Chinese Farmers’ Low-Carbon Agricultural Technology Adoption Behavior and Its Influencing Factors

Abstract

Low-carbon agricultural technology (LCAT) is essential for China to achieve its carbon emissions peak by 2030 and neutrality by 2060. Farmers’ adoption of LCAT is crucial for adapting to and mitigating climate change risks. This study explores the social-psychological factors shaping farmers’ LCAT adoption behavior, utilizing the Theory of Planned Behavior and the Normative Activation Model. Survey data from 360 farmers in Wuxi, Jiangsu Province, were analyzed using structural equation modeling. Findings show that behavioral attitude, perceived behavioral control, subjective norms, and personal norms have positive and direct effects on farmers’ LCAT adoption. The analyses also discovered four mediation paths that indirectly influence farmers’ LCAT adoption, including Subjective Norms → Personal Norms → Adoption Level; Consequence Awareness → Personal Norms → Adoption Level; Responsibility Attribution → Personal Norms → Adoption Level; and Consequence Awareness → Responsibility Attribution → Personal Norms → Adoption Level. The study deepens our understanding of the social-psychological mechanism underlying farmers’ LCAT adoption behavior. The findings offer valuable insights for promoting low-carbon agricultural technologies and guiding policy development. Recommendations include promoting LCAT by leveraging social influence to enhance social norms, educating farmers on ethical environmental stewardship, raising awareness of farming’s environmental impacts, and providing policy incentives and technical support to reduce adoption barriers.

1. Introduction

Global climate change affects agroecosystems, and agricultural development is adapting in response to these changes [1]. The primary driver of global warming is the emission of greenhouse gases resulting from human activities [2]. Globally, agroforestry and other land uses are the second largest source of GHGs from human activities, accounting for 24.5% of total GHG emissions, higher than transportation, buildings, and industry [3]. Specifically, CH₄ and N₂O dominate the GHG emissions from agricultural activities, accounting for 47% and 58% of total anthropogenic emissions, respectively [4]. The World Food and Agriculture Organization (FAO) projects that emissions from agroforestry are expected to increase by more than 30% by 2050 if current patterns of agroforestry activities are not changed [5]. In China, the greenhouse effect caused by agricultural production is more prominent, with total carbon emissions as high as 8.28~10.97% [6]. Climate change directly affects China’s national food security by shortening the reproductive period of major food crops (e.g., maize, rice, wheat, etc.) and increasing the incidence of pests. Therefore, promoting low-carbon or low-emission agriculture is one of the effective ways to mitigate global climate change, particularly within the agricultural sector [7,8,9].

In May 2022, the Ministry of Agriculture and Rural Affairs of the People’s Republic of China launched the “Implementation Program for Carbon Sequestration and Emission Reduction in Agriculture and Rural Areas”, which is the first all-encompassing scheme driven by China’s response to climate change [10]. Given the planting sector’s ongoing role in carbon emissions, the government’s focus and investment are increasingly directed towards low-carbon farming methods in crop production. A core feature of this initiative is the government’s active involvement. By channeling resources (potentially through incentives, subsidies, or technical assistance), the policy encourages farmers to adopt these low-carbon methods. This support is crucial, as transitioning to new practices often requires financial and educational assistance, especially for farmers who are accustomed to traditional, emission-heavy techniques.

Low-carbon agriculture (LCA) refers to farming practices that reduce greenhouse gas emissions and increase carbon sinks during agricultural production [11,12]. As a system, it is integrated with several agricultural technologies [13]. Low-carbon agricultural technology (LCAT) has been gradually introduced to reduce carbon emissions from agriculture, increase soil carbon storage, and improve air quality. In rice fields, LCAT can help mitigate CH₄ emissions through practices such as alternate wetting and drying, soil testing and precision fertilization, organic fertilization, and no-tillage cultivation [14,15]. For dryland crops and wheat, LCAT reduces the N₂O emissions from chemical fertilizers and utilizes elements such as mixed organic and inorganic fertilizers, the application of controlled or slow-release fertilizers, the deep placement of nitrogen fertilizers, and conservation tillage [15,16]. It is clear that LCATs are complex and tend to be applied in a composite form. In technology diffusion, farmers are more inclined to accept technology clusters [17]. In order to identify the carbon effects of each stage of agricultural production activities, scholars have proposed five systems for LCATs, including reduced tillage, 4R fertilization (right source, right rate, right time, and right place), eco-friendly pesticide application, agricultural film recycling and covering, and straw resource utilization [13]. Meanwhile, promoting the adoption of LCAT by farmers has also become a national policy direction to address climate change [8]. The government has adopted a range of interventions, including economic incentives (e.g., subsidies and tax credits), market regulation (e.g., through business entities and farming cooperatives), and information dissemination (e.g., policies, publicity, and training) [18,19,20]. Despite the substantial benefits of LCATs and extensive efforts made to promote their adoption, farmers’ adoption continues to be limited [18,21].

Addressing the issue of farmers’ LCAT adoption is urgent. Therefore, it is essential to develop targeted policies that cater to farmers’ distinct needs to improve adoption rates. Since the adoption of LCATs is ultimately an individual voluntary choice made by farmers, it is crucial to investigate the mechanism inherent in the farmers’ adoption decisions [21]. For this reason, a large amount of literature on farming technology adoption has emerged. Studies show the importance of demographic, social-psychological, and environmental factors [22,23,24]. Factors affecting farmers’ adoption behavior during the decision-making process have been analyzed through various personal, social, psychological, and farm-related attributes. Two of the common areas are the personal attributes of the adopter (i.e., gender, age, educational background, health status, and social networks) and family farm-related characteristics (i.e., land area, household income, lease duration, land transfer status, degree of mechanization, and number of family members engaged in farming activities) [25,26]. External factors are mainly related to farm management and operations, including policy support, market demand, technological advances, and natural conditions (i.e., soil quality, moisture, and light) [27,28,29]. Apart from this, social-psychological attributes, including attitudes, perceptions beliefs, and norms underlying behaviors, are studied in this area as well. Recent research underscores that farmers’ attitudes, perceived control, and norms tied to specific agricultural technologies significantly shape their decision-making processes, greatly influencing the development of proactive choices [30,31,32,33,34]. Moral attributes and self-perceived values can be regarded as two distinct types of social-psychological factors in farmers’ decisions [35,36]. A moral attribute is a principle or standard that regulates personal behavior and decision-making [37]. It may be derived from societal expectations or personal ethical duties, and it may act as a mediator between personal attitudes and behavioral outcomes [38,39]. Dong and Ogiemwonyi et al. [40,41] pointed out that moral obligation significantly shapes farmers’ willingness to adopt green production, which emphasizes the importance of ethical standards over economic incentives. Similarly, research on forest conservation has illustrated that sustainable harvesting practices reflect the ethical duty to balance ecological and economic demands [42]. Self-perceived value is a personal evaluation of farmers’ choices, competencies, and abilities, shaped by their self-concept (e.g., regarding oneself as environmentally aware) [43,44,45]. Within the agricultural context, LCAT adoption by farmers may depend on their sense of resource availability and perceived benefits, such as financial gain, personal success, or long-term outcomes of their farms. Overall, these psychological factors are interrelated and exert a complex influence on farmers’ behavior. However, the relationships between these constructs are rarely explored, especially in LCATs.

From a more theoretical perspective, models including the Theory of Planned Behavior (TPB) [46] and the Norm Activation Model (NAM) [47] have been employed to explain farmers’ various behaviors, such as green prevention and control [48,49], eco-breeding management [50], integrated pest management [51], organic adoption [52,53,54], and sustainable agricultural technology usage [55]. Each theory possesses distinct features and strengths. The TPB excels in explaining the rational decision-making view, with a strong focus on the role played by behavioral attitude, subjective norms, and perceived behavioral control in affecting their intentions [56,57]. In contrast, the NAM provides a better explanation with regard to a moral and normative view, with a focus on the role played by outcome awareness, responsibility attribution, and personal norms in affecting behavioral intentions [58,59]. However, both theories have a narrow scope in the sense that they focus either on psychological motivations related to self-interest and social pressures or moral obligations linked to altruism. Consequently, neither theory on its own fully captures the complex psychological processes involved in farmers’ decision making. The integrated TPB-NAM model offers a broader framework for understanding farmers’ green production behavior from the perspectives of social-psychological factors, addressing the limitations of each theory when applied independently. This combined approach allows for a holistic analysis of both self-perceived values and moral norms. Furthermore, this dual perspective is particularly relevant in the context of sustainable agriculture, where personal benefits and moral obligations often intersect.

In summary, despite the benefits of LCAT and government support, farmers’ adoption rates remain low. This phenomenon can be attributed to three primary factors. First, current research on LCAT adoption has been scarce, with unclear mechanisms driving the adoption process, particularly in the social-psychological realm. This uncertainty hinders policymakers’ ability to design targeted intervention strategies with scientific evidence, ultimately reducing farmers’ willingness to adopt such practices. Second, LCATs vary widely in type, allowing farmers to choose one or more technologies during the adoption process. The adoption spectrum framework proposed by Han et al. (2023) [60] outlines four critical components, including entirety, variability, sophistication, and longevity. They argue that assessing the degree of adoption can yield richer insights into farmers’ decision making. However, previous studies have predominantly relied on a binary adoption model, which oversimplifies LCAT adoption and fails to capture its complexity and the diversity of behaviors observed. Third, investigations into farmers’ environmental behavior often overlook the interplay of moral attributes and self-values, potentially leading to an incomplete grasp of what drives LCAT adoption. This study aims to deepen the understanding of how farmers adopt LCATs by constructing an integrated TPB-NAM framework. The resulting model is applied to empirically analyze a set of socio-psychological factors that affect farmers’ decisions regarding LCAT adoption. By focusing on the interplay between these socio-psychological factors and adoption behaviors, the study provides new insights into the drivers of LCAT adoption, addressing shortcomings in previous research.

2. Theoretical Analysis and Hypotheses

2.1. Concept Definition

According to the Chinese government’s Implementation Plan for Emission Reduction and Carbon Sequestration in Agriculture and Rural Areas, we defined LCAT as a technical measure to achieve low greenhouse gas emissions and increase carbon sink in agriculture systems. Based on existing research, we categorized LCAT into five systems, including reduced tillage, 4R fertilization (right source, right rate, right time, and right place), eco-friendly pesticide application, agricultural film recycling, and straw resource utilization [61,62] (as shown in

Table 1. Low-carbon agricultural technology system and specific practices.

LCAT System	Specific Practices
Reduced tillage system	No-till, Reduced tillage, Shallow tillage, Deep loosening, Crop rotation, Intercropping, and Relay planting
4R fertilizing system	Right-source fertilization, Right-rate fertilization, Right-time fertilization, and Right-place fertilization
Eco-friendly pesticide application system	Physical pest control, Use of biological pesticides
Agricultural film system	Agricultural film recycling, Agricultural film covering
Straw resource utilization system	Straw biogas treatment, Straw returning to the field

).

2.2. Conceptual Framework

The Theory of Planned Behavior (TPB) was proposed by Ajzen [46]; it is an extension of the earlier Theory of Reasoned Action (TRA). According to TPB, human behavior is primarily driven by intention, which is developed through three essential factors: behavioral attitude, subjective norms, and perceived behavioral control [62]. Schwartz’s [47] Norm Activation Model (NAM) acknowledges the psychological factors driving pro-environmental behavioral intentions through the framework of moral norms [56,57]. It is mainly composed of three important variables: consequence awareness, responsibility attribution, and personal norms. On this basis, we propose a conceptual framework that integrates the TPB and NAM to explain farmers’ LCAT adoption levels from both self-value and moral perspectives, as shown in Figure 1. In TPB, behavioral attitude (manifested by cognition, judgment, and evaluation of LCAT), perceived behavioral control (manifested by self-capacity, available time, and disposable income), and subjective norms (manifested by social support and peer discussions) directly influence adoption levels. These TPB factors are driven by self-perceived value, which shapes farmers’ attitudes and perceptions of their resources. In NAM, responsibility attribution (manifested by a sense of responsibility for the outcomes of one’s actions) and consequence awareness (manifested by awareness that not acting altruistically causes negative outcomes) both contribute to the formation of personal norms. These personal norms, representing an internalized sense of moral responsibility, directly influence adoption levels. Furthermore, moral attributes serve as a bridge between TPB’s subjective norms and NAM’s personal norms. By internalizing social influences into moral obligations, moral attributes enhance the impact of both theories on farmers’ adoption behaviors.

2.3. Hypotheses

Behavioral attitude is the degree factor that guides an individual’s particular behavior, shaped by beliefs or assessments about the behavior, whether known or unknown [63]. If farmers perceive a certain behavior to have better development prospects, they are more likely to adopt it [64]. When considering LCAT adoption, farmers evaluate factors such as costs, benefits, and risks. They also compare these technologies with traditional farming methods to form objective views on their adoption. If farmers have a comprehensive understanding of LCAT and believe that its benefits, such as higher returns, lower costs, and reduced risks, outweigh its disadvantages, they are more inclined to adopt LCATs [65]. Based on the discussion above, we propose the hypothesis below:

H1: 

Farmers’ behavioral attitude has a positive effect on their adoption level of LCATs.

Subjective norms capture an individual’s belief regarding social expectations and to what degree they perceive others would affect their behavior in encouraging or discouraging a certain action [66]. Under such pressure, individuals are likely to align their behavior with societal expectations [67]. In the context of farmers’ LCAT adoption, subjective norms include the encouragement or resistance farmers encounter from neighbors, large growers, and cooperative organizations. When neighboring farmers in a village adopt LCATs, the adoption can be seen as becoming a social norm, leading other farmers to voluntarily adopt similar technologies [67]. However, Liu et al. (2020b) found that the neighborhood effect had no significant impact on rice farmers’ willingness to adopt low-carbon practices [36]. On the other hand, active promotion by village leaders or adoption by large growers can create a demonstration effect, encouraging many farmers to follow suit [68]. Based on the above analysis, we posit the variable as follows:

H2: 

Farmers’ subjective norms positively influence their adoption level of LCATs.

Perceived behavioral control refers to an individual’s perception of the ease or difficulty in performing a particular behavior [68]. According to Daxini et al. (2018), when individuals perceive that they have more resources and opportunities, and when they expect fewer obstacles, their perceived behavioral control regarding behavior will be stronger [69]. If farmers perceive they have sufficient time, energy, and economic capital to master the needed knowledge and acquire LCAT services, they will be likely to adopt such technologies [70]. Vaske et al. (2020) further noted that farmers are more inclined to recognize and adopt LCATs if they perceive economic benefits [71]. Based on this discussion, we propose the hypothesis below:

H3: 

Farmers’ perceived behavioral control positively influences the adoption level of LCATs.

Personal norms are internalized beliefs about moral duty that guide behavior to meet self-set ethical standards [71]. The norms often develop through the internalization of subjective norms based on their social environment and group influences [72]. Unlike subjective norms, personal norms stem from social pressure but focus on an individual’s cognition and judgment regarding what is considered morally right or wrong [73]. Farmers feel content and validated when they perceive their behaviors as consistent with their personal ethical standards [36]. Moreover, some farmers view safeguarding the environment as a moral obligation. When driven to better the ecosystem, beautify their community, and promote sustainability, they tend to implement at least one LCAT [51]. Based on this, we propose the following hypothesis:

H4: 

Farmers’ personal norms positively influence their adoption level of LCATs.

Subjective norms significantly influence individuals’ decisions by establishing the standards and values of specific actions [74]. In agriculture, farmers often base their choices on observing others or considering their opinions [74]. When faced with LCAT decisions, farmers internalize external social pressures and expectations. These gradually convert into personal behavioral norms through complex social psychological processes [75]. Zhao et al. (2022) argued that this internalization of social norms is a multifaceted process shaped by context-specific rules and reference groups [76]. On the one hand, social norms heighten farmers’ recognition of the necessary transition to low-carbon farming by promoting such practices across the agricultural industry and the broader community. On the other hand, when farmers perceive societal approval of LCATs, they become more aware of traditional methods’ negative impacts. This awareness cultivates a sense of personal responsibility to adopt low-carbon farming practices. As a result, it develops personal norms for LCATs [77]. Based on this analysis, hypothesis H5 is proposed as follows:

H5: 

Farmers’ subjective norms have a significant positive impact on the personal norms of LCATs.

Responsibility attribution involves individuals feeling accountable for their actions’ consequences [77]. This reflects farmers’ sense of personal obligation to mitigate the environmental impact of their practices. When farmers believe their actions produce positive outcomes and feel obligated to continue those actions, their personal norms become activated [76]. As key decision makers in rural areas [78], farmers often perceive themselves as environmental stewards. If farmers feel responsible for protecting and beautifying their countryside, they are likely to view LCAT adoption as a moral duty. This drives them toward adopting sustainable practices [79]. Based on this reasoning, hypothesis H6 is proposed as follows:

H6: 

Farmers’ responsibility attribution has a positive effect on their personal norms.

Consequence awareness is recognizing that failing to perform certain altruistic behaviors leads to negative outcomes. This pertains to farmers’ awareness of the environmental harm caused by not adopting LCATs. According to Zhang et al. (2021), farmers in China are more concerned about environmental issues that threaten their personal values [80]. This awareness shapes farmers’ perceptions of how their farming methods impact the environment [81]. Environmentally conscious farmers are more inclined to accept their responsibility to save the environment, especially in terms of mitigating climate change. They also understand the adverse ecological consequences that arise from not utilizing LCAT [49]. Thus, hypothesis H7 is proposed as follows:

H7: 

Farmers’ consequence awareness has a positive effect on their responsibility attribution.

Studies reveal that recognizing the consequence awareness of resource depletion and environmental harm can indirectly or directly engage personal norms, such as saving water, sustainable agricultural land conservation practices, and green agricultural production behavior [48,82,83]. As the agroecological environment deteriorates and society increasingly advocates for green and sustainable development, responsible agricultural operators are evolving from “rational individuals” to “social individuals” [84]. This analysis suggests that the more pronounced farmers’ perceived negative impacts are of traditional agricultural technologies on ecology, resources, quality, and sustainable development, the stronger the sense of responsibility among farmers, increasing the likelihood of personal norm formation. Once engaged, these norms trigger moral responsibilities, ethical values, and a sense of guilt, which in turn enhance farmers’ readiness to adopt LCATs. Following the analysis above, hypothesis H8 is introduced:

H8: 

Farmers’ consequence awareness has a positive effect on their personal norms.

3. Data and Methods

3.1. Survey Design

The survey questionnaire collected information in three parts. The first part gathered data on farmer households, including individual characteristics such as gender (0 for male, 1 for female), age (in years), education level (years), and physical condition (classified as good, normal, or worse). It also gathered information on family characteristics, including the number of household members capable of productive work, farming scale (≤0.67 hm², 0.67–2 hm², or ≥2 hm²) [85], number of fields, and annual household income (measured in CNY 10,000). Part 2 was about farmers’ LCAT cognitions within a combined model of TPB and NAM, including the perception of technology, the impact of technical training, and the assessment of economic and ecological benefits associated with LCAT. We utilized 19 observed variables to represent six latent variables, including three specific indicators for behavioral attitude derived from Wang et al. (2018) [59], four specific indicators for subjective norms based on Guo et al. (2022) [86], three specific indicators for perceived behavioral control informed by Sui et al. (2023) [87], three specific indicators for personal norms based on Xiong et al. (2020) [88], three specific indicators for responsibility attribution following Zou et al. (2023) [89], and three specific indicators for consequence awareness referencing Chen et al. (2024) [49]. In Part 3, we evaluated farmers’ actual uptake of LCAT. The question, “Among the following five low-carbon agricultural technology systems, how many specific technologies have you adopted?” (the value range is 0–16) measures the level of farmers’ adoption behavior for LCAT. A five-point Likert scale was used to measure the farmers’ adoption level of LCAT as informed by Waiswa et al. (2025) and Zhao et al. (2022) [18,76], with values ranging from 1 to 5, representing strongly disagree, disagree, neutral, agree, and strongly agree. The measurement variables and specific questions are shown in

Table 2. Variables and questionnaire questions.

Latent Variables	Observed Variables	Questionnaire Questions
Behavioral Attitude	BA1	The benefits of adopting low-carbon agricultural technologies outweigh the potential risks.
	BA2	Low-carbon agricultural technologies can increase both crop yields and household incomes.
	BA3	The cost of adopting low-carbon agricultural technologies is relatively low.
Subjective Norms	SN1	Many farmers in my community have adopted low-carbon farming techniques.
	SN2	Neighbors frequently discuss low-carbon agricultural technologies with each other.
	SN3	Village officials provide strong technical support for low-carbon agriculture.
	SN4	My family supports my adoption of low-carbon agricultural technologies.
Perceived Behavioral Control	PBC1	I can quickly learn and apply low-carbon farming techniques.
	PBC2	I have sufficient time to learn and master low-carbon agricultural technologies.
	PBC3	I have enough financial resources to support my learning and mastery of low-carbon agricultural technologies.
Personal Norms	PN1	I feel a moral obligation to use low-carbon farming techniques.
	PN2	Practicing low-carbon agriculture aligns with my core values.
	PN3	I would feel guilty if I chose not to adopt low-carbon farming techniques.
Responsibility Attribution	RA1	I am responsible for the environmental pollution caused by my agricultural production.
	RA2	I take responsibility for the damage caused to the soil during agricultural production.
	RA3	I am accountable for the water wastage in my agricultural production process.
Consequence Awareness	CA1	The failure to adopt low-carbon agricultural technologies hampers sustainable agricultural development.
	CA2	Not adopting low-carbon agricultural technologies conflicts with one’s ethical principles.
	CA3	The failure to practice low-carbon agricultural technologies is against the family’s wishes.
Adoption Level	AB	How many low-carbon agricultural technologies have been adopted? (0–16)

.

3.2. Data Collection

Farmer surveys were conducted to clarify adoption behavior and its psychological factors in a rural area of Wuxi, Jiangsu, from July to September 2023. Located in the thriving Yangtze Delta region of East China, Wuxi benefits from abundant rainfall and favorable agricultural conditions. Supported by national policies, initiatives such as the Carbon Incentive Scheme (CIS), public low-carbon scenarios, and carbon emission reduction programs are widely implemented [90]. Wuxi has become one of the first agricultural modernization demonstration zones in southern Jiangsu and a key area for advancing low-carbon agriculture in China, making it a representative suitable study area [91]. Xishan, Xinwu, and Yixing were selected as sample locations because of their agricultural prominence, diversity in farming practices, and varying levels of LCAT promotion. We randomly chose 1–2 townships from each county and 2–4 villages from each township, with 10–15 households per village. In total, 360 farmers were sampled and visited from 8 townships within the Wuxi area through face-to-face interviews, with an effective rate of 92.5% (360/389).

3.3. Data Analysis

A structural equation model (SEM) was employed to carry out hypothesis tests. SEM is a statistical method that analyzes the relationship between variables based on their covariance matrix [92]. Additionally, a separate regression-based mediation analysis was conducted with the PROCESS tool [93]. In comparison to the conventional regression approaches, SEM excels in addressing complex interactions among numerous latent and observed variables. This study considers the LCAT adoption level by farmers (i = 0–16) as a continuous indicator, that is, the dependent variable can be treated as continuous. The design of each dimension in the questionnaire adopts the Likert five-point scale method [94]. SEM includes 6 latent variables and 19 observed variables, which are usually expressed as follows:

$$X={\lambda }_X\xi +\delta$$

(1)

$$Y={\lambda }_Y\eta +\epsilon$$

(2)

where Equations (1) and (2) are measurement equations to describe the connections between latent variables, observed variables, and farmers’ LCAT adoption level. λ_X and λ_Y represent the factor loadings for the latent variables $X$ and $Y$. $\delta$ and $\epsilon$ denote the error terms associated with the observed variables.

$$\eta =\gamma \xi +\beta \eta +\zeta$$

(3)

where Equation (3) is a structural equation to describe the relationship between six exogenous latent variables, endogenous latent variables, and farmers’ LCAT adoption level. The path coefficients (denoted as $\gamma$ and $\beta$) indicate the strength of these relationships. $\zeta$ is the random error term.

4. Results

4.1. Sample Characteristics

Table 3. Basic characteristics of surveyed farmers.

Variable	Definition	Number	Percentage (%)	Variable	Definition	Number	Percentage (%)
Gender	Male	251	69.7%	Family Number	≤3	66	18.3%
	Female	109	30.3%		4–6	250	69.4%
Age (Years)	30–40	29	8.1%		≥7	44	12.2%
	41–50	50	13.9%	Farming Scale (hm2)	≤0.67	244	67.8%
	51–60	151	41.9%		0.67–2	105	29.2%
	61–70	91	25.3%		≥2	11	3.1%
	≥71	39	10.8%	Plot Number	1–2	114	31.7%
Physical Condition	Good	305	84.7%		3–4	234	65.0%
	Normal	50	13.9%		≥5	12	3.3%
	Poor	5	1.4%	Annual Household Income (CNY 10,000)	0–10	137	38.1%
Education Level (Years)	Primary School and Below	168	45.8%		10–20	177	49.2%
	Middle School	96	26.7%		20–30	41	11.4%
	High School	49	13.6%		≥30	5	1.4%
	College or Above	47	13.1%

shows the descriptive statistics of the interviewed farmers and their family characteristics. The majority of respondents were male (69.7%, 251 individuals). This demographic distribution reflects the prevalent pattern where men predominantly guide farming decisions, a feature often documented in rural agricultural studies. The study’s participants were predominantly middle-aged or older, with a notable 81.1% of farmers aged 41 or above, reflecting the aging demographic within the agricultural workforce. The majority of farmers (84.7%) reported excellent physical fitness. This could imply that farmers are largely in good health, which is crucial for the effective management of farming activities. In terms of education, many interviewed farmers had low to moderate education levels. This limitation might impede their ability to adopt innovative technologies or understand contemporary agricultural techniques; 45.8% of farmers completed only primary school, while a smaller portion (13.1%) pursued higher education, with a college degree or beyond.

The survey also indicates that 69.4% of households (260 families) had four to six members, indicating that most farmers have small to medium-sized households. This may affect the availability of labor for farm activities. Notably, 67.8% of these farmers manage small plots (≤0.67 hm²) and continue to engage in small-scale farming operations. Family-oriented farming on a modest scale was widespread, with 65% of farmers owning three to four plots of land. This may indicate that land holdings are relatively concentrated and less fragmented, which could lead to more efficient operations and potentially higher productivity. Additionally, the data show that 49.2% of farmers have an annual household income between CNY 10,000 and 20,000. The proportion of farmers with incomes above CNY 20,000 is relatively small. This emphasizes the idea that the interview group was smallholder farmers with limited financial resources [45].

4.2. Farmers’ LCAT Adoption Status

Table 4. Farmers’ adoption of different LCAT systems.

LCAT System	Number	Percentage (%)
Reduced tillage system	161	44.72%
4R fertilizing system	95	26.39%
Eco-friendly pesticide application system	52	14.44%
Agricultural film system	33	9.17%
Straw resource utilization system	28	7.78%

presents surveyed farmers’ adoption status for different LCAT systems. Specifically, 63.06% of farmers adopted at least one LCAT system. Among them, the reduced tillage system was the most widely adopted technology (44.72% adoption rate). The reduced tillage system may be considered the most accessible and beneficial option for farmers, as it supports soil health and reduces carbon emissions. In addition, the 4R fertilizing system followed, with an adoption rate of 26.39%. This system reduces carbon emissions and enhances fertilizer effectiveness through the precise application of fertilizer, including selecting the appropriate fertilizer, controlling its quantity, and optimizing the timing and location of its application. The eco-friendly pesticide application system was adopted by 14.44% of farmers, which reflects their awareness of the need to reduce the environmental impact of chemicals. However, there is still potential for broader adoption of eco-friendly pesticide application system. The agricultural film system and the straw resource utilization system were adopted by 9.17% and 7.78% of farmers, respectively. These systems aim to reduce carbon emissions from plastic waste (primarily from agricultural films) and increase soil organic matter on farmland.

Table 5. Farmers’ binary adoption status and number of adopted LCATs.

	LCAT Adoption Status	Number	Percentage (%)
Binary adoption	Unadopted LCAT	133	36.94%
	Adopted LCAT	227	63.06%
Number adopted LCATs	1 LCAT Adopted	137	38.06%
	2 LCAT Adopted	16	4.44%
	3 LCAT Adopted	14	3.89%
	4 LCAT Adopted	20	5.56%
	5 LCAT Adopted	18	5.00%
	6 LCAT Adopted	22	6.11%

presents the quantitative characteristics of LCAT technology adoption among the surveyed farmers, detailing their binary adoption status and the distribution of specific adoption frequencies. The data indicate that out of 360 respondents, 63.06% (227 farmers) adopted at least one LCAT technology, while 36.94% (133 farmers) remain non-adopters. Within the adopting group, the number of technologies adopted exhibited significant heterogeneity. Most farmers (38.06%) adopted only one LCAT technology. Smaller groups adopted multiple LCATs: 4.44% used two, 3.89% used three, 5.56% used four, 5.00% used five, and 6.11% used six. From Table 5, it can be seen that the majority of adopters engaged with only a single LCAT, but a notable minority integrated multiple LCATs into their agricultural practices. No farmers adopted seven or more low-carbon agricultural technologies. This finding highlights an opportunity to promote the broader adoption of the full spectrum of low-carbon technologies through targeted education and incentives [33].

4.3. Measurement Model

To examine the validity and reliability of the latent variables, a confirmatory factor analysis was carried out. The results are presented in

Table 6. Reliability and validity test results.

Latent Variables	Observed Variables	Std. Estimate	Construct Reliability (CR)	Average Variance Extracted (AVE)
Behavioral Attitude	BA1	0.94	0.95	0.87
	BA2	0.91
	BA3	0.91
Subjective Norms	SN1	0.93	0.95	0.84
	SN2	0.91
	SN3	0.91
	SN4	0.92
Perceived Behavioral Control	PBC1	0.95	0.96	0.88
	PBC2	0.94
	PBC3	0.92
Personal Norms	PN1	0.93	0.95	0.86
	PN2	0.93
	PN3	0.92
Responsibility Attribution	RA1	0.94	0.95	0.86
	RA2	0.93
	RA3	0.92
Consequence Awareness	CA1	0.91	0.94	0.84
	CA2	0.90
	CA3	0.94

. The reliability coefficients for each dimension and the overall scale were greater than 0.9, indicating exceptional consistency across individual dimensions and the entire measure [94]. The Kaiser–Meyer–Olkin (KMO) value for the overall scale reached 0.93, and Bartlett’s test of sphericity produced statistically significant results, affirming the data’s appropriateness for factor analysis. The construct reliability (CR) values for the six latent variables exceeded 0.7, reflecting strong internal consistency [95]. This indicates that the questionnaire items reliably captured the latent variables. As a result, the questionnaire proved highly reliable, and the data collected were considered trustworthy. Regarding validity, tests showed that the standardized factor loadings of the 19 observed variables ranged from 0.90 to 0.95, all exceeding the minimum threshold of 0.6 [94]. In addition, the average variation extracted (AVE) for each of the six latent variables was above 0.5 [96]. This suggests that the latent variables effectively explained the observed variables. Collectively, these findings demonstrate that the survey data are of good convergent validity and meet the required validity requirements.

4.4. Structural Model

4.4.1. Model Fitness

After the measurement model was established, a full structural model was subsequently developed to test our hypotheses. To assess the model fit, a comprehensive set of fit indices was employed, encompassing absolute fit, incremental fit, and parsimonious fit indicators. Per

Table 7. SEM goodness-of-fit indices.

Fitness Indicators	Acceptable Fit Values	Post Modification	Result
Absolute Fitness Indicator
CMIN/DF	<3 Ideal, <5 Acceptable	2.36	Accept
RMSEA	<0.05 Ideal, <0.08 Acceptable	0.06	Accept
Value-Added Fitness Indicators
NFI	>0.9	0.96	Accept
IFI	>0.9	0.97	Accept
CFI	>0.9	0.97	Accept
TLI	>0.9	0.97	Accept
Concise Fitness Indicators
PNFI	>0.5	0.81	Accept
PCFI	>0.5	0.82	Accept

, the CMIN/DF value in the absolute fit indicator was 2.36, which falls within the ideal range of less than 3. The RMSEA value was 0.06, indicating a satisfactory fit (less than 0.08) [95]. Among the selected incremental fit indicators (NFI, IFI, CFI, TLI), the values ranged from 0.96 to 0.97, all exceeding the recommended threshold of 0.9 [95]. The values for the parsimonious fit indicators (PNFI, PCFI) were greater than 0.8, surpassing the critical value of 0.5, which suggests that the data from the field questionnaire survey in this study fit the model well [96].

4.4.2. Direct Effect and Hypotheses Test

Table 8. SEM direct estimation and hypothesis.

Type	Hypothesis	Path	Coefficient	S.E.	C.R.	Test Result
Direct effect	H1	Behavioral Attitude → Adoption Level	0.23 ***	0.02	3.95	Acceptable
	H2	Subjective Norms → Adoption Level	0.31 ***	0.02	5.32	Acceptable
	H3	Perceived Behavioral Control → Adoption Level	0.21 ***	0.02	3.60	Acceptable
	H4	Personal Norms → Adoption Level	0.26 ***	0.02	5.05	Acceptable
	H5	Subjective Norms → Personal Norms	0.24 ***	0.06	4.00	Acceptable
	H6	Responsibility Attribution → Personal Norms	0.16 ***	0.06	2.65	Acceptable
	H7	Consequence Awareness → Responsibility Attribution	0.81 ***	0.04	18.76	Acceptable
	H8	Consequence Awareness → Personal Norms	0.50 ***	0.09	5.96	Acceptable

Note. The significance level is set at p < 0.01, denoted by ***; the significance level is set at p < 0.05, denoted by **; the significance level is set at p < 0.1, denoted by *.

and Figure 2 together present the results of the structural model’s direct estimation, along with the hypothesis and path analysis diagram.

Behavioral attitude was one of the primary factors influencing farmers’ attitudes toward the adoption level of LCATs (Figure 2). Specifically, farmers’ cognition, judgment, and evaluation of LCATs significantly impacted their adoption decisions. The path coefficient for behavioral attitude was 0.23, significant at the 1% level (Table 8). This confirms Hypothesis 1 and indicates that farmers with more favorable attitudes toward LCATs are more likely to adopt them, although this effect is relatively moderate. As farmers gain more knowledge of LCATs, they become more likely to value their benefits in farming, such as increased productivity, reduced costs, and environmental improvements. These recognized benefits can further increase understanding and enhance their trust and support for LCATs and a positive attitude towards its adoption and consequent incorporation into farming practices. These findings align with research by Hou et al. (2015) and Sarkar et al. (2022) [97,98]. Regarding benefit and risk perceptions, farmers who prioritize economic gains tend to adopt LCATs with lower costs and risks compared to conventional farming practices, and their attitudes influence their perceptions of the potential outcomes.

Subjective norms had the greatest effect on farmers’ adoption level toward LCATs, ranking higher than behavioral attitude (Figure 2). The path coefficient for subjective norms was 0.31, significant at the 1% level, confirming Hypothesis 2 (Table 8). This result corresponds with the findings of Sarkar et al. (2022) and Gowda et al. (2021) [98,99]. This can be attributed to the close-knit nature of social relations in Chinese rural communities. Farmers are highly influenced by social support and peer interactions toward adopting cost-effective and efficient farming practices. Specifically, the factor loadings for neighbors, village cadres, and family members were 0.91, 0.91, and 0.92, respectively. This indicates that the influence of farmers’ family behavior is greater than that of neighbors and village officials. Family members exert the most powerful influence. This suggests that family advice, opinions, and expectations critically shape farm household actions. Neighborhood dynamics and village leaders (e.g., local officials and government organizations) also significantly affect farmers’ LCAT adoption decisions through the provision of guidance, information, and assistance. Farmers near others are more exposed to demonstration effects. This proximity increases their likelihood of recognizing the practical advantages of LCATs. Moreover, farmers often emulate the practices of their peers. Village leaders, wielding notable authority in rural settings, play a vital role in promoting LCAT awareness and uptake. These findings are consistent with Doran et al. (2020) and Garmendia-Lemuset et al. (2024) [100,101].

Perceived behavioral control significantly affected farmers’ adoption level toward LCATs. However, it had a relatively lower impact compared to behavioral attitude (Figure 2). Perceived behavioral control showed a positive path coefficient of 0.21 (p < 0.01), validating Hypothesis 3 (Table 8). This indicates that farmers’ perceived control in adopting LCATs has a significant positive impact on their adoption behavior, matching the findings by Yu et al. (2023) [28]. Farmers confident in rapidly mastering agricultural technologies, with sufficient time and finances, are more likely to accept LCATs. Specifically, factor loadings for self-capacity, available time, and disposable income were 0.95, 0.94, and 0.92, respectively. This suggests that self-capacity outweighs the available time and financial resources in driving behavior. Farmers’ confidence in adopting LCATs is key to integrating these technologies into farming practices. This confidence directly strengthens their perceived behavioral control, which reflects how capable individuals feel about performing a specific action. When farmers believe they can successfully use LCATs, they are far more likely to take the leap and implement them [102]. Additionally, perceived behavioral control reflects farmers’ evaluation of the resources essential for effectively implementing LCATs. Farmers who see the benefits of using LCATs show increased readiness to allocate time and monetary resources to tackle obstacles hindering adoption [30].

Personal norms ranked second in influencing farmers’ LCAT adoption level (Figure 2), following subjective norms, with a path coefficient of 0.26, and passed the significance test at the 1% level (Table 8). Therefore, Hypothesis 4 was supported. The decision to adopt LCATs can be strongly influenced by farmers’ views on whether it is an ethical behavior. When farmers see LCAT use as matching their moral values, they view it as a social duty [103]. Higher personal norms will motivate farmers to feel pride and satisfaction in the adoption behavior of LCATs and to recognize the act as a right moral decision. In contrast, farmers who do not adopt LCATs will internally perceive themselves as violating their moral obligations and feel very guilty [104]. This will prompt farmers to reflect and decide to change their behavior. This echoes the findings of Shi et al. (2020) on the drivers of sustainable production [105].

Subjective norms moderately shaped personal norms (path coefficient: 0.24, p < 0.01) (see Figure 2 and Table 8). Hypothesis 5 reflects the idea that standards of behavior shared within a society shape a farmer’s internal sense of responsibility regarding LCAT adoption. This also implies that social expectations and pressures influence both LCAT acceptance level and the strength of personal values and obligations. This is supported by Niu et al. (2020) and Zhang et al. (2023) [106,107]. When farmers perceive LCAT adoption as socially endorsed, they notice the downsides of traditional methods more readily. This fosters a personal duty to pursue sustainable practices and reinforces personal norms supporting low-carbon farming [104].

Responsibility attribution affected farmers’ personal norms (see Figure 2) (path coefficient: 0.16, p < 0.01), verifying H6 (Table 8). As farmers develop a stronger sense of environmental duty, their intrinsic moral commitment to environmentally friendly practices will also increase. Farmers who attribute ecological damage (e.g., global warming, soil degradation) to their own activities are more likely to internalize a belief in environmental responsibility and feel a personal obligation to adopt low-carbon farming practices. This is similar to the findings of Xie et al. (2021) on eco-friendly behavior [81].

Hypothesis H7 examined the relationship between consequence awareness and responsibility attribution. Farmers’ awareness of environmental harm from their actions, alongside LCAT benefits, affects how they attribute responsibility (Figure 2). The coefficient of 0.81 implied that consequence awareness is a strong driver of responsibility attribution (Table 8). Farmers recognizing their ecological impact are more likely to accept responsibility for reducing harm. This is a critical finding because those committed to environmental protection see themselves as agents of responsible farming [102]. Awareness of issues like carbon emissions and soil pollution sparks feelings of remorse and guilt among farmers. This awareness, in turn, awakens their moral duty to safeguard the environment [103,104]. Consequently, they tend to accept accountability for these ecological outcomes. This acceptance motivates them to pursue sustainable farming practices. Such findings reinforce the observations of Ricart et al. (2025) [108].

Consequence awareness positively and significantly influenced farmers’ personal norms (path coefficient: 0.50, p < 0.01) (see Figure 2), verifying H8 (Table 8). This also means that farmers’ consequence awareness of traditional agricultural practices can activate personal norms both directly and indirectly, through responsibility attribution. Positive beliefs and perceptions of farmers on environmental issues can enhance farmers’ internal motivation to adopt LCATs. The positive direction means that farmers with a clearer grasp of traditional methods’ negative effects feel a stronger obligation to adopt LCATs that meet societal norms. This awareness of consequences will activate ethical standards, reshaping behaviors and values through internalized principles. This is similar to the research of Guo et al. (2019) and Badsar et al. (2023), who showed that farmers’ environmental beliefs can promote their environmental behaviors [109,110].

4.4.3. Indirect Effect Result Analysis

Table 9. Indirect estimation.

Type	Path	Coefficient	S.E.	C.R.	Mediation Type
Indirect effect	Subjective Norms → Personal Norms → Adoption Level	0.06 **	0.02	2.52	Partial mediation
	Consequence Awareness → Personal Norms → Adoption Level	0.13 ***	0.03	2.65	Partial mediation
	Responsibility Attribution → Personal Norms → Adoption Level	0.04 **	0.02	2.03	Partial mediation
	Consequence Awareness → Responsibility Attribution → Personal Norms	0.13 ***	0.05	4.68	Partial mediation
	Consequence Awareness → Responsibility Attribution → Personal Norms → Adoption Level	0.03 ***	0.02	3.20	Chain mediation

Note. The significance level is set at p < 0.01, denoted by ***; the significance level is set at p < 0.05, denoted by **; the significance level is set at p < 0.1, denoted by *.

lists five distinct indirect paths among various mediating variables.

Interestingly, our indirect effect analysis revealed that farmers’ internal and external psychological factors exert multiple serial mediation effects on LCAT adoption (see Table 9). Personal norms emerged as the primary mediator that connects subjective norms, consequence awareness, and responsibility attribution to the adoption level. This result indicates that the internalization of norms is critical for translating external influences into behavioral outcomes. The analysis also revealed that the three sets of pathways showed significant indirect effects of 0.06, 0.13, and 0.04, respectively (Table 9). Specifically, subjective norms, as a form of social influence, affect behavior indirectly via the mediation of personal norms. By incorporating societal expectations (via peers, community leaders, or agricultural networks) into their own moral codes, individuals are predisposed to adopt behaviors that align with established social norms. Once personal norms are established, they directly influence adoption behavior by creating a sense of moral duty. According to the NAM, personal norms reflecting a sense of moral duty are key drivers of pro-environmental actions. Consequence awareness, as a cognitive-level antecedent factor, affects adoption behavior by triggering personal norms. Awareness of potential outcomes strengthens individuals’ moral obligations, thereby encouraging corresponding actions. Responsibility attribution, as an antecedent factor at the moral level, affects behavior by intensifying personal norms. The perception of personal responsibility for an outcome activates individuals’ internalized moral standards, reinforcing their commitment to ethical conduct. This finding is consistent with observations in research on other environmentally conscious practices, such as ecological breeding and eco-friendly product consumption [50,111].

Responsibility attribution played a dual mediating role in this path model. First, it connected consequence awareness to personal norms, indicating that it functions as a crucial mechanism in the transition from cognition to moral commitment (indirect effect = 0.13, p < 0.01) (Table 9). Second, responsibility attribution indirectly influenced the adoption level through personal norms, underscoring that it acts as a fundamental antecedent variable in morally guided behavior. This finding emphasizes responsibility attribution in the Norm Activation Theory and matches conclusions from Westerink et al. (2021) [111,112]. Such studies highlight the decisive role that responsibility attribution plays in shaping behavior.

Further analyses revealed the existence of a chain mediating path: consequence awareness → responsibility attribution → personal norms → adoption level (indirect effect = 0.03, p < 0.01), as presented in Table 9. This sequence illustrates the mental process driving LCAT behavioral adoption. Responsibility attribution acts as a link between cognitive awareness and moral considerations, translating consequence awareness into personal norms. Meanwhile, these personal norms serve as intrinsic motivators, translating responsibility attribution into concrete adoption actions. This step-by-step mediation process underscores an ordered connection between responsibility attribution and personal norms in behavioral decision making, confirming the altruistic dispositions of local farmers. Similar observations are reported by Huo et al. (2025) [113]. This study provides a fresh viewpoint on LCAT adoption by incorporating these elements into a dual-path model and analyzing their role as stimuli within an innovative environmental framework.

5. Discussion

Norms play a crucial role in shaping farmers’ decision behavior to adopt new agricultural technologies, influencing their behavior through both social expectations and personal moral frameworks, which reflects the combined use of the TPB and NAM. Specifically, subjective norms are the dominant driver in influencing farmers’ adoption behavior in the TPB, as shown by their higher path coefficients compared to other factors. This suggests that social influences such as community trust, peer pressure, and social networks are crucial factors that drive farmers’ LCAT adoption. In collectivistic rural communities, farmers are closely connected to social networks [102]. Neighborhood effects, peer demonstrations, and the influence of local agricultural leaders play a critical role in amplifying the impact of social norms, through either advocacy or hands-on demonstrations, helping to integrate LCAT adoption into social practices [103]. When these social bonds encourage farmers to adopt related technologies as a desirable practice, they are more likely to comply in order to maintain social coherence, gain social acceptance, and build social capital through their personal behavior. This is consistent with other studies that utilize the TPB framework [38,62,63].

Furthermore, personal norms represent the moral obligation driving LCAT adoption, shaped by personal ethics and societal expectations. As the second strongest driver, personal norms act as a mediator linking subjective norms, consequence awareness, and responsibility attribution to adoption level. Subjective norms shape individual beliefs, attitudes, and behaviors, emphasizing the role of external pressures. Even without fully understanding agricultural extension practices, farmers adopt behaviors by observing role models or influential figures. Meanwhile, they internalize community norms as personal norms to avoid criticism and promote these behaviors independently. This finding has been acknowledged in previous studies on farming behaviors and other environmental farming behaviors [48,49,50,51].

Additionally, moral emotions like guilt and empathy, along with perceived responsibilities such as environmental stewardship, motivate behavior beyond immediate economic benefits. This reflects altruistic values rooted in social and personal beliefs. For this reason, the level of farmers’ adoption can be further enhanced by incorporating social influence, moral responsibility and awareness-related activities. This is also supported by the study conducted by Waiswa et al. (2025) [18].

This finding further supports the idea that farmers prioritize moral attributes over self-perceived values, meaning that ethics-conscious farmers are driven to choose altruism over self-interest. This is evident from the higher path coefficients for subjective norms and personal norms compared to perceived behavioral control and behavioral attitude. Unlike some prior studies, this suggests that both social and personal norms can offer stronger motivations for behavior adoption, likely due to their ties to social norms and rural society structures [71,112]. First, the pressure farmers feel from community advocacy and expectations is an important aspect to consider. When rural organizations establish sustainable agricultural development as the accepted norm for farming practices, farmers prioritize collective well-being and long-term sustainability over immediate economic benefits to avoid criticism or exclusion [114]. This view is consistent with the concept of environmental citizenship [115]. Additionally, many sustainable practices lack immediate economic benefits, reducing their appeal as based on self-interest [116]. If farmers do not see significant short-term economic returns, their behavioral attitudes may not strongly encourage the adoption of these practices. However, subjective norms and personal norms can offer stronger incentives to act in ways that benefit the collective good or align with individual perceived values.

Moreover, consequence awareness and responsibility attribution are the critical and foundational steps in the moral activation of normative mechanisms that drive the development of personal norms, emphasizing NAM’s focus on activating moral attributes. Consequence awareness, which reflects farmers’ understanding of ecological conservation, has the dominant influence on both responsibility attribution and personal norms. On one hand, responsibility attribution is the catalyst for consequence awareness. This is a dual cognitive psychological process: the first question is whether farmers have a personal responsibility to mitigate any negative consequences, and the second is whether they have a moral obligation to take action [58,59]. Farmers who recognize environmental consequences and attribute responsibility will activate their personal norms, psychologically motivating themselves to proactively adopt new sustainable farming practices. However, the effectiveness of responsibility attribution is contingent on the depth of consequence awareness. Although responsibility attribution is important, it has limited influence on personal norms if farmers’ understanding of ecological impacts is superficial [35,66]. Otherwise, a weak sense of consequences will hinder adoption behavior. On the other hand, moral emotions triggered by consequence awareness directly reinforce personal norms [80]. Consequence awareness involves not only rational cognition but also the reinforcement of emotions, such as guilt or empathy [117]. These emotions convert personal norms into behavioral motivations that rational awareness alone cannot produce [80,81]. Therefore, environmental education covers both elements. It is suggested to develop educational programs that integrate factual knowledge of environmental effects with strategies to evoke emotional responses, such as empathy or guilt. This approach connects cognitive understanding with affective engagement, thereby strengthening personal norms and motivating farmers’ ethical dedication to sustainable practices [73,83]. Future research might usefully examine these emotional aspects within social psychology, particularly as linked to altruistic behavior.

6. Conclusions and Implications

The “dual carbon” goal of achieving “carbon peak” in 2030 and “carbon neutrality” in 2060 was proposed by China in September 2020. Reducing agricultural production emissions is central to this strategy. It is also essential for building a robust agricultural sector and achieving green, low-carbon progress [61]. While previous research has primarily focused on the demographic, socioeconomic, and environmental factors influencing LCAT adoption, social-psychological drivers have received comparatively less attention. To address this gap, our study contributes to understanding the social-psychological factors behind farmers’ LCAT adoption in China. Specifically, we constructed a structural equation model (SEM) using the TPB and NAM to empirically analyze the direct and indirect effects of moral attributes and self-values on adoption behavior, utilizing data from 360 farmers in Wuxi. The main conclusions are as follows: farmers’ LCAT adoption behavior was significantly and directly affected by behavioral attitude (manifested by understanding, perceived benefits, and perceived risks), subjective norms (manifested by neighbors’ experience and community supports), perceived behavioral control (manifested by time, resources, and competences to learn about LCATs), and personal norms (manifested by ethical obligations and environmental responsibility). Among these, first, subjective norms had the greatest direct impact and indirectly influenced adoption behavior by shaping personal norms, indicating that social pressure is not only an external driver but also capable of being internalized as a personal moral obligation. Secondly, personal norms were second only to subjective norms in terms of direct influence and served as a key mediator in the relationship between subjective norms, consequence awareness (manifested by the understanding of the outcome), and responsibility attribution (manifested by the allocation of responsibility for the consequences) and adoption level. This further underscores that social pressure (via subjective norms) and moral commitment (via personal norms) are more effective than financial incentives when immediate financial rewards for sustainable agricultural farming practices are absent. Further, consequence awareness is the most critical driver in the farmers’ LCAT adoption behavior model, as it significantly shapes the formation of responsibility attribution and personal norms. The chain mediating path “consequence awareness → responsibility attribution → personal norms → adoption level” serves as the foundation for understanding the psychological mechanism underlying farmers’ adoption behavior. Additionally, responsibility attribution serves a dual role as a mediator between consequence awareness and personal norms, though its direct influence is limited. It is difficult to simply develop a sense of environmental responsibility attributed to farmers that translates strongly into personal norms or actions. A combination of increased awareness of consequences and social influence is needed to have more impact.

The results of this study offer valuable insights for promoting low-carbon agricultural technologies and guiding policy formulation. Firstly, a primary strategy involves enhancing social norms through promotion or exemplary practices. Given the importance of subjective norms in influencing farmers’ behavior, it is essential to leverage the influence of community figures (such as neighbors, relatives, and local leaders) to support LCAT adoption. Suggested measures include launching peer-led outreach initiatives, scheduling visits to LCAT demonstration sites, and encouraging seasoned farmers and agricultural experts to model best practices. These steps can strengthen social influence and expand LCAT implementation. Second, educating farmers to align ethical principles with environmental stewardship is vital. Recognizing the role of personal norms in shaping environmental awareness and decision making, as well as ethical rural education and ecological management programs, can instill lasting ecological values. Farmers’ sense of duty and ownership can be reinforced through partnerships with agro-industrial groups, cooperatives, training workshops, and informational materials. Third, targeted campaigns should heighten awareness of the environmental impacts of farming practices. Since consequence awareness strongly affects responsibility attribution and personal norms, on-site events and media efforts emphasizing ecological outcomes are recommended. Such initiatives can clarify the long-term advantages of sustainable methods. For example, government agencies and academic institutes might highlight LCAT’s economic and ecological benefits (e.g., cost savings, soil resilience) over conventional approaches while weighing climate change risks against LCAT adoption gains. Finally, tackling responsibility attribution requires offering policy incentives and technical support to lower adoption hurdles. To overcome its limited effect, efforts should emphasize environmental impact awareness and social reinforcement. Interventions like LCAT-focused subsidies, paired with training-integrated policy messaging, can address both individual accountability and collective social norms, fostering behavioral shifts.

This study has several limitations that should be disclosed. First, reliance on self-reported data may introduce biases, such as social desirability or recall errors, which could affect the accuracy of the adoption behavior reported by farmers. Future research should tackle this by incorporating objective measures, cross-validating data, and tracking behavior over time. Second, due to time and financial resources constraints, only 360 valid questionnaires were included in this study. Given the limited sample size of this small-scale study, caution is necessary when interpreting the psychological mechanism behind farmers’ adoption behavior, as the study may not be fully representative. Therefore, the findings are preliminary and require further validation through research in Jiangsu or other plains regions of China. In addition, Duong and Onwezen et al. have demonstrated that emotional factors, such as emotional state, threat perception, and stress awareness, play a significant role in shaping behavior [79,118]. Further investigation of these factors is essential, particularly within the established frameworks of the TPB and NAM. However, the current study does not address these emotional dimensions. Future research should enhance the TPB and NAM models by incorporating these emotional variables. In particular, special attention should be given to measuring farmers’ mental states to gain a deeper understanding of the psychological drivers behind LCAT adoption behavior. Moreover, given the significant role of subjective norms, future research should use social network analysis to explore how social ties influence farmers’ LCAT adoption and identify key influencers for targeted outreach. Furthermore, while our study emphasizes social-psychological factors, external barriers such as high costs, limited awareness, and inadequate support (e.g., training, subsidies) may hinder multi-technology adoption and require further investigation. Finally, the current study offers a cross-sectional view of farmers’ adoption behaviors, limiting insights into how social-psychological factors evolve over time. Future research should adopt a longitudinal approach, tracking farmers’ attitudes, norms, and adoption behaviors over an extended period. This would reveal the sustainability of LCAT adoption and the long-term impact of factors like subjective norms and personal norms, providing a dynamic understanding of behavioral change.