The mathematisation of economics took place in three different ways, each based on an image of mathematics that was endorsed by a practice of mathematisation and which remained the dominant image until another image became dominant. The first dominant image was that of mathematics as a science of forms, which was embraced in the 1920s by the early econometric programme of scientisation. The resulting methodology was the modelling style developed at MIT in the 1940s and 1950s. The next dominant image was that of mathematics as a science of structure. It was introduced into economics by von Neumann and opened the door for Bourbaki mathematics which set the standards for mathematical rigour. The image of algorithms entered economics through research into decision making under uncertainty. By turning the decisionmaker into a 'representative agent' who takes decisions at a macro-level, this approach reshaped macroeconomic analysis in the 1970s and 1980s.
