Phone No.
(+352) 247-85187
Email us
secretariat@script.lu
Address
33, Rives de Clausen
L-2165 Luxembourg

www.bibliotc.lu

Travaux de candidature

Bibliothèque en ligne

jeudi 9 août 2018 / Catégories: Informatique

Collaborative problem solving: Reducing trial and error in a Tangible User Interface (TUI)-based scenario in order to increase the learning effect

by Lahure Cathia

In our rapidly changing society, education is searching for new methods and new tools to reach those students who have difficulties in the traditional system.
At the LIST, the Luxembourg Institute of Science and Technologies, such methods are developed around tangible tables. In the context of the Erasmus project RE-ENGAGE, teachers devised scenarios containing a problem to be solved by students. The images of the scenario are projected onto the surface of the table. Learners interact with the scenario by moving physical objects on the table’s surface.
If the learners use a pure trial-and-error approach to get to an expected result, the problem-solving scenarios do not always reach their best learning effect. This project aims to find a way to reduce the resolution of a given problem by trial and error to get the learners to reflect more on the situation and understand its result.
As a first step, I designed and implemented scenarios. After an exploratory development phase, I used iterative development to create a scenario for a case study.
The case study revealed trial-and-error behaviour by test users. Based on the analysis results, I implemented a design solution for reducing trial and error. In the redesigned scenario, output values remained hidden unless a button was pressed, and a score was awarded more marks if the problem was solved in less tries.
In a comparative study, both scenarios have been evaluated. Groups of students tested one of the two scenarios.
The comparative study concludes that the design solution for less trial and error caused a higher workload, a longer reflection, but did not generate a higher learning effect. I justify this result by the high complexity of the scenario.