Lahure, CathiaMaquil, Valérie2018-11-152018-11-152018https://dl.gi.de/handle/20.500.12116/18234This paper describes the results from a comparative study with 14 pupils using two different versions of a tangible tabletop application on satellite communication. While one of the versions was designed in a way to allow the resolution of the tasks in a pure trial-and-error approach, the second version prevented this by adding a button which had to be pressed in order to calculate and display results. The results of the study show that the design of the button and the associated scoring system was indeed successful in slowing down interactions and increasing thinking time. However, the knowledge acquisition was lower for the version with the button as compared to the one supporting trial-and-error. We discuss the results of this study and, in particular, argue for the need to carefully balance usability, task complexity and the learning dimension in the design of interactive tabletops for learning.entangible user interfacestabletop interfacesuser studytrial-and-errorcollaborative problem solvinglearning-by-doingText/Journal Article10.1515/icom-2018-00221618-162X