Category representations can be broadly classified as containing within-category information or between-category information. Hélie, Shamloo, & Ell (2017) showed that regular classification learning instructions (A/B) promote between-category representation in rule-based categorization. whereas conceptual learning instructions (YES/NO) promote learning within-category representation with the same categories. This presentation explores how these task instructions affect brain activity using fMRI. Participants learned two sets of two categories and then transferred to comparing new sets of already learned categories. Computational models from Hélie et al. (2017) were fit to the behavioral data to determine the type of representation learned by each participant. fMRI contrasts were computed to compare BOLD signal between the two task instructions and between the two types of representation. The results show that participants learn different types of rules in the A/B and YES/NO tasks (univalent vs. bivalent) and that fMRI can be used to distinguish between these types of rule representation. More importantly, participants doing the same task but using different strategies can also be distinguished using fMRI.