Promoting climate resilience through learning-based behavioural change: Insights from an agent-based model of a coastal farming community in Guangxi, China

Individual climate-adaptive behaviour (CAB) plays an important role in strengthening local climate resilience, yet many farmers tend to form biased judgements about climate risks under uncertainty, which can hinder the uptake of CAB. To address this issue, this study develops a Learning–Protection Motivation Theory–Climate-Adaptive Behaviour model (LPMT-CAB) to simulate CAB decision-making under social and experiential learning interventions. Using household survey data from coastal Guangxi, China, the model simulates how learning mechanisms, including social learning (SL) and experiential learning (EL), shape CAB decisions over time. The model operationalizes PMT by linking threat appraisal, coping appraisal, and benefit-based behavioural adjustment to the dynamic adoption of CAB. The results indicate that learning strategies can significantly reduce farmers' risk perception biases and increase the adoption of CAB, particularly passive adaptive behaviours (PCAB), which require lower upfront costs, compared with no-intervention scenarios. SL accelerates early uptake and diffusion but may fall short in fostering long-term adaptive capacity, whereas EL strengthens persistence and stability by embedding adaptive practices through "learning by doing". These complementary approaches help balance short-term needs with a long-term perspective. This study provides theoretical insights for designing climate resilience education policies for smallholder farmers in similar socio-ecological settings and proposes a behavioural analysis modelling framework that can be extended to other contexts through context-specific calibration.