810 (0.85) |
| Age | 0.0028 (0.40) | −0.0130(−1.50) | 0.0107 (1.22) |
| Local resident | −0.3953** (−2.27) | ||
| Household size | −0.0389 (−1.38) | 0.0631**(1.99) | |
| Health status | 0.0171 (0.22) | −0.0066 (−0.07) | |
| Education | −0.2900*** (−2.83) | ||
| Farming experience | 0.0024 (0.44) | ||
| Farm size | 0.0006*** (3.12) | ||
| Constant | 0.0062 (0.01) | 0.4272(0.79) | −0.2893 (−0.38) |
| R2 | 0.122 | 0.147 | 0.109 |
3SLS system estimation results.
4.2 Endogeneity and robustness tests
4.2.1 Endogeneity analysis
To address potential endogeneity of risk perception, participation in agricultural skills training programs is introduced as an instrumental variable. Training participation is associated with enhanced meteorological hazard literacy, a cognitive precursor to risk perception, satisfying the relevance condition (first-stage F = 8.36, p = 0.004).
As shown in Column 2 of Table 3, the coefficients for risk perception and government trust remain positive and significant at the 1% level, consistent with the baseline results in Table 2. Columns 3 and 4 confirm that government information services remain positively associated with both risk perception and government trust. These results further support the proposed mediation mechanisms and suggest that the baseline estimates are reasonably robust to potential endogeneity concerns.
Table 3
| Variable | Adaptive behavior | Risk perception | Government trust |
|---|---|---|---|
| Government Information Services | 0.1449** (2.51) | 0.2432*** (2.96) | |
| Risk Perception | 0.2270*** (3.80) | ||
| Government Trust | 0.1771*** (3.06) | ||
| Income | 0.0766 (0.98) | 0.0330 (0.41) | 0.0570 (0.61) |
| Age | 0.0029 (0.41) | −0.0130 (−1.50) | 0.0116 (1.36) |
| Local resident | −0.3946** (−2.27) | ||
| Household size | −0.0382 (−1.36) | 0.0631** (1.99) | |
| Health | 0.0169 (0.22) | 0.0096 (0.11) | |
| Education | −0.2700*** (−2.68) | ||
| Farming experience | 0.0024(0.44) | ||
| Farm size | 0.0006*** (3.12) | ||
| Skills training | 0.170*** (3.97) | ||
| Constant | 0.0050 (0.01) | 0.4272 (0.79) | −0.6674 (−0.88) |
| R2 | 0.121 | 0.140 | 0.117 |
3SLS results after introducing instrumental variables.
4.2.2 Robustness tests
Two robustness checks were conducted: (1) replacing 3SLS with OLS, and (2) redefining adaptive behavior as the count of strategies adopted. As shown in Table 4, coefficient signs and significance levels are consistent with the main results across both specifications, confirming the reliability of the baseline estimates.
Table 4
| Variable | Replacement model (OLS) | Replacement variable | ||||
|---|---|---|---|---|---|---|
| Adaptive behavior | Risk perception | Government trust | Adaptive behavior | Risk perception | Government trust | |
| Risk perception | 0.1906*** (3.13) | 0.4428** (2.31) | ||||
| Government trust | 0.1764*** (2.99) | 0.8018** (2.55) | ||||
| Government information services | 0.1449** (2.46) | 0.2700*** (3.15) | 0.3352** (2.51) | 0.3675*** (3.23) | ||
| Constant | 0.0062 (0.01) | 0.4272 (0.77) | −0.2893 (−0.37) | 4.1543** (2.46) | 2.9771*** (6.15) | 0.9764** (2.23) |
| R2 | 0.122 | 0.140 | 0.123 | 0.112 | 0.140 | 0.123 |
Robustness test results.
4.3 Mechanism analysis
4.3.1 Mediation analysis
Although the baseline model was estimated using 3SLS to account for simultaneity, the mediation effects were tested using Sobel tests and Bootstrap procedures applied to each pathway separately, and the sequential testing of individual pathways provides conservative estimates of the indirect effects Table 5.
Table 5
| Effect type | Coef. | Std. err. | z | P > |z| | [95% conf. interval] | Proportion | |
|---|---|---|---|---|---|---|---|
| lower limit | upper limit | ||||||
| Total effect | 0.171 | 0.077 | 2.22 | 0.026 | 0.025 | 0.158 | — |
| Indirect effect | 0.077 | 0.029 | 2.13 | 0.033 | 0.002 | 0.059 | 44.86% |
| Direct effect | 0.094 | 0.074 | 1.43 | 0.024 | 0.013 | 0.181 | 54.97% |
Mediation effect of risk perception.
The mediation effects of risk perception and government trust were tested using Sobel tests and Bootstrap procedures (5,000 resamples). Both indirect pathways are statistically significant: risk perception (Z = 2.13 > 1.96) and government trust (Z = 2.33 > 1.96), confirming H1 and H2.
To compare the relative strength of the two pathways, all variables were standardized. Both standardized and unstandardized results consistently show that the mediating effect of government trust (standardized indirect effect = 0.082) exceeds that of risk perception (0.077). This finding suggests that, in the context of rural China, institutional trust appears to play a somewhat stronger role than threat appraisal in translating government information into adaptive action Table 6.
Table 6
| Effect type | Coef. | Std. err. | z | P > |z| | [95% conf. interval] | Proportion | |
|---|---|---|---|---|---|---|---|
| lower limit | upper limit | ||||||
| Total effect | 0.174 | 0.083 | 2.09 | 0.036 | 0.009 | 0.337 | — |
| Indirect effect | 0.082 | 0.021 | 2.33 | 0.020 | 0.008 | 0.090 | 47.02% |
| Direct effect | 0.092 | 0.063 | 0.73 | 0.468 | −0.077 | 0.169 | 52.97% |
Mediation effect of government trust.
4.3.2 Moderation analysis
Table 7 presents the moderation analysis. Social capital is operationalized through three dimensions: intra-village relational networks, governmental liaison frequency, and participation in village activities.
Table 7
| Variable | Adaptive behavior | Risk perception | Government trust |
|---|---|---|---|
| Government information services | 0.2908**(2.25) | 0.4876***(3.13) | |
| Intra-village relational networks (SC1) | 0.0418(1.46) | 0.0829**(2.40) | |
| Interaction term (GIS × SC1) | 0.5920***(4.67) | 0.1766(1.35) | |
| Risk perception | 0.1906***(3.19) | ||
| Government trust | 0.1764***(3.05) | ||
| Constant | 0.0062(0.01) | 1.4416*(1.92) | 0.0420(0.07) |
| Governmental liaison Frequency (SC2) | 0.1372*(1.89) | 0.0564(0.64) | |
| Interaction term (GIS × SC2) | 0.0783**(2.54) | 0.0751**(2.15) | |
| Risk perception | 0.1906***(3.19) | ||
| Government trust | 0.1764***(3.05) | ||
| Constant | 0.0062(0.01) | 1.4806**(2.52) | 0.9606(1.30) |
| Participation in village activities (SC3) | 0.0197(0.51) | 0.0778*(1.66) | |
| Interaction term (GIS × SC3) | 0.1545***(2.81) | 0.0394(0.67) | |
| Risk perception | 0.1906***(3.19) | ||
| Government trust | 0.1764***(3.05) | ||
| Constant | 0.0062(0.01) | 0.4920(0.68) | −0.1419(−0.22) |
Moderation effects of social capital.
The interaction between government information services and intra-village relational networks is positively significant for risk perception but not for government trust. The interaction with governmental liaison frequency is positively significant for both risk perception and government trust. The interaction with participation in village activities is positively significant for risk perception but not for government trust. Overall, the results indicate that social capital is associated with stronger relationships between government information services and cognitive responses on risk perception across all three dimensions, while its moderating effect on government trust is significant only for governmental liaison frequency.
4.4 Heterogeneity analysis by human capital
4.4.1 Education level
Growers were divided into two groups: high educational attainment (senior high school or above) and low educational attainment (below senior high school). Table 8 presents the stratified results.
Table 8
| Variable | High education level | Low education level | ||||
|---|---|---|---|---|---|---|
| Adaptive behavior | Risk perception | Government trust | Adaptive behavior | Risk perception | Government trust | |
| Risk perception | 0.2445*** (3.65) | 0.0096 (0.07) | ||||
| Government trust | 0.2178*** (3.29) | 0.0726 (0.56) | ||||
| Government information services | 0.1737** (2.22) | 0.2660** (2.26) | 0.0391 (0.46) | 0.2238* (1.96) | ||
| Constant | −0.1223 (−0.17) | 0.3115 (0.45) | −0.7791 (−0.74) | 0.0220 (0.02) | 1.1309 (1.26) | −0.1750 (−0.16) |
| R2 | 0.167 | 0.140 | 0.073 | 0.113 | 0.229 | 0.185 |
Heterogeneity analysis by education level.
Among high-education growers (Columns 1–3), both risk perception and government trust have significant positive effects on adaptive behavior. Among low-education growers, neither effect is statistically significant. This pattern indicates that the cognitive pathways identified in this study are relatively more evident among growers with higher educational attainment.
4.4.2 Income level
Growers were divided into high and low-income groups based on self-reported income thresholds. Table 9 presents the results.
Table 9
| Variable | High-income | Low-income | ||||
|---|---|---|---|---|---|---|
| Adaptive behavior | Risk perception | Government trust | Adaptive behavior | Risk perception | Government trust | |
| Risk perception | 0.2087*** (3.43) | −0.0807 (−0.31) | ||||
| Government trust | 0.1875*** (3.15) | 0.2083 (1.01) | ||||
| Government information services | 0.1288** (2.16) | 0.2546*** (2.90) | 0.4266** (2.03) | −0.0603 (−0.21) | ||
| Constant | 0.6900 (1.19) | 0.5903 (1.20) | 0.0132 (0.02) | −0.3661 (−0.18) | −1.9909 (−0.92) | −2.8805 (−1.34) |
| R2 | 0.132 | 0.142 | 0.124 | 0.530 | 0.542 | 0.656 |
Heterogeneity analysis by income level.
Among high-income growers, both risk perception and government trust have significant positive effects on adaptive behavior. Among low-income growers, neither relationship is significant. Additionally, while government information services significantly predict risk perception across both income groups, their effect on government trust is only significant among high-income growers. These findings suggest that income conditions may shape the extent to which information services are associated with adaptive behavior through cognitive mechanisms.
5 Discussion
The findings of this study both confirm and extend existing scholarship on information-driven climate adaptation. While prior research has established that government information services promote adaptive behavior (Madhuri and Sharma, 2020; Wang et al., 2023), the mechanisms through which this occurs have remained underspecified. Our results clarify these mechanisms by suggesting that risk perception and government trust operate as parallel yet asymmetric cognitive pathways, with institutional trust exhibiting a somewhat stronger indirect effect than threat appraisal. This asymmetry diverges from smallholder-focused studies in sub-Saharan Africa and South Asia, where risk perception consistently emerges as the dominant driver (Anshuka et al., 2021; Truelove et al., 2015), but aligns with evidence from hierarchical governance contexts showing that source credibility independently shapes behavioral compliance (Lu et al., 2014; Xu et al., 2023). We attribute this divergence to the structural position of government in rural China as the near-exclusive provider of meteorological disaster information: when source diversity is limited, the legitimacy of the dominant source may become relatively more important. The moderating role of social capital further nuances existing findings. Rather than treating social capital as a uniform amplifier of information effects (Lin et al., 2021; Xu, 2017), our dimension-level analysis shows that governmental liaison frequency is associated with stronger cognitive responses, while peer networks primarily amplify risk perception, suggesting that vertical and horizontal social ties serve qualitatively distinct functions in translating information into action. Finally, the absence of significant mediating effects among low-education and low-income producers challenges the implicit assumption in much of the adaptation literature that information provision benefits all farmers equally (Hansen et al., 2011; Feng et al., 2016). and points instead to a structural equity gap in information-based governance that warrants targeted policy attention.
The relatively moderate effect sizes observed in this study also suggest that government information services represent only one component of farmers’ adaptation decision-making. Adaptive behavior is influenced not only by information acquisition and cognitive responses, but also by economic resources, production conditions, institutional support, and local environmental contexts. Therefore, while information services contribute to adaptation through risk perception and government trust, their influence should be understood as incremental rather than transformative. This finding highlights the importance of combining information-based interventions with broader institutional and socioeconomic support measures to improve adaptation capacity among large-scale grain producers.
6 Conclusion
This study investigated how government information services influence meteorological disaster adaptive behaviors among large-scale grain producers in Jiangxi Province, China, using a simultaneous equations model estimated by three-stage least squares (3SLS). By integrating the Information-Cognition-Behavior (ICB) framework with empirical data from 414 large-scale grain producers, this study examined the dual mediating roles of risk perception and government trust, the moderating role of social capital, and the boundary conditions imposed by human capital heterogeneity.
The results demonstrate that government information services are positively associated with both risk perception and government trust, which in turn promote adaptive behavior. The mediating effect of government trust is stronger than that of risk perception, suggesting that institutional credibility is a somewhat stronger behavioral mechanism than threat appraisal alone in governance contexts where government serves as the primary information authority. Social capital exhibits moderating effects on the relationships between government information services and cognitive responses, with governmental liaison frequency showing the most consistent amplifying effects. Heterogeneity analysis reveals that these mediating effects are significant only among producers with higher educational attainment and income levels, highlighting a structural equity gap in the effectiveness of uniform information-based governance strategies.
These findings carry practical implications for agricultural disaster governance in China. Our field interviews with large-scale grain producers revealed that farmers are highly dependent on and eager for accurate government meteorological information; yet, current weather forecasts are typically issued only at the prefecture level, failing to capture the micro-level climate variability that directly affects farm-level production decisions. This gap underscores the need to downscale forecasting systems to the county or township level and develop producer-specific early warning platforms delivering real-time alerts through accessible digital channels. Beyond improving information precision, policymakers should prioritize institutional trust-building, as our findings show that perceived government credibility drives adaptive action more powerfully than information volume alone. Social capital networks—particularly government–farmer liaison mechanisms—should also be leveraged to amplify information effectiveness. Finally, for low-education and low-income producers who may not respond to information-based strategies alone, complementary interventions, including simplified guidance materials and subsidized adaptive inputs, are needed to ensure more equitable adaptation outcomes.
This study has several limitations, including its cross-sectional design, single-province sample, and reliance on 2019 survey data. Future research should employ longitudinal or experimental designs, extend the analysis to other agricultural regions and producer types, and incorporate stronger instrumental variables to improve causal identification. Replicating the dual-mediation model in different governance and agro-ecological contexts would further test the generalizability of the ICB framework and its applicability to broader climate adaptation scholarship.
Statements
Data availability statement
The raw data supporting the conclusions of this article will be made available by the authors, without undue reservation.
Author contributions
TL: Writing – original draft, Data curation, Conceptualization, Formal analysis. HL: Writing – original draft, Methodology, Investigation, Conceptualization. ZH: Supervision, Writing – review & editing, Funding acquisition.
Funding
The author(s) declared that financial support was received for this work and/or its publication. This study was supported by the Jiangxi Agricultural University Interdisciplinary Innovation Cultivation Project (Grant No. JXAU-02-2025-06) titled “Government Agricultural Cultivation Services and Farmers’ Green Production Behavior: A Case Study of the’Touyan’Project for Rural Industry Revitalization in Jiangxi Province” [2025–2028, 50,000 RMB, Principal Investigator: ZH].
Conflict of interest
The author(s) declared that this work was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.
Generative AI statement
The author(s) declared that Generative AI was not used in the creation of this manuscript.
Any alternative text (alt text) provided alongside figures in this article has been generated by Frontiers with the support of artificial intelligence and reasonable efforts have been made to ensure accuracy, including review by the authors wherever possible. If you identify any issues, please contact us.
Publisher’s note
All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher.
Supplementary material
The Supplementary material for this article can be found online at: https://www.frontiersin.org/articles/10.3389/fsufs.2026.1874810/full#supplementary-material
References
1
AD.LiaoY. F.WuW.LiuX. (2022). Research on information service mode of government disaster risk prevention. J. Catastrophology37, 186–191.
2
AjzenI. (1991). The theory of planned behavior. Organ. Behav. Hum. Decis. Process.50, 179–211. doi: 10.1016/0749-5978(91)90020-T
3
AkerJ. C. (2011). Dial ‘A’ for agriculture: A review of information and communication technologies for agricultural extension in developing countries. Agric. Econ.42, 631–647. doi: 10.1111/j.1574-0862.2011.00545.x
4
AnshukaA.van OgtropF. F.SandersonD.ThomasE.NeefA. (2021). Vulnerabilities shape risk perception and influence adaptive strategies to hydro-meteorological hazards: a case study of Indo-Fijian farming communities. Int. J. Disaster Risk Reduct.:102401. doi: 10.1016/j.ijdrr.2021.102401
5
ArbuckleJ. G.MortonL. W.HobbsJ. (2015). Understanding farmer perspectives on climate change adaptation and mitigation: the roles of trust in sources of climate information, climate change beliefs, and perceived risk. Environ. Behav.47, 205–234. doi: 10.1177/0013916513503832,
6
AvenT.RennO. (2009). On risk defined as an event where the outcome is uncertain. J. Risk Res.12, 1–11. doi: 10.1080/13669870802488883
7
BaxterL.EgbertN.HoE. (2008). Everyday health communication experiences of college students. J. Am. Coll. Heal.56, 427–436. doi: 10.3200/jach.56.44.427-436
8
BurnhamM.MaZ. (2017). Climate change adaptation: factors influencing Chinese smallholder farmers' perceived self-efficacy and adaptation intent. Reg. Environ. Chang.17, 171–186. doi: 10.1007/s10113-016-0975-6
9
CarrerasM.BowlerS. (2019). Community size, social capital, and political participation in Latin America. Polit. Behav.3, 723–745. doi: 10.1007/s11109-018-9470-8
10
ChaiZ. H.ZhaoY. F. (2013). Farmers’ awareness of agricultural insurance premium subsidy policies and trust in government: a survey of 500 farmers in Inner Mongolia, China. Rural Econ.4, 66–69. doi: 10.26957/j.cnki.cn51-1029/f.2013.04.016
11
ChenQ. (2014). Advanced Econometrics and Stata Applications. Beijing: Higher Education Press.
12
ChenY.LiS. X. (2009). Group identity and social preferences. Am. Econ. Rev.99, 431–457. doi: 10.1257/aer.99.1.431
13
ConleyT. G.UdryC. R. (2010). Learning about a new technology: pineapple in Ghana. Am. Econ. Rev.100, 35–69. doi: 10.1257/aer.100.1.35
14
DeressaT. T.HassanR. M.RinglerC.AlemuT.YesufM. (2009). Determinants of farmers' choice of adaptation methods to climate change in the Nile Basin of Ethiopia. Glob. Environ. Change19, 248–255. doi: 10.1016/j.gloenvcha.2009.01.002
15
DFID (1999). Sustainable Livelihoods Guidance Sheets. London: Department for International Development.
16
FengX. L.ChenZ. X.HuoX. X. (2016). Factors influencing farmers adapting behavior to drought: based on 1079 apple household survey data. J Arid Land Resour Environ.30, 43–49. doi: 10.13448/j.cnki.jalre.2016.078
17
FischbacherU.GächterS.FehrE. (2001). Are people conditionally cooperative? Evidence from a public goods experiment. Econ. Lett.71, 397–404. doi: 10.1016/s0165-1765(01)00394-9
18
FischhoffB.SlovicP.LichtensteinS.ReadS.CombsB. (1978). How safe is safe enough? A psychometric study of attitudes towards technological risks and benefits. Policy Sci.9, 127–152. doi: 10.1007/BF00143739
19
FuW. F.SunJ.ZhangY. T. (2025). The impact of government response on farmers’ willingness to participate in environmental improvement: analysis of the chain mediating effects of policy value identification and government trust. Agric. Technol.45, 155–160. doi: 10.19754/j.nyyjs.20250228031
20
GaiH.YanT. W.ZhouX. S. (2021). How to achieve long-term effects of policy publicity? An analysis on farmers’ behaviors of continued straw returning to the field in Hubei province. China Rural Surv.6, 65–84. doi: 10.20074/j.cnki.11-3586/f.2021.06.004
21
GrothmannT.PattA. (2005). Adaptive capacity and human cognition: the process of individual adaptation to climate change. Glob. Environ. Change15, 199–213. doi: 10.1016/j.gloenvcha.2005.01.002
22
HansenJ. W.MasonS. J.SunL.TallA. (2011). Review of seasonal climate forecasting for agriculture in Sub-Saharan Africa. Exp. Agric.47, 205–240. doi: 10.1017/S0014479710000876
23
HuangK. Y.GuanH. Y.YangZ. M.ChenY. C. (2017). Farmers' perceptions of climate change and their adaptive behaviors in poor areas. Environ. Sci. Technol.40, 200–205.
24
LiM.LuQ.QiaoD. (2017). Technological cognition, government support and farmers’ adoption of water-saving irrigation technology. J Arid Land Resour Environ.31, 27–32. doi: 10.13448/j.cnki.jalre.2017.375
25
LiZ. Y.ZhangM. X.WeiR.ZhuY. C. (2021). WeChat use and altruistic behavior in the COVID-19 crisis: the mediated effects of risk perception and public trust. Chin J Journal Commun.43, 6–22. doi: 10.20077/j.cnki.11-1262/f.2017.01.004
26
LinL. M.HeX. L.HanY. Q. (2021). Impact of rural residents’ ecological cognition and relationship on their cooperative behavior in the domestic garbage management: based on the survey data of fixed observation points in rural areas of Fujian province. J Ecol Rural Environ.37, 1301–1309. doi: 10.19741/j.issn.1673-4831.2020.1028
27
LuJ.AldrichJ.ShiT. J. (2014). Revisiting media effects in authoritarian societies: democratic conceptions, collectivistic norms, and media access in urban China. Polit. Soc.42, 253–282.
28
LuH. G.ChengX. B.ZhangJ. B. (2025). Impact of digital meteorological information services on the climate adaptation behavior of rice farmers: based on TAM and TPB integration frameworks. J. China Agric. Univ.30, 189–202.
29
LuoL.ZhaoX. Y.WangY. R.ZhangQ. (2016). Farmers’ perception and its influence factors of climate change in the Shiyang River basin. J. Desert Res.36, 1171–1181.
30
MaF. C.XiaY. K. (2003). The status-quo and innovation of government information service. Libr. Inf. Serv.12, 19–23.
31
MaddisonD. (2007). The perception of and adaptation to climate change in Africa. World Bank Policy Research Working PaperNo. 4308.
32
MadhuriM.SharmaU. (2020). How do farmers perceive climate change? A systematic review. Clim Change162, 991–1010.
33
McClurgS. D. (2003). Social networks and political participation: the role of social interaction in explaining political participation. Polit. Res. Q.4, 448–465.
34
MsangiS.RosegrantM. W.YouL. (2006). Ex post assessment methods of climate forecast impacts. Clim. Res.33, 67–79. doi: 10.3354/cr033067,
35
MuemaE. M. (2018). Determinants of Access and Use of Climate Information Services among Smallholder Farmers in Makueni County, Kenya. Nairobi: University of Nairobi.
36
QiQ. (2024). Research on social trust, risk perception and farm households’ pesticide use behavior. J. Shenyang Univ. Technol. Soc. Sci.17, 663–672.
37
QiY. J.YuanY. (2017). Analysis of the construction of rural public meteorological information service system in Yangcheng County. Mod. Agric. Sci. Technol.16, 211–214.
38
RogersR. W. (1975). A protection motivation theory of fear appeals and attitude change. J. Psychol.91, 93–114. doi: 10.1080/00223980.1975.9915803,
39
SmitB.WandelJ. (2006). Adaptation, adaptive capacity and vulnerability. Glob. Environ. Change.16, 282–292. doi: 10.1016/j.gloenvcha.2006.03.008
40
SimonH. A. (1959). Theories of decision-making in economics and behavioral science. Am. Econ. Rev.49, 253–283.
41
SongY.ChenL.LiQ.HeJ. X.WangY. (2022). AI ethical risk perception, trust and public participation. Stud. Sci. Sci.40, 1153–1162. doi: 10.16192/j.cnki.1003-2053.20211101.001
42
TaoY.QiuX. W.ZhouY. X.HuJ. L. (2022). The conflict between risk perception, social trust and farmers’ organic fertilizer substitution behavior. Aust. J. Agric. Econ.5, 49–64. doi: 10.13246/j.cnki.jae.2022.05.005
43
TianS. Y.ChenJ. Y. (2014). Farmers' adaptation to climate change within the sustainable livelihood framework. China Popul. Resour. Environ.24, 31–37.
44
TongQ. M.ZhangL.ZhangJ. B. (2018). Study on the influence of characteristics of family endowment on households’ adaptive behavior to climate change. Soft Sci.32, 136–139. doi: 10.13956/j.ss.1001-8409.2018.01.29
45
TrueloveH. B.AmandaR. C.LankaT. (2015). A socio-psychological model for analyzing climate change adaptation: a case study of Sri Lankan paddy farmers. Glob. Environ. Change31, 85–97. doi: 10.1016/j.gloenvcha.2014.12.010
46
WangX. Y.ZhengR.WeiM. Z.GaoZ. H. (2023). Research on the formation mechanism and promotion path of government public information service accessibility in crisis situations: taking major public health emergencies for instance. Information and Documentation Services44, 102–112.
47
WeiG. R.ShiX. M.SunL. F.WangL.LiH. J. (2020). The farmers′ perception of climate change and diversity of adaptive behavior in dry farming region. Journal of Shaanxi Normal University:Natural Science Edition48, 116–124. doi: 10.15983/j.cnki.jsnu.2020.04.018
48
WenH.WuC. H. (2022). Information redundancy, event cognition and the formation mechanism of public coping behavior: taking six public health emergencies of international concern as examples. J Xi’an Univ Financ Econ.35, 118–128. doi: 10.19331/j.cnki.jxufe.2022.02.011
49
WuN.ChenT.ChenY. (2022). On food security in China in the new era. Journal of Fuzhou University (Philosophy and Social Sciences)36, 1–10.
50
XiJ. P. (2022) Hold high the great banner of socialism with Chinese characteristics and strive in unity to build a great modern socialist country in all respects—report to the 20th national congress of the communist party of China
51
XuX. X. (2017). Democratic participation, social capital and government trust in rural public service supply: an empirical analysis based on data from 95 counties in 10 provinces of China. J Sichuan Adm Inst.2, 86–93. doi: 10.13448/j.cnki.jalre.2021.092
52
XuM. Y.GaoJ.WangQ.LiaoY. E. (2023). The relationships among government trust, perception of COVID-19 risk and residents’ tourism support: destination social responsibility as a moderator. J Chin Ecotourism.13, 1–12.
53
XuR. F.LiuW. X.NiQ.ZhaoM. J. (2021). Risk perception, government trust and urban residents’ ecological compensation payment level in Weihe river basin. J Arid Land Resour Environ.35, 10–16.
54
YangY. T.ChenZ. J.ZhongJ. (2012). Factors affecting farmers’ satisfaction with agriculture technology promotion policy: a study based on Jiangxi province. J Jiangxi Agric Univ Soc Sci Ed.11, 23–29.
55
YangJ.HuangZ.ZhangX.ReardonT. (2013). The rapid rise of cross-regional agricultural mechanization services in China. Am. J. Agric. Econ.95, 1245–1251. doi: 10.1093/ajae/aat027
56
ZhangK. J.LiF. D.YinC. B. (2023). How does knowledge diffusion affect the transformation of rice cropping systems?: chain mediation effect based on structural equation model. J. China Agric. Univ.28, 48–66.
57
ZhangL.XieX. Q.KeJ. Z. (2021). Study on the impact of livelihood capital on farmers adaptive behavior to meteorological disasters: mediating effect based on risk perception. Inner Mongolia Sci Technol Econ.18, 46–49.
58
ZhaoX. K.ZhouB. (2012). An analysis on factors influencing large-scale rice-plant farmers' cognition of up-to-date agriculture technology: based on the survey of 1077 farmers in Jiangxi province. China Rural Surv.4, 29–36. doi: 10.20074/j.cnki.11-3586/f.2012.04.003
59
ZhuH. G.ZhouS. D. (2011). Factors influencing southern rice farmers adapting to climate change behavior——based on 346 household survey data of 36 counties in Jiangxi province. J. Nat. Resour.26, 1119–1128.
Summary
Keywords
adaptive behavior, government information services, government trust, meteorological disasters, risk perception
Citation
Liu T, Li H and Hu Z (2026) Government information services and farmers’ adaptive behavior toward meteorological disasters: evidence from large-scale grain producers in China. Front. Sustain. Food Syst. 10:1874810. doi: 10.3389/fsufs.2026.1874810
Received
07 May 2026
Revised
28 June 2026
Accepted
29 June 2026
Published
15 July 2026
Volume
10 - 2026
Edited by
Frank Baffour-Ata, Kwame Nkrumah University of Science and Technology, Ghana
Reviewed by
Nur Muttaqien Zuhri, Universitas Muhammadiyah Semarang, Indonesia
Huang Wei, North China University of Water Conservancy and Electric Power, China
Updates
Copyright
© 2026 Liu, Li and Hu.
This is an open-access article distributed under the terms of the Creative Commons Attribution License (CC BY). The use, distribution or reproduction in other forums is permitted, provided the original author(s) and the copyright owner(s) are credited and that the original publication in this journal is cited, in accordance with accepted academic practice. No use, distribution or reproduction is permitted which does not comply with these terms.
*Correspondence: Zhen Hu, huzhen@jxau.edu.cn
†These authors have contributed equally to this work
Disclaimer
All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article or claim that may be made by its manufacturer is not guaranteed or endorsed by the publisher.