The main aim of this paper is to present results of an experimental test focused on the validity and effectiveness of composed methodology aimed at increasing the student's attention in Virtual Learning Environment. Areas of presented methodology which were subject of our research is students behavior during learning. The behavioral part of methodology is focused on calculating estimated time for reading/learning specific learning material based on mathematical model. The model has been created as linear regression model trained and validated with student's access logs from Moodle instance used at the university. The methodology created by us, which is the subject of the experiment, uses the mentioned mathematical model in order to more optimally calculate the time-on-task and thus assume the time required for the correct understanding of the presented curriculum in the form of e-learning. In this paper, we will focus on confirming the didactic effectiveness of our proposed methodology for monitoring student attention. We describe an experiment with which we verified the proposed methodology. Methodology itself is focused on improving students learning experiences while using e-learning platforms like Moodle.