Computational thinking is a type of analytical thinking that is developed intentionally, that is, under a 'learned' approach, and the best way to learn it is through computer programming. In this paper, the experience of using block-based programming environments such as Scratch, Lightbot and the PSelnt pseudo-code tool for the understanding of computational concepts and development of computational thinking skills is reported. A quasi-experimental methodological design was used, since it did not work with a control group. A mixed exploratory investigation was carried out. 85 first-year students from the Accounting Professional School of the National University of San Agustin de Arequipa participated in the experience. The results show that in introductory programming courses of short time period, teaching with block-based programming environments compared to the use of pseudocode with the PSelnt tool, facilitates the appropriation of computational concepts and the development of computational thinking skills; thus obtaining better learning results. It is concluded that block-based programming environments favor the construction of knowledge and foster the development of creativity and algorithmic thinking.