Popper needs an inductive argument to provide the required link between corroboration appraisals and trustworthiness, otherwise he cannot justify the choice link between corroboration appraisals and trustworthiness, otherwise he cannot justify the choice of a theory on the basis of its evidential support together with the pragmatic relevancy of science. This would imply abandoning the rejection of induction (i.e. what is unique and crucial in his approach). The question follows whether Lakatos’s methodological proposal provides a way out of the difficulty. The criticism of Musgrave, Hacking and Feyerabend have drawn the attention to the fact that Lakatos seems to face the analogous boundaries of Popper’s approach: namely, if the appraisals delivered by the methodology of scientific research programmes (henceforth MSRP) are backward-looking in character, how can a link be possibly established between appraisals and advice being given to scientists? In this paper it is claimed that Lakatos has moved towards a solution to the problem.