According to transition science, system innovation requires experimentation and social learning to explore the potential of innovations for sustainable development. However, the transition science literature does not elaborate much on the learning processes involved. Senge's Field of Change provides a more detailed approach to the role of learning and action in innovation. We linked the Field of Change to transition management literature in order to explore social learning in an agricultural innovation experiment in the Netherlands called the ‘New Mixed Farm’.

More and more, development organizations are under pressure to demonstrate that their programs result in significant and lasting changes in the well-being of their intended beneficiaries. However, such "impacts" are often the product of a confluence of events for which no single agency or group of agencies can realistically claim full credit. As a result, assessing development impacts is problematic, yet many organizations continue to struggle to measure results far beyond the reach of their programs

he agricultural sector is under increasing pressure to bridge a growing concern for hunger and economic deprivation. At the centre of discussion is increase in agricultural productivity at a scale increasingly complex. This complexity challenges the capacity of both extension workers, farmers, farming systems and even the environment. This means that what matters for agricultural development and achieving the above situation is the capability of people to be effective and productive economic agents. It is here that capacity building comes in.

The LIVES project works to increase adoption of value chain interventions through use of improved knowledge and capacity by value chain actors and service providers. Knowledge management and capacity development are important components of the project to fill gaps in knowledge and capacity of value chain actors and service providers. Capacity is defined as the capabilities (knowledge, skills, experience, values, motivations, organizational processes, and linkages) that determine how well value chain actors and service providers utilize resources, market opportunities, and relationships.

The usual linear top-down approach of the innovation process has been transformed into an 'ecological vision' in which regions make up the appropriate scale for the elaboration of agricultural innovation systems (AIS). Interfaces such as institutional arrangements have been created in industrial countries. However, there is still a lack of interaction among parties involved in innovation for the agricultural sector, especially in the outermost regions and in developing countries.

In this paper the authors used a network perspective to study the micro level of agricultural innovation systems and investigate the different roles and functions that collaborating actors have to perform to spread their innovation both horizontally and vertically. Based on a literature review, we distinguish between three separate network functions: (1) learning and knowledge co-creation, (2) upscaling and institutional entrepreneurship and (3) outscaling and innovation brokerage.