In parallel with the popular development of opinion mining research, sorting methods know a renewed interest today. In these approaches a small sample of individuals sort a set of items in a pre-fixed or free set of categories and, roughly speaking, their aim is to better understand the perceived similarities and the subjectivity involved in the decisions. But, their analyzes only support the final rankings and do not take into account the informations associated with the construction dynamics of the ranking. In this communication, we present an original approach that explores the contribution to the analysis of the mouse trajectories and the associated events to the sorting task. An experiment is carried out within the framework of the Q-methodology which has been originally developed in psychology and which is now used in many domains. Our first results allow to identify different sorting strategies and atypical individual behaviors. We evaluate their roles in the construction of the final results based on a factorial approach.