The article discusses whether and to what extent experiments can contribute to a research paradigm based on the study of human behaviour in complex evolving environments and on the problem of asymmetric adjustment among different components of economic system along certain trajectories, focusing on the possibility that experimental evidence may represent an external consistency check on this type of heterodox modelling. It considers the evidence on rationality of human agents, and the possibility to identify a microfoundation alternative to homo oeconomicus, discussing the evidence on humans as strong reciprocators, as trusting individuals and as embedded in social norms.