The aims of the research is to determine whether there were significant differences in the solution of density problems between two groups of students: one group that is trained with a computing environment that approach the concept of density fragmented (disarticulation of the concept and approach decontextualized problems) and another group that is trained with a computing environment that approach the same concept but an integrated manner (integration of the concept and approach contextualized problems). The research is quasi-experimental, it was worked with two groups of seventh grade students, the equivalence between these two groups was studied through Student t-test, where it was found that there were no significant differences in their background. The research was conducted during two weeks. Initially, each of the groups, give them a guide formed for a theoretical part and other experimental. With the theoretical part we were looking to know the background and with the experimental part to propitiate cognitive conflict. When students complete the experimental part, they socialized the results. Subsequently, it proceeded to train each of these groups with the computing environment, after this stage of training, the teacher raise questions related with the concept developed during the session. Finally, was applied a written test consisting of 12 density problems, the operationalization of these results was made through evaluation criteria that were designed for each of the problem. The results show significant differences in the solution of density problems between two groups of students, these differences favouring students who were trained with the integrated computing environment; it was characterized by to approach then concept of density taking into account the relationship with mass, volume or temperature, also, it was addressed contextualized problems.