Behavioral and innovation drivers of farmers’ support for forest policy at the forest agriculture interface

Abstract

Sustainable forest management in agro-ecological landscapes requires policies that are both innovative and socially inclusive. This study investigates the behavioral and perceptual drivers of farmers’ support for participatory forest governance at the forest–agriculture interface of the Zagros region, western Iran. By integrating the Theory of Planned Behavior (TPB) and the Diffusion of Innovation (DOI) framework, the research examines how farmers’ perceptions of policy innovation properties influence their attitudes, subjective norms, perceived behavioral control, and intentions to support forest policy. Data were gathered from 385 rural households using a structured questionnaire and analyzed through Partial Least Squares Structural Equation Modeling (PLS-SEM). The integrated model explained 63% of the variance in behavioral intention, demonstrating substantial predictive power. Compatibility, relative advantages, and observability emerged as the strongest predictors of farmers’ support, while complexity negatively affected intention. The results highlight that policy acceptance depends on both psychological motivation and perceived innovation attributes. These findings emphasize the need for participatory, context-sensitive forest governance that aligns with local livelihood systems, enhances transparency, and simplifies administrative procedures. The study contributes to advancing innovation-driven approaches to forest policy, offering actionable insights for climate-resilient and community-based ecosystem management.

Introduction

Forests and agricultural landscapes are critical for biodiversity, climate regulation, and rural livelihoods^1,2. Yet, they face increasing threats from deforestation, land-use change, climate variability, and unsustainable resource use^3,4. Addressing these challenges requires integrative approaches linking ecological science, human behavior, and policy innovation⁵. Sustainable management of forest–agriculture interfaces, particularly in communities reliant on natural resources, is therefore a global conservation and development priority⁶. Climate change has heightened ecological and socioeconomic vulnerabilities, with altered rainfall, rising temperatures, and frequent wildfires undermining forest productivity and agricultural resilience^7,8. Effective management strategies must protect ecological integrity while supporting local livelihoods, relying not only on technical design but also on community perception, acceptance, and participation⁹.

In many regions, particularly in developing countries, forests and agricultural systems form socio-ecological mosaics where livelihoods, culture, and conservation are closely intertwined^10,11. While these landscapes risk forest conversion, they also offer opportunities for local communities to engage in participatory forest management and conservation programs. Such involvement aligns with the United Nations Sustainable Development Goals, including responsible consumption (SDG 12), climate action (SDG 13), life on land (SDG 15), partnerships (SDG 17), and poverty reduction (SDG 1). In western Iran, for instance, traditional practices like under-story cultivation and rotational grazing demonstrate agriculture integrated within over five million hectares of forest–agriculture mosaics^12,13. These landscapes highlight both the potential and tension between livelihoods and ecological protection¹⁴, emphasizing the need to understand behavioral, institutional, and innovation-related factors shaping public support for conservation policies¹⁵.

In the context of climate change, forest management increasingly relies on innovative policy instruments encompassing institutional, technological, procedural, and participatory dimensions. Institutional innovations involve new governance arrangements, incentives, and coordination mechanisms between state agencies and local communities^16,17. Technological innovations introduce improved tools, monitoring systems, and climate-smart practices^18,19, while procedural innovations simplify administration and enhance transparency^20,21. Participatory innovations expand local involvement in planning, decision-making, and knowledge sharing²². The success of these initiatives depends on how communities perceive their novelty, relevance, and feasibility²³. Despite the growing adoption of approaches such as payments for ecosystem services, community-based management, and carbon offset programs, few studies examine behavioral mechanisms shaping support^24,25. In the Zagros forests, many policies have failed due to perceptions of impracticality and unfairness²⁶, highlighting the need for context-sensitive, participatory designs that align with local expectations and foster durable conservation outcomes²⁷.

This study addresses this gap by proposing integrating behavioral and innovation theories to explain farmers’ intentions to participate in sustainable forest management under climate change. By developing an integrated framework, the research offers a novel interdisciplinary lens linking psychological determinants with perceptions of policy innovation. present study contributes to this interdisciplinary discourse by integrated framework to explain local farmers intention to participate in sustainable forest management under climate change. By linking behavioral psychology with innovation aspect, the study offers new insights into how perceptions of policy innovation shape pro-environmental attitudes, social norms, and perceived control. This approach contributes to global debates on forest–agriculture conservation by highlighting the behavioral foundations of policy innovation, offering practical guidance for designing socially inclusive, innovation-driven, and ecologically sound management systems. Such integration connects behavioral science with innovation and policy studies, bridging micro-level cognition with macro-level institutional change. It also aligns with current international calls to incorporate behavioral, ecological, and technological insights into forest policy and ecosystem management, particularly in the context of climate adaptation and biodiversity conservation.

Theoretical framework

Many forest policy and initiatives in developing countries are top-down and insufficiently aligned with local peoples, especially forest-dependent farmers’ expectations about feasibility, clarity, and local benefit^28,29. By conceptualizing perceived innovation attributes as expectations rather than post-experience judgments, this study responds directly to these historical challenges. It proposes that the success of future sustainable forest management policies hinges on the degree to which they meet individuals’ cognitive expectations of simplicity, compatibility, and fairness^30,31. Understanding farmers’ support for sustainable forest policies requires a framework that captures both the psychological mechanisms shaping behavior and the contextual drivers influencing how new policies are perceived. Conventional behavioral models often explain individual intentions but fail to account for how people interpret and respond to innovation in policy design. Conversely, innovation diffusion models explain the spread of new ideas but seldom explore the underlying motivational processes that lead individuals to adopt or reject them. This study proposes an integrative perspective that bridges these two analytical traditions. By combining the Theory of Planned Behavior (TPB) and the Diffusion of Innovation (DOI) theory, the framework captures how behavioral intention is simultaneously shaped by psychological readiness (attitude, social norms, perceived control) and innovation perception (relative advantage, compatibility, complexity, trialability, and observability). In this research, farmers’ intention to support sustainable forest management under climate change is not only a function of personal motivation but also of how they evaluate the novelty, relevance, and practicality of policy innovations introduced in their socio-ecological context. This conceptual synthesis allows for a more nuanced understanding of policy acceptance recognizing that behavioral commitment to sustainability depends as much on how innovations are framed and communicated as on how individuals think and decide.

Theory of planned behavior

The TPB³² is one of the most widely applied psychological models for explaining and predicting human behavior across social, environmental, and policy contexts. The TPB posits that an individual’s behavioral intention the motivational readiness to perform a specific behavior is the most immediate determinant of actual behavior³³. Behavioral intention, in turn, is influenced by three core psychological constructs: attitude, subjective norms, and perceived behavioral control. Attitude refers to an individual’s overall positive or negative evaluation of performing the behavior, reflecting both instrumental (useful–harmful) and affective (pleasant–unpleasant) judgments³⁴. Subjective norms represent the perceived social pressure to engage or not engage in the behavior, based on expectations from significant others such as family members, peers, or community leaders³⁵. Perceived Behavioral Control (PBC) denotes the perceived ease or difficulty of performing the behavior, which depends on one’s skills, resources, and opportunities and is conceptually similar to self-efficacy³⁶. PBC in this study reflects two complementary dimensions including self-efficacy and perceived control over external resources. From a self-efficacy perspective, PBC represents farmers’ confidence in their ability to perform sustainable forest-management behaviors, aligning with Bandura’s argument that perceived competence strongly predicts action³⁷. From a resource-based perspective, PBC captures farmers’ perceptions of having sufficient time, knowledge, institutional support, and access to tools, emphasized in environmental psychology as essential for overcoming barriers to pro-environmental action. This dual interpretation highlights that strengthening farmers’ skills and reducing institutional constraints are equally critical for enhancing participation in forest policy initiatives. According to the TPB, individuals are more likely to perform a behavior when they evaluate it favorably, believe that important others expect them to do so, and feel capable of carrying it out. The TPB has been extensively employed to study pro-environmental and conservation behaviors, including energy saving, recycling, and participation in natural resource management, making it a robust theoretical foundation for examining behavioral support for sustainable forest management under climate change^38–40.

The diffusion of innovation theory

According to the DOI theory⁴¹, the adoption of new ideas, technologies, or policy mechanisms depends largely on how individuals perceive the characteristics of the innovation. The five key perceived attributes including relative advantages, compatibility, complexity, trialability, and observability, play a central role in shaping adoption decisions. In the context of sustainable forest management, these attributes determine how local people evaluate, accept, and engage with conservation-oriented innovations such as participatory governance, payment for ecosystem services, or climate-smart forestry programs. Within the integrated DOI–TPB framework, these perceived innovation attributes are proposed to influence the three motivational determinants of behavioral intention identified in the TPB.

Relative advantage is the degree to which an individual perceives an innovation as offering superior benefits compared to existing practices⁴¹. Within the framework, the perception of relative advantage is expected to positively influence attitude, as individuals are likely to evaluate innovations favorably when they are seen as beneficial or effective⁴². In addition, widely recognized benefits may generate community support and positive social discourse, shaping subjective norms that encourage participation⁴³. Perceived advantages may also strengthen PBC, as individuals who recognize tangible benefits might feel more capable and motivated to engage⁴². Research confirmed the influence of this variable on farmers’ intention to adopt eco-friendly behaviors or new technologies^43,44. Thus, higher perceived relative advantage is hypothesized to enhance favorable attitudes, social expectations, and perceived control over supporting sustainable forest management.

Compatibility describes the extent to which an innovation aligns with individuals’ existing values, beliefs, and livelihood practices⁴¹. Innovations that fit local traditions or resource-use systems are expected to foster favorable attitudes⁴⁵, since individuals perceive them as congruent with their socio-cultural context. Compatibility is also assumed to shape subjective norms by reinforcing collective approval and shared expectations within communities⁴⁶. Moreover, when an innovation aligns with local livelihoods and institutional settings, it likely reduces perceived barriers, thereby strengthening PBC^44,47. Consequently, compatibility is proposed to contribute to more positive evaluations, greater normative support, and higher perceived ease of engagement.

Complexity refers to the perceived difficulty of understanding or implementing an innovation⁴⁸. In the integrated framework, perceived complexity is expected to be negatively associated with the TPB constructs. High levels of complexity may reduce attitude favorability by increasing uncertainty or perceived risk, whereas simpler innovations may promote more positive evaluations⁴⁹. Complexity can also weaken subjective norms, as innovations that are difficult to understand or apply are less likely to be widely endorsed or observed within a social system^44,50. In addition, greater perceived difficulty is assumed to lower PBC, as individuals may feel less confident in their ability to engage. Therefore, lower perceived complexity is anticipated to strengthen all three motivational components of the TPB.

Trialability represents the degree to which an innovation can be tested or experimented with before full implementation⁵¹. This attribute is theorized to positively influence attitude, as opportunities for limited experimentation may allow individuals to form more favorable evaluations based on experiential learning⁵². Trialability may also affect subjective norms by enabling visible pilot projects and peer discussions that increase social endorsement. Furthermore, the ability to test an innovation on a small scale can enhance PBC, as individuals gain confidence and reduce uncertainty regarding participation⁵³. Therefore, trialability is expected to encourage stronger motivational readiness to support innovative forest management practices.

Observability refers to the visibility of the innovation’s results and outcomes to others⁵³. Within the TPB framework, visible evidence of success is assumed to strengthen attitude, as observable benefits increase perceived effectiveness and credibility⁵⁴. Observability may also reinforce subjective norms by increasing social visibility, imitation, and peer influence within communities⁵⁵. Finally, observing positive results in other contexts can enhance PBC, as individuals gain confidence in their own ability to replicate the behavior⁵⁶. Consequently, higher observability is expected to contribute to more favorable attitudes, stronger normative pressures, and greater perceived capability to engage in sustainable forest management.

Based on the integrated theoretical framework, the following hypotheses are proposed:

• H1: Perceived innovation attributes positively influence attitude toward supporting forest policy and management.

• H2: Perceived innovation attributes positively influence subjective norms regarding support for forest policy and management.

• H3: Perceived innovation attributes positively influence PBC over participation in forest policy and management.

• H4: Attitude positively affects intention to support forest policy and management.

• H5: Subjective norms positively affect intention to support forest policy and management.

• H6: PBC positively affects Behavioral Intention to support forest policy and management.

• H7: Perceived innovation attributes affect intention.

The summary of influences of research constructs and theoretical framework of research are illustrated in Table 1; Fig. 1.

Table 1.

Summary of influences of constructs.

DOI components	TPB components
	Attitude	Subjective norms	Perceived behavioral control
Relative advantage	Increases favorable evaluation of the innovation’s outcomes.	Generates positive social endorsement.	Build confidence in beneficial participation.
Compatibility	Aligns with personal values, creating positive feelings.	Reinforces shared cultural expectations.	Reduces barriers, enhancing control.
Complexity	High complexity reduces positive attitude; simplicity enhances it.	Simplicity encourages social diffusion.	Simplicity increases perceived ability.
Trialability	Testing leads to experiential trust and positive evaluation.	Demonstrations strengthen community approval.	Trial opportunities enhance capability confidence.
Observability	Visible success fosters positive attitudes.	Visibility amplifies peer influence.	Seeing others succeed increases self-efficacy.

Fig. 1.

The theoretical framework of research, integrated model of diffusion of innovation and theory of planned behavior.

Materials and methods

Study area

This research was carried out in Dowreh Chegeni County, located in Lorestan Province, western Iran (Fig. 2). The county consists of three districts including Chegeni, Veysian, and Shahivand, which together include six rural districts. The area represents a typical socio-ecological system within the Zagros forest region, where human livelihoods and forest resources are intricately connected. It offers a representative case for examining the behavioral and institutional dimensions of sustainable forest policy implementation in a context characterized by ecological fragility and livelihood dependency. According to official statistics, approximately 95% of the land in the county is publicly owned, while only about 5% is privately held⁵⁷. Multiple forms of traditional land use persist in Zagros. A significant portion of forestland is used for understory cultivation, and the region’s rangelands and forests serve as a major source of livestock grazing. These overlapping land uses have led to extensive forest degradation, soil erosion, and a decline in natural regeneration capacity^58,59. Such pressures are intensified by population growth, land-use conflicts, and recurrent droughts linked to climate change.

Fig. 2.

The geographical map of the study location.

In response, governmental and local initiatives have sought to promote sustainable forest management and reduce dependence on extractive resource use. Programs such as the Toba initiative, economic afforestation projects, and tree planting on sloped lands have been introduced to enhance both environmental and socioeconomic resilience. More recently, participatory and incentive-based approaches such as community involvement in reforestation and livelihood diversification schemes have been implemented to strengthen local cooperation and compliance. Dowreh Chegeni’s combination of strong livelihood dependence on forest resources, repeated exposure to conservation interventions, and evident tensions between state-led policies and local practices makes it an ideal setting for this study. The region provides a realistic environment for analyzing how farmers perceive the novelty, fairness, and practicality of new forest policies, and how these perceptions shape their attitudes, social norms, and behavioral intentions.

Study population and sampling method

The target population consisted of rural households in Dowreh Chegeni County, whose livelihoods are directly linked to forest and natural resources. Based on local administrative records and prior surveys, the sampling frame comprised approximately 10,000 adult household members who are directly engaged in farming, livestock grazing, or other forest-related resource use. Eligible respondents were adults (≥ 18 years) who were either the household head or another adult household member primarily responsible for land and livestock-management decisions. Individuals who had lived in the study area for less than one year or who were unwilling or unable to provide informed consent were excluded. A sample of 385 was predetermined based on a statistical power analysis tailored to structural equation modelling. A stratified, multistage random sampling approach was employed to ensure representativeness across the county’s three main districts and their constituent rural districts. The county was stratified into three main districts. Proportional allocation was used so that the number of interviews per district reflected the relative size of the rural population in each stratum. Within each district, a list of eligible villages was compiled. Villages were selected at random with probability proportional to population size. In each selected village, a household listing was used to randomly select households. This multi-stage stratified sampling design was adopted to ensure adequate geographic and demographic representativeness across study county. Stratification by district was used because these areas differ in population size, livelihood dependence, and exposure to forest policies. Villages were then selected proportionally to population size, followed by random household selection to minimize selection bias. This approach ensured that the final sample of 385 respondents reflected the spatial, socioeconomic, and livelihood diversity of the region.

Data collection

Data for this research were collected through a structured questionnaire survey complemented by expert consultation and a pilot test to ensure conceptual and methodological validity. The questionnaire was designed to capture farmers’ perceptions of forest policy innovation, attitudes, subjective norms, PBC, and intention to support sustainable forest management (Table 2). All measurement items were adapted from validated scales used in prior studies on the TPB³² and the DOI⁴¹, with wording modified to suit the local forest-policy context of the Zagros region. An initial pool of statements was drafted based on the constructs of the integrated DOI–TPB framework. The draft questionnaire was then reviewed by a panel of seven experts comprising university faculty, forest policy specialists, and local natural resource officers. Their feedback helped refine the clarity, cultural relevance, and technical accuracy of the items. Content validity was ensured by evaluating each item for relevance, comprehensiveness, and simplicity, resulting in minor adjustments to language and response format. to ensure the conceptual and cultural validity of the measurement instrument, we assessed the content validity of the Persian questionnaire using both the Content Validity Index (CVI) and Content Validity Ratio (CVR). The CVI is a quantitative indicator that reflects the extent to which each item is judged by experts as relevant, clear, and representative of the intended construct. Values were above 0.79, which are considered acceptable for instruments evaluated by more than six experts⁶⁰. The CVR, proposed by Lawshe (1975), quantifies the degree of essentiality of each item, reflecting whether an item is necessary for measuring the construct. According to Lawshe’s critical values, for a panel of seven experts used in this study, CVR values were above 0.99 which is recommended to consider an item essential. To ensure cultural adaptation, items were reviewed for alignment with local terminology, customary forest-management practices, and sociocultural norms of the Zagros rural communities. Statements that could be misunderstood, carried culturally specific meanings, or did not reflect local policy experiences were modified to improve clarity and contextual relevance. This stage was supported by consultations with local forestry practitioners and extension agents familiar with local communication styles and institutional arrangements.

Table 2.

The measurement statements of constructs of research model.

Constructs	Codes	Statements	Derived from
Relative advantage	RA1	The forest management policy should provide clear benefits.	DOI theory
	RA2	Implementing policy should improve both environmental and livelihood outcomes.
	RA3	Innovation should make forest management more effective and sustainable.
	RA4	The policy must offer advantages that justify the effort required to adopt it.
	RA5	New approach should bring better long-term results for my community.
Compatibility	CO1	The new forest policy should fit well with our local forest management traditions.
	CO2	Innovation must be consistent with my values about forest use and conservation.
	CO3	The program must align with the needs of local farmers and forest users.
	CO4	The proposed practices should suit our existing livelihood conditions.
	CO5	Adopting this policy must not conflict with my other activities or responsibilities.
Complexity	CX1	The new policy should be easy to understand and implement.
	CX2	The procedures required by the policy must be clear and straightforward.
	CX3	I feel confident that I can learn how to apply new policy effectively.
	CX4	The new management approach should not be complicated for implementation.
	CX5	The rules and requirements should be simple enough for farmers to follow.
Trialability	TR1	It would be possible to test new policy on a small scale before full implementation.
	TR2	Local people can experiment with innovation without taking major risks.
	TR3	The policy should allow gradual participation or step-by-step involvement.
	TR4	I need to try new approach and see the results before deciding to fully adopt it.
	TR5	The government should allow flexibility for local adaptation during the trial phase.
Observability	OB1	I have seen other communities successfully apply similar forest management innovations.
	OB2	The positive results of policy should be visible to people around me.
	OB3	I can easily observe the environmental benefits of adopting new approach.
	OB4	The outcomes of the program should be well publicized or shared among farmers.
	OB5	Seeing others adopt this policy motivates me to participate as well.
Attitude	ATT1	Supporting sustainable forest management is beneficial for the environment.	TPB theory
	ATT2	I believe participating in sustainable forest management improves my livelihood.
	ATT3	I have a positive opinion about engaging in forest management activities.
	ATT4	Participating in such programs is a worthwhile and responsible action.
	ATT5	I feel that supporting sustainable forest management is a good idea.
Subjective norms	SN1	People whose opinions I value think I should support forest management.
	SN2	My family and community members expect me to participate in forest conservation.
	SN3	Local leaders and officials encourage me to engage in forest management.
	SN4	Most people important to me approve of my involvement in such initiatives.
	SN5	There is social pressure in my community to take part in forest management.
Perceived behavioral control	PBC1	I have the knowledge necessary to participate in forest management.
	PBC2	I have the skills necessary to participate in forest management.
	PBC3	I feel confident that I can overcome obstacles to participate in forest management.
	PBC4	I have access to the resources (time, tools, support) needed to be involved.
	PBC5	It would be easy for me to take part in forest management.
Intention	IN1	I intend to participate in forest management programs in the near future.
	IN2	I will actively support government or community forest conservation initiatives.
	IN3	I plan to adopt the recommended forest management practices on my land.
	IN4	I am willing to collaborate with others to promote sustainable forest use.
	IN5	I will encourage other farmers to join sustainable forest management efforts

A pilot test involving 30 farmers was conducted to assess the reliability, comprehension, and timing of the instrument. Feedback from the pilot led to revisions in item wording, sequencing, and scale balance to enhance respondent understanding. Internal consistency was checked using Cronbach’s alpha, and all constructs exceeded the 0.70 threshold, confirming acceptable reliability before the main survey⁶². Responses were recorded on a five-point Likert scale ranging from 1 (strongly disagree) to 5 (strongly agree)⁶³. Trained enumerators conducted face-to-face interviews with the selected respondents during June to October 2024. Interviews were conducted in the local language to ensure comprehension, and immediate on-site checks were performed to verify completeness and logical consistency. After completing data collection and analysis, the finalized Persian questionnaire was translated into English solely for the purpose of publication. This translation followed a careful verification process to ensure conceptual equivalence rather than literal word-for-word transfer. The research team cross-checked the translated items against the original Persian version to confirm accuracy, preserve the meaning of culturally adapted phrases, and maintain fidelity to the theoretical constructs.

Ethical principles were strictly followed throughout the research. The data collection process for this study was conducted in accordance with institutional and international ethical standards for research involving human participants. This study was non-interventional social science research. Nevertheless, all procedures strictly followed the principles of voluntary participation, informed consent, confidentiality, and anonymity. Participants were informed about the purpose of the study and their right to withdraw at any time without consequence. Written informed consent was obtained from all respondents prior to participation. The study complied with the ethical principles outlined in the Declaration of Helsinki and with the Ethical Guidelines for Social Science Research Involving Human Participants approved by Research Ethics Committees of Lorestan University of Medicine Sciences.

Data analysis

This study applied Partial Least Squares Structural Equation Modeling (PLS-SEM) to analyze the interrelationships among constructs derived from the DOI and the TPB theories. PLS-SEM was chosen because it enables simultaneous assessment of multiple latent variables, their indicators, and causal pathways while accounting for measurement errors⁶⁴. The study’s objective is primarily predictive and exploratory, focusing on explaining variance in farmers’ behavioral intentions rather than confirming a well-established theory. The integrated DOI–TPB framework involves complex causal links through which innovation attributes influence behavioral determinants and, in turn, behavioral intention. The research model includes multiple latent constructs, numerous indicators, and several indirect pathways, making it structurally complex and better suited to PLS-SEM’s flexibility. Additionally, preliminary tests indicated deviations from multivariate normality, for which PLS-SEM is more robust than Covariance based Structural Equation Modeling (CB-SEM)⁶⁵. The moderate sample size (n = 385) also aligns with PLS-SEM requirements, ensuring adequate statistical power. Together, these considerations make PLS-SEM the more appropriate analytical approach.

Measurement model assessment

The reliability and validity of the reflective measurement model were assessed. To ensure the internal consistency and convergent validity of the constructs, multiple reliability indices were examined, including Cronbach’s alpha, rho_A, and Composite Reliability (CR). Cronbach’s alpha and rho_A values greater than 0.70 were considered indicative of satisfactory internal consistency, while CR values exceeding 0.70 confirmed the reliability of the latent constructs. Convergent validity was evaluated through the Average Variance Extracted (AVE), where values above 0.50 indicate that a construct explains more than half of the variance in its observed indicators (Fornell and Larcker, 1981).

To establish discriminant validity, both the Fornell–Larcker criterion and the Heterotrait–Monotrait ratio (HTMT) were employed. The Fornell–Larcker approach compares the square root of the AVE for each construct with its correlations with other constructs; discriminant validity is supported when each construct’s AVE square root exceeds the corresponding inter-construct correlations⁶⁶. In addition, the HTMT ratio was calculated to provide a more stringent assessment of discriminant validity⁶⁷. HTMT values below 0.85 (and, more conservatively, below 0.90) indicate sufficient discriminant separation among constructs. Together, these procedures ensured that all constructs in the integrated DOI–TPB framework demonstrated acceptable levels of reliability, convergent validity, and discriminant validity prior to structural model evaluation.

Structural model assessment and hypothesis testing

After confirming the reliability and validity of the measurement model, the structural model was evaluated to test the hypothesized relationships among the latent constructs of the integrated DOI–TPB framework. The analysis was carried out with SmartPLS₃ software. This method is appropriate for complex models involving multiple mediating relationships and relatively moderate sample sizes, as it focuses on maximizing the explained variance of dependent constructs rather than assuming multivariate normality. The evaluation of the structural model involved examining path coefficients (β values), their t-statistics, and p-values obtained through a bootstrapping procedure with 5,000 subsamples. Each hypothesized path was tested for statistical significance at the 0.05 level to determine whether the proposed relationships were supported. The direction and magnitude of each coefficient were used to interpret the strength and significance of influence among the constructs. Model explanatory power was assessed using the coefficient of determination (R²) for each endogenous variable. Based on these criteria, the hypothesis testing results were interpreted according to the significance and direction of the path coefficients. Significant positive relationships among the constructs were considered empirical support for the study’s hypotheses. Specifically, the results examined how perceived innovation attributes influenced the core psychological determinants of behavior and, in turn, how these determinants shaped farmers’ behavioral intention to support sustainable forest policy under climate change.

Results

Reliability and validity

As presented in Table 3, all constructs demonstrated satisfactory internal consistency. Cronbach’s alpha values ranged from 0.771 to 0.833, exceeding the widely accepted threshold of 0.70, which indicates that the items for each construct consistently measured the same underlying concept⁶². The rho_A coefficients followed a similar pattern (0.776–0.21), further confirming the robustness of construct reliability. CR values were all above 0.844, ranging between 0.844 and 0.872, thereby satisfying the recommended minimum of 0.70 and evidencing strong overall measurement reliability. The AVE values varied from 0.521 to 0.578, which surpasses the 0.50 benchmark⁶⁸, confirming that each construct explains more than half of the variance in its indicators. Collectively, these results demonstrate that all latent constructs achieved adequate levels of reliability and convergent validity. Hence, the measurement model exhibits solid psychometric properties, providing a reliable foundation for subsequent structural model assessment.

Table 3.

The results of reliability and validity assessment.

Constructs	Cronbach’s α	rho_A	CR	AVE
Attitude	0.800	0.804	0.862	0.555
Compatibility	0.798	0.802	0.861	0.554
Complexity	0.833	0.908	0.863	0.562
Intention	0.816	0.821	0.872	0.577
Observability	0.807	0.809	0.866	0.564
Perceived behavioral control	0.813	0.817	0.870	0.573
Relative advantage	0.800	0.803	0.862	0.555
Subjective norms	0.817	0.818	0.872	0.578
Trialability	0.771	0.776	0.844	0.521

Discriminant validity

Discriminant validity was assessed through both the Fornell–Larcker criterion and the heterotrait–monotrait ratio (HTMT) of correlations to ensure the empirical distinctiveness of the latent constructs. As presented in Table 4, the square roots of the AVE, reported on the diagonal of the correlation matrix, were all greater than the corresponding inter-construct correlations. The diagonal values ranged from 0.722 to 0.76, indicating that each construct shared more variance with its own indicators than with those of other constructs⁶⁶. This pattern demonstrates that constructs are statistically distinct from one another. Complementing these results, the HTMT analysis (Table 5) yielded values between 0.150 and 0.757, all of which fall below the conservative threshold of 0.85 ⁶⁷, providing additional evidence of discriminant validity. These findings collectively confirm that multicollinearity is not a concern and that each latent construct is empirically unique. Taken together, the Fornell–Larcker and HTMT results affirm that the measurement model satisfies the criteria for discriminant validity, thereby ensuring the robustness and conceptual distinctiveness of the constructs used in this study.

Table 4.

The Fornell and larcker assessment of model.

	1	2	3	4	5	6	7	8	9
Attitude	0.745
Compatibility	0.557	0.744
Complexity	-0.15	-0.125	0.75
Intention	0.589	0.599	– 0.293	0.76
Observability	0.464	0.391	– 0.097	0.513	0.751
Perceived behavioral control	0.543	0.494	– 0.05	0.608	0.49	0.757
Relative advantage	0.521	0.478	– 0.081	0.59	0.434	0.491	0.745
Subjective norms	0.529	0.534	– 0.083	0.621	0.46	0.551	0.609	0.76
Trialability	0.418	0.465	– 0.013	0.412	0.235	0.436	0.402	0.44	0.722

Table 5.

The HTMT assessment of model.

	1	2	3	4	5	6	7	8
Attitude
Compatibility	0.696
Complexity	0.15	0.134
Intention	0.719	0.741	0.307
Observability	0.569	0.48	0.119	0.626
Perceived behavioral control	0.676	0.614	0.105	0.744	0.598
Relative advantage	0.643	0.59	0.105	0.727	0.531	0.606
Subjective norms	0.65	0.656	0.131	0.757	0.562	0.678	0.75
Trialability	0.525	0.592	0.127	0.511	0.285	0.539	0.503	0.547

Structural model evaluation

The results of structural model (Fig. 3) demonstrated substantial explanatory capability. As presented in Table 6, the R² values for attitude (0.449), subjective norms (0.489), and PBC (0.419) indicated moderate explanatory power, while intention exhibited a strong R² of 0.631, showing that over 63% of the variance in farmers’ intention to support sustainable forest policy was explained by its predictors. Adjusted R² values were only marginally lower, confirming model stability and the absence of overfitting. Similarly, the Stone–Geisser’s Q² values were all greater than zero (0.228–0.335), establishing satisfactory predictive relevance for all endogenous constructs. These results suggest that the integrated DOI–TPB model offers a robust predictive structure for explaining behavioral intention in the context of sustainable forest management.

Fig. 3.

The structural model of research with coefficients and R².

Table 6.

R² and adjusted R² and Q² of independent constructs.

Constructs	R 2	R2 Adjusted	Q2
Attitude	0.449	0.439	0.228
Subjective norms	0.489	0.480	0.262
Perceived behavioral control	0.419	0.409	0.222
Intention	0.631	0.620	0.335

Hypothesis testing

Influence of innovation attributes on TPB constructs

The results of effects and hypotheses test are shown in Tables 7 and 8, respectively. Perceived innovation attributes exhibited substantial effects on the three psychological determinants of behavioral intention. Relative advantage positively and significantly influenced attitude (β = 0.227, t = 3.984, p < 0.001), subjective norms (β = 0.366, t = 6.717, p < 0.001), and PBC (β = 0.194, t = 3.325, p < 0.001) (H1). Likewise, compatibility (H2) exerted significant positive effects on attitude (β = 0.291, t = 5.208, p < 0.001), subjective norms (β = 0.217, t = 3.636, p < 0.001), and PBC (β = 0.201, t = 3.695, p < 0.001). These findings indicate that when forest policies are perceived as advantageous and aligned with local values and practices, farmers exhibit more favorable attitudes, stronger social endorsement, and greater confidence in their ability to participate. Complexity showed negative but mostly insignificant relationships with the TPB constructs (β = −0.073 to 0.021, p > 0.05) (H3), suggesting that the perceived difficulty of understanding or implementing new policies had a limited effect on motivational determinants. This outcome implies that while simplicity remains desirable, it was not a decisive barrier in this context possibly because farmers are accustomed to adaptive management under complex ecological and administrative conditions.

Table 7.

Direct, indirect and total effects of constructs of research model.

Constructs	Attitude	Subjective norms	Perceived behavioral control	Intention (Indirect)	Intention (Direct)	Intention (Total)
Attitude					0.101	0.101
Subjective norms					0.173	0.173
Perceived behavioral control					0.208	0.208
Relative advantage	0.227	0.366	0.194	0.127	0.166	0.293
Compatibility	0.291	0.217	0.201	0.109	0.189	0.298
Complexity	– 0.073	– 0.007	0.021	– 0.004	– 0.205	– 0.209
Trialability	0.141	0.150	0.199	0.082	0.018	0.100
Observability	0.212	0.181	0.283	0.112	0.114	0.226

Table 8.

The results of hypotheses of integrated model of research.

		t statistics	p values	Direction	Supported
H1	Relative advantage > attitude	3.984	0.00	+	Yes
	Relative advantage > subjective norms	6.717	0.00	+
	Relative advantage > perceived behavioral control	3.325	0.00	+
H2	Compatibility > attitude	5.208	0.00	+	Yes
	Compatibility > subjective norms	3.636	0.00	+
	Compatibility > perceived behavioral control	3.695	0.00	+
H3	Complexity > attitude	1.655	0.09	–	No
	Complexity > subjective norms	0.145	0.88	–
	Complexity > perceived behavioral control	0.435	0.66	–
H4	Trialability > attitude	2.776	0.00	+	Yes
	Trialability > subjective norms	2.926	0.00	+
	Trialability > perceived behavioral control	4.106	0.00	+
H5	Observability > attitude	4.005	0.00	+	Yes
	Observability > subjective norms	3.357	0.00	+
	Observability > perceived behavioral control	5.61	0.00	+
H6	Attitude -> intention	1.963	0.05	+	Yes
H7	Subjective norms > intention	2.991	0.00	+	Yes
H8	Perceived behavioral control > intention	4.295	0.00	+	Yes
H9	Relative advantage > intention	3.102	0.00	+	Yes
	Compatibility > intention	3.502	0.00	+
	Complexity > intention	5.091	0.00	+
	Trialability > intention	0.41	0.68	–
	Observability > intention	2.531	0.01	+

Trialability (H4) was found to significantly enhance all three TPB components; attitude (β = 0.141, t = 2.776, p < 0.01), subjective norms (β = 0.150, t = 2.926, p < 0.01), and PBC (β = 0.199, t = 4.106, p < 0.001). These results highlight the importance of experiential engagement and pilot initiatives that allow farmers to test and observe forest policies before full-scale adoption. Similarly, observability (H5) had strong positive effects on attitude (β = 0.212, t = 4.005, p < 0.001), subjective norms (β = 0.181, t = 3.357, p < 0.01), and PBC (β = 0.283, t = 5.610, p < 0.001). These findings confirm that visible and demonstrable policy outcomes strengthen both individual and collective motivation to support innovation in forest management.

Effects of TPB constructs on intention

All three motivational determinants of the TPB significantly predicted behavioral intention. Attitude (H6) had a positive and marginally significant effect (β = 0.101, t = 1.963, p < 0.05), indicating that favorable evaluations of forest policy contribute to higher support intentions. Subjective norms (H7) exerted a significant positive influence (β = 0.173, t = 2.991, p < 0.01), emphasizing the role of social expectations and peer endorsement in shaping participation behavior. PBC (H8) demonstrated the strongest effect (β = 0.208, t = 4.295, p < 0.001), suggesting that a sense of capability and opportunity is a critical factor driving farmers’ willingness to support sustainable forest policies.

Direct effects of innovation attributes on intention

Beyond their indirect effects through the TPB constructs, certain innovation attributes (H9) also had significant direct impacts on behavioral intention. Relative advantage (β = 0.166, t = 3.102, p < 0.01), compatibility (β = 0.189, t = 3.502, p < 0.001), and observability (β = 0.114, t = 2.531, p < 0.05) were all significant positive predictors of intention. In contrast, complexity exhibited a significant negative effect (β = −0.205, t = 5.091, p < 0.001), reaffirming that perceived difficulty in policy implementation directly reduces support intentions. Trialability, however, did not show a significant direct influence (β = 0.018, t = 0.410, p > 0.05), indicating that its role operates mainly through enhancing intermediate psychological constructs.

The total effects analysis revealed that compatibility (β = 0.298) and relative advantage (β = 0.293) were the two strongest overall predictors of behavioral intention, followed by observability (β = 0.226). These findings underscore that policies perceived as both beneficial and locally appropriate are most likely to gain sustained support. Conversely, perceived complexity had a substantial negative total effect (β = −0.209), confirming that administrative or procedural burdens undermine behavioral commitment. Overall, the results validate the integrated DOI–TPB framework as a meaningful explanatory model for farmers’ behavioral intentions toward sustainable forest management under climate change. The findings suggest that beyond traditional attitudinal and normative factors, perceptions of policy innovation, especially its compatibility, advantages, and visibility play a decisive role in shaping community participation in forest conservation initiatives.

Discussion

The findings of this study provide robust empirical evidence supporting the integration of the TPB and the DOI theory to explain farmers’ intention to support sustainable forest policy under climate change. The proposed framework effectively captured how perceptions of policy innovation, particularly compatibility, relative advantage, and observability, interact with psychological determinants such as attitude, subjective norms, and PBC to shape behavioral intentions. The results underscore that farmers’ willingness to engage in sustainable forest management is not driven solely by individual motivation but also by the perceived relevance, simplicity, and visibility of policy measures. Policies that are perceived as compatible with local livelihood systems, beneficial compared to existing practices, and observable in their outcomes are more likely to gain community acceptance. These findings reinforce calls for participatory, context-sensitive, and innovation-oriented approaches to forest governance, particularly in regions like the Zagros where top-down conservation measures have historically failed due to weak local engagement and administrative complexity. The integrated DOI–TPB framework thus bridges a critical gap between behavioral theory and innovation diffusion by demonstrating how farmers’ cognitive and social evaluations of policy characteristics influence their readiness to cooperate with conservation initiatives.

The results strongly support Hypotheses 1–3, showing that perceived innovation attributes significantly influenced the three motivational components of the TPB. Among these attributes, relative advantages, compatibility, and observability exhibited the most consistent and significant effects across all three constructs. This implies that farmers’ psychological readiness to support sustainable forest management is closely tied to how they evaluate the usefulness, appropriateness, and visibility of policy innovations. The positive impact of relative advantage (H1) confirms that farmers are more likely to form favorable attitudes and perceive stronger social and personal control when policies promise tangible benefits such as improved livelihoods, resource efficiency, or equitable participation. This finding aligns with the DOI theory, which posits that innovations perceived as advantageous diffuse more rapidly⁴¹, and with empirical studies showing similar patterns in environmental and agricultural contexts^42,43. In the Zagros region, where farmers’ livelihoods are highly dependent on forest resources, the perceived capacity of new policies to balance conservation with livelihood gains is central to shaping supportive attitudes.

Similarly, the strong effects of compatibility (H2) on all three TPB constructs indicate that the success of forest policy innovations depends on their fit with local cultural and livelihood systems. Farmers in the study area value continuity with traditional practices and local norms and thus perceive compatible policies as legitimate and feasible. This finding corroborates earlier research showing that innovations congruent with local traditions are more likely to be accepted and sustained^45,46. Compatibility also reinforces collective norms, as community members tend to endorse innovations that align with shared beliefs and customary land-use systems.

In contrast, complexity (H3) demonstrated weak and nonsignificant relationships with the TPB constructs, suggesting that perceived difficulty was not a dominant psychological barrier. The research confirmed the negative role of complexity in intention to acceptance a new practice^48,50. The weak effect of complexity in this study suggests that the construct may function differently in the socio-ecological and institutional context of the Zagros region. Unlike many settings where procedural difficulty strongly deters adoption^48,69, farmers here are already accustomed to navigating bureaucratic and ecological complexity in state-led forest programs. As a result, complexity may not be perceived as a novel barrier but rather as an expected feature of government interventions. It is also possible that complexity carries a different cognitive meaning, reflecting administrative opacity rather than technical difficulty. Because such opacity is already normalized, its marginal effect on attitudes and norms may be muted. This contextual interpretation explains why complexity exerted a direct negative effect on intention but had limited influence on psychological antecedents.

Trialability (H4) and observability (H5) both showed significant positive influences. These results indicate that experiential learning and visible outcomes are vital for building trust and social endorsement. Previous studies have similarly emphasized that demonstration plots, participatory experimentation, and transparent communication enhance the diffusion of conservation practices^51,52. Farmers who can experiment with pilot programs or witness the success of peers gain confidence and social validation, reinforcing their willingness to participate. The lack of tangible or observable benefits diminishes both attitude and perceptions of relative advantage, leading farmers to question whether participation will meaningfully improve their livelihoods or forest conditions. These conditions are reinforced by farmers’ past negative experiences with forest regulations, which contribute to skepticism regarding new initiatives; constraints in time, labor, and technical knowledge, which limit farmers’ perceived capability to comply with policy requirements; and restricted access to timely and comprehensible information, which reduces both perceived control and confidence in decision-making.

The results provide strong empirical support for Hypotheses 4–6, consistent with the TPB assumptions. All three psychological determinants significantly predicted farmers’ intention to support sustainable forest management, with PBC exerting the strongest influence. The positive relationship between attitude (H6) and behavioral intention indicates that farmers with favorable evaluations of sustainable forest policies are more inclined to support them. This finding aligns with the TPB and previous environmental studies emphasizing that positive beliefs about outcomes enhance intention^70,71. However, the relatively smaller effect of attitude compared to other factors may reflect a pragmatic mindset among rural communities in Zagros, where supportive attitudes do not always translate into active participation due to contextual constraints such as limited resources or institutional barriers. This highlights the importance of strengthening attitudes in promotional and educational programs. It is essential to give specific attention to attitudes and their enhancement within such programs. Since attitude serves as a fundamental factor, its improvement can lead to the reinforcement of other influencing factors.

The significant role of subjective norms (H7) underscores the importance of social approval and peer influence in shaping farmers’ behavioral decisions. This aligns with findings from^72,73, who reported that community expectations and the influence of local leaders are decisive in collective environmental action. In tightly knit rural societies, conformity to group norms and the desire for social harmony often outweigh individual reasoning, making social endorsement a crucial determinant of pro-environmental intentions. The significant effect of subjective norms on farmers’ intentions highlights the need to better understand the specific social referents shaping compliance with forest policies in the Zagros context. Rather than reflecting undifferentiated community pressure, these normative influences likely emerge from distinct and hierarchically layered sources, including respected village elders and religious authorities whose endorsement can legitimize policy initiatives; peer farmers whose visible adoption of practices such as rotational grazing or restoration activities reinforces behavioral expectations within close social networks; and forestry officials whose formal authority and regular interactions create expectations of compliance. Recognizing these differentiated normative pathways suggests that policy communication should strategically mobilize trusted local leaders and strengthen peer-to-peer learning to amplify supportive social norms. At the same time, the study’s findings underscore the importance of addressing key barriers that undermine participation.

The strong effect of PBC (H8) reflects the practical dimension of behavioral intention. This finding is consistent with prior research in conservation and agriculture indicating that self-efficacy and perceived resource availability are among the most powerful predictors of behavioral intention^69,74. Farmers are more willing to engage in sustainable management when they feel capable of doing so, supported by adequate knowledge, institutional backing, and manageable procedures. In the context of the Zagros, the perceived ability to navigate policy requirements and access necessary resources may be particularly critical given past experiences with restrictive or complex state-led programs. Administrative complexity and institutional opacity continue to discourage engagement, indicating the need for streamlined procedures, localized documentation support, and multilingual, visually accessible communication tools. Moreover, limited technical skills and resource constraints reduce farmers’ PBC, underscoring the value of skill-building workshops, stepwise participation mechanisms, and decentralized technical assistance delivered through village councils.

The analysis of direct and total effects further confirms the mediating structure proposed by the integrated DOI–TPB model. Beyond their indirect influence through TPB constructs, relative advantage, compatibility, and observability also directly predicted behavioral intention (H9). These findings suggest that farmers’ intention to support sustainable forest policy is shaped both by underlying psychological mechanisms and by direct cognitive assessments of innovation characteristics. The significant direct effects of relative advantage and compatibility imply that farmers’ decision-making is partly instrumental: they evaluate forest policies not only through attitudinal and normative lenses but also in terms of their practical benefits and contextual fit. This echoes previous studies highlighting that policy acceptance increases when innovations are perceived as economically beneficial and culturally aligned^42,47. In the Zagros, policies that reduce livelihood risk or complement traditional practices (e.g., rotational grazing, understory cultivation) are viewed as more legitimate and attainable. The significant positive influence of observability on intention emphasizes the power of demonstration and transparency. Visible success stories and peer experiences enhance credibility and trust, reducing uncertainty regarding policy outcomes. This is consistent with findings from research^53,54, who argued that observability accelerates behavioral diffusion in environmental initiatives by fostering imitation and perceived self-efficacy.

Conversely, complexity exhibited a significant negative direct effect on intention, confirming that administrative or procedural burdens deter participation. This pattern mirrors the historical challenges of forest policy implementation in the Zagros, where unclear procedures, bureaucratic requirements, and poor communication have discouraged engagement^26,27. Simplifying policy processes and improving accessibility could therefore be critical for enhancing public support. Finally, the nonsignificant direct effect of trialability suggests that its influence operates indirectly through improved attitudes and perceived control rather than directly shaping intention. This finding aligns with⁵¹ who found that experimentation fosters behavioral readiness primarily by building trust and reducing uncertainty, which in turn influence intention indirectly. The combined evidence from the hypothesis testing validates the integrated DOI–TPB model as a powerful explanatory tool for understanding behavioral support for sustainable forest policy in socio-ecological systems like the Zagros. Farmers’ behavioral intention emerges as a function of both their internal psychological evaluations and their perceptions of the policy environment. Compatibility and relative advantage are particularly decisive, highlighting that conservation policies must align with existing livelihood structures and demonstrate clear socioeconomic benefits.

Implications

Theoretical implications

This study contributes several theoretical advancements to the existing literature on behavioral and innovation research in environmental policy contexts. First, by integrating the TPB with the DOI theory, it extends both frameworks beyond their conventional domains. Whereas the TPB has been widely used to explain pro-environmental and conservation behaviors and the DOI primarily addresses the spread of technological or organizational innovations, this study demonstrates their complementary value in explaining behavioral support for policy innovations in natural resource management. The integrated model captures the dual nature of behavioral intention rooted both in individual cognitive motivation and in perceptions of innovation attributes. This synthesis therefore bridges psychological and contextual explanations of behavior, providing a more comprehensive understanding of why individuals accept or resist environmental policies.

Second, the findings advance theoretical knowledge by showing that perceptions of innovation attributes, particularly compatibility, relative advantage, and observability, not only influence the core motivational components of the TPB but also exert direct effects on behavioral intention. This dual influence highlights that innovation-related perceptions function both as antecedents and independent drivers of behavioral intention. By empirically validating these relationships, the study extends the TPB to account for contextual and perceptual dimensions that are often neglected in its classical formulation. It also refines the DOI by embedding its innovation attributes within a psychological decision-making framework, moving beyond diffusion dynamics to examine motivational processes. Third, the results provide theoretical insight into the mediating mechanisms that link innovation perceptions to behavior. The significant indirect effects through attitude, subjective norms, and PBC reveal that innovation perceptions shape behavioral intention largely by enhancing cognitive and social readiness. This highlights the importance of considering psychological mediators in diffusion studies and supports a more nuanced conceptualization of innovation adoption as a socially embedded, cognitively driven process rather than a purely informational one.

Fourth, this study contributes to the literature on conservation behavior by contextualizing behavioral theory within a socio-ecological setting characterized by policy complexity, livelihood dependence, and institutional distrust. The validation of the integrated model in the Zagros region demonstrates that the TPB and DOI can be effectively applied in developing-country contexts where resource users face both economic constraints and policy uncertainty. This contextual application extends the theoretical generalizability of both models and provides a framework adaptable to other rural and resource-dependent environments where innovation adoption intersects with social norms and institutional legacies. Finally, the study introduces a theoretical shift from viewing behavioral support as a response to external policy enforcement toward understanding it as an outcome of perceived innovation fairness, feasibility, and visibility. This conceptual reframing aligns with recent interdisciplinary movements in behavioral policy research that emphasize expectation-based cognition—how individuals form intentions based on perceived future outcomes rather than direct experience. By integrating expectation-driven innovation perceptions with the motivational structure of the TPB, this research advances a more dynamic theoretical lens for analyzing environmental policy acceptance under conditions of uncertainty and change.

Policy and managerial implications

The findings of this study offer clear, context-grounded guidance for improving the design and implementation of sustainable forest policies in the Zagros region. The results emphasize that farmers’ support is shaped not only by regulatory incentives but also by how policies are perceived in terms of compatibility, relative advantage, observability, complexity, and PBC. Based on these empirical patterns, several targeted interventions are recommended. First, enhancing compatibility is essential. Because compatibility had the strongest total effect on intention, policies should be co-designed with farmers to ensure alignment with local livelihood systems. Practical measures include integrating rotational grazing, supporting understory cultivation where ecologically feasible, and aligning reforestation schedules with farming calendars. Participatory mapping and village-level “compatibility assessments” can help identify livelihood–policy mismatches before implementation. Second, leveraging relative advantage requires linking policy measures to visible livelihood benefits. Farmers were more supportive when policies appeared economically and practically beneficial. Policymakers should therefore establish payment for ecosystem services, small livelihood grants for sustainable fodder production, and market incentives for non-timber forest products. Communicating direct benefits such as improved soil moisture, reduced erosion, or income diversification can shift perceptions from compliance to collaboration.

Third, observability should be strengthened by investing in demonstration-based learning. Given observability’s strong influence on intention, forest agencies should establish demonstration plots, model villages, and farmer-to-farmer learning exchanges. Public field days, local exhibitions of successful reforestation outcomes, and short visual updates shared through village communication channels can enhance transparency and build trust. As farmers see tangible ecological and economic results, social diffusion and local endorsement increase. Fourth, administrative complexity must be substantially reduced. Complexity had a significant negative effect on intention, highlighting the urgent need to simplify procedures. Streamlined permit processes, clear step-by-step guidelines, and local documentation support centers can reduce confusion and transaction costs. Providing multilingual materials (Persian and local Kurdish/Luri dialects) and visually simplified instructions would address the administrative opacity frequently cited in the Zagros region. Fifth, strengthening PBC demands targeted capacity building. Since PBC exerted the strongest effect among TPB predictors, policy implementation should be supported by skill-building workshops, hands-on training on climate-smart forestry, and accessible technical assistance. Establishing decentralized support units within village councils can help ensure stable access to tools, seedlings, and advisory services, thereby improving farmers’ confidence in their ability to comply with forest programs.

Sixth, social norms should be more explicitly activated. Subjective norms significantly influenced behavioral intention, suggesting that community leaders, elders, and village councils can play strategic roles. Policymakers should engage respected local actors as policy ambassadors, organize community dialogues to create shared commitments, and promote public recognition programs for villages demonstrating exemplary forest stewardship. Strengthening positive normative pressure can accelerate adoption in tightly knit rural communities. Collectively, these targeted measures demonstrate how each recommendation directly addresses the empirically identified barriers (complexity, administrative opacity) and motivators (compatibility, relative advantage, observability, social endorsement, and perceived control). By grounding interventions in socio-ecological realities of the Zagros and aligning them with farmers’ lived experiences, policymakers can promote more equitable, feasible, and sustainable forest management outcomes.

Implications for similar forest–agriculture systems worldwide

Although this study was conducted in the Zagros region of western Iran, the findings have relevance for a wide range of socio-ecological systems where forest–agriculture mosaics are central to rural livelihoods. Similar dynamics can be observed in the Himalayan forest–agriculture interfaces, the dryland forest communities of North Africa, and the Mediterranean agroforestry landscapes. These regions share several structural characteristics with the study area including strong livelihood dependence on forests and rangelands, exposure to high levels of climate variability and resource scarcity, and the prevalence of mixed land-use systems in which agriculture, grazing, and forest extraction coexist. In each of these contexts, policy innovations must balance ecological goals with local livelihood needs a challenge closely aligned with the compatibility and relative advantage dimensions identified in this study. Likewise, the importance of observability and social norms is highly transferable to settings where community-based management, collective decision-making, and peer learning remain central features of resource governance. The negative influence of complexity also resonates strongly in regions where administrative procedures and institutional fragmentation create barriers to participation. Therefore, the behavioral and innovation-related mechanisms identified here may offer valuable insights for designing participatory and context-sensitive forest policies across diverse forest–agriculture mosaics globally. Future applications of this framework in these regions could help identify common patterns, validate cross-cultural robustness, and refine behavioral interventions for climate-resilient natural resource governance.

Limitations and suggestions for future research

While this study provides valuable insights into the behavioral and innovation-related factors associated with farmers’ intentions to support sustainable forest policy, several limitations should be acknowledged to ensure an appropriate interpretation of the findings. First, the study relied on a cross-sectional research design based on self-reported questionnaire data. As a result, the observed relationships among constructs represent associations rather than causal effects. Although the integrated DOI–TPB framework offers a theoretically grounded explanation of behavioral intention, the temporal ordering of variables cannot be empirically confirmed. Future research could employ longitudinal designs or repeated measurements before and after policy interventions to better capture changes in attitudes, perceptions of innovation, and behavioral intentions over time, and to assess potential causal dynamics. Second, the dependent variable in this study was behavioral intention rather than observed behavior. Intentions are widely recognized as strong predictors of behavior; however, they do not always translate into actual participation due to contextual constraints, institutional barriers, or unforeseen livelihood pressures. Future studies should incorporate behavioral indicators, such as participation in forest programs, compliance records, or observed land-use practices, to more directly assess the link between intention and actual behavior in forest governance contexts.

Third, the study was conducted in a specific socio-ecological and institutional setting—the Zagros forest–agriculture interface in western Iran. This region is characterized by strong livelihood dependence on forest resources, a history of top-down policy implementation, and ongoing climate-related pressures. These contextual characteristics may influence farmers’ perceptions and behavioral responses, thereby limiting the generalizability of the findings to other regions. Replication of the integrated DOI–TPB framework in different ecological, cultural, and governance contexts would help assess the broader applicability and robustness of the model. Fourth, although the study incorporated key innovation attributes derived from the Diffusion of Innovation theory, it did not explicitly account for other contextual and institutional factors that may shape policy acceptance. Variables such as institutional trust, perceived fairness, procedural justice, risk perception, and past experiences with enforcement were not included in the model but may significantly influence farmers’ responses to forest policy. Future research could extend the framework by integrating these dimensions to provide a more comprehensive understanding of behavioral support for environmental policy innovations.

Fifth, the analysis focused primarily on individual-level perceptions and motivations, without explicitly modeling collective or multi-level influences. In rural and forest-dependent communities, social capital, power relations, and local governance structures often play a critical role in shaping participation. Future studies could adopt multilevel or mixed-methods approaches that combine individual survey data with community-level or institutional data to capture the interaction between personal cognition and broader governance dynamics. Finally, while Partial Least Squares Structural Equation Modeling was appropriate for the study’s exploratory and predictive objectives, alternative analytical approaches could offer complementary insights. Covariance-based SEM, longitudinal SEM, or experimental designs could be used in future research to test model stability, assess causal mechanisms, and compare competing theoretical frameworks. Acknowledging these limitations does not undermine the contribution of the present study; rather, it clarifies the scope of inference and highlights meaningful opportunities for future research. By addressing these limitations, subsequent studies can build on the integrated DOI–TPB framework to further advance understanding of behavioral support for sustainable forest governance under conditions of climate change.

Conclusion

This study explored the behavioral and perceptual factors associated with farmers’ intentions to support sustainable forest policy at the forest–agriculture interface by integrating the TPB and the DOI framework. The results indicate that farmers’ support intentions are shaped by a combination of psychological determinants and perceptions of policy innovation, highlighting the importance of considering both individual motivation and contextual policy characteristics when designing forest governance interventions. Among the innovation attributes examined, perceived compatibility, relative advantage, and observability were most strongly associated with higher levels of intention to support participatory forest governance. These findings suggest that farmers are more inclined to express support for forest policies when such policies are perceived as aligned with local livelihood systems, offering tangible benefits, and producing visible outcomes. In contrast, perceived complexity was negatively associated with intention, indicating that administrative or procedural difficulty may discourage supportive intentions even when policies are viewed as environmentally beneficial. Together, these results underscore that farmers’ willingness to engage with forest policy is closely linked to how innovations are perceived in terms of feasibility, relevance, and transparency.

From a behavioral perspective, all three components of the TPB were significantly associated with intention, with perceived behavioral control exerting the strongest influence. This finding highlights the importance of farmers’ confidence in their ability to participate and their access to necessary resources and institutional support. Social influences, reflected in subjective norms, also played a meaningful role, suggesting that community expectations and peer endorsement are important in shaping support for forest policy initiatives in rural contexts. The integrated DOI–TPB framework demonstrated substantial explanatory power, accounting for a considerable proportion of the variance in behavioral intention. Rather than implying causal relationships, the study provides explanatory evidence on how innovation perceptions and psychological factors interact to shape farmers’ stated intentions to support sustainable forest management under climate change. These insights contribute to a more nuanced understanding of policy acceptance in forest-dependent communities, where past experiences, livelihood constraints, and institutional conditions influence responses to conservation initiatives. From a practical perspective, the findings indicate that forest policies may gain greater local acceptance when they are designed and communicated in ways that emphasize compatibility with existing practices, clearly articulated benefits, visible outcomes, and manageable procedures. While the results do not predict actual behavior, they offer evidence-informed guidance for policymakers and practitioners seeking to improve the social acceptance of forest governance interventions in contexts similar to the Zagros region. Overall, this study highlights the value of integrating behavioral theory with innovation perspectives to better understand the factors associated with farmers’ support for participatory and climate-responsive forest policy.

Author contributions

R.M, M.P and F.H conceptualized the paper, R.M and M.P did data collection, R.M did data analysis and wrote the main manuscript, M.P and F.H reviewed and edited the paper. All authors reviewed the manuscript.

Data availability

Data will be available based on request from corresponding author.

Declarations

Competing interests

The authors declare no competing interests.