Modelling agricultural innovations as a social-ecological phenomenon

Agricultural innovations involve both social and social-ecological dynamics where outcomes emerge from interactions of innovation actors embedded within their ecological environments. Neglecting the interconnected nature of social-ecological innovations can lead to a flawed understanding and assessment of innovations. In this paper, we present an empirically informed, stylized agent-based model of agricultural innovation systems in Mali, West Africa. The study aimed to understand the emergence of food security and income inequality outcomes through two distinct model structures: top-down, aid-driven (exogenous) innovation and bottom-up, community-driven (endogenous) innovation. Our research questions were: i) How does the inclusion of social-ecological interactions in the model affect food security and income inequality outcomes? ii) How do exogenous and endogenous mechanisms influence food insecurity and income inequality? iii) What are the conditions under which exogenous and endogenous mechanisms would improve food security? The structural design of the model was based on a combination of theory, empirics, and mapping of social-ecological dynamics within innovation systems. Using the Social-Ecological Action Situation framework, we mapped the social, social-ecological, and ecological interactions that jointly produce food security outcomes. The exploratory model analysis reveals three key insights: i) Incorporation of social-ecological interactions influences model outcomes. Scenarios with social-ecological interactions showed a stronger relationship between income inequality and food security, lower levels of food security, and higher levels of income inequality than scenarios with social interactions. ii) Endogenous mechanism leads to higher food security and income inequality than the exogenous mechanism. iii) Bidirectional outreach is more effective than unidirectional outreach in improving food security. Inclusion of social-ecological dynamics and interactions such as the role of climate risk perception, social learning and formation of innovation beliefs and desires is key for modelling and analysis of agricultural innovations.