With the goal of applying learning styles in both individual and collaborative learning, some adaptation mechanisms have been developed. These mechanisms try to improve the process of learning by matching the teaching style with the student?s learning style and by grouping students in some specific ways. We use the Felder-Silverman Learning Style Model and its Index of Learning Styles (ILS) questionnaire, in order to classify students, depending on their preferences on four dimensions (active/reflective, sensing/intuitive, visual/verbal, and sequential/global). The benefits of learning styles, it can be of interest for adaptive hypermedia learning systems in both individual and collaborative activities, especially if they support automatic grouping of users. From the results obtained by a case study with students of Computer Science in a collaborative task can be concluded that some dimensions of the learning style model seem to affect the quality of the resulting work. With this aim, new grouping rules have been incorporated in the TANGOW/WOTAN system to be used in the corresponding courses.