This paper presents an extension of Fuzzy Deformable Prototypes (FDPs) based on the use of interval type-2 fuzzy sets. The aim is to improve FDPs' capabilities for managing uncertainty and imprecision. This extension is applied to predict the academic performance of the students who make use of Web-based tutoring systems. The prediction model contains patterns of behavior that are used to determine the future academic performance of new students based on their affinity with the prototypes previously discovered. Interval Type 2 Fuzzy Sets (IT2FS) were used to handle the imprecision of the academic data caused by the overlapping between the fuzzy representations of prototypes.