When facing a problem, it is necessary to go through two steps. First, understanding the problem, or creating its mental model. Second, solving the problem, or finding the best answer. In this article, the authors focus on problems which are finite dynamic interactions with perfect and complete information. They experimentally analyze the stage of understanding of such interactions.

Abstract
Dealing with a problem consists of understanding and solving. While there is vast literature analyzing problem solving, this article focuses on understanding. The problems subjects deal with are finite dynamic games with complete and perfect information. Understanding involves comprehending the relationships between choices and their consequences. The same backward-induction problem is presented using two distinct representations: tree and non-tree. The crucial difference between them is that the relevant connections between choices and consequences are directly provided in a tree but must be identified in a non-tree. Our study comprises 27 pairs of tree/non-tree interactions with a varying depth. We measure and analyze understanding by looking at success in understanding and the effort it involves. Our subjects struggle with understanding, even more so than solving. Greater depth increases the cognitive effort of understanding and, simultaneously, lowers the success rate in understanding.

Read the full article at Science Direct

Konrad Grabiszewski, Alex Horenstein, Understanding dynamic interactions, Games and Economic Behavior, Volume 149, 2025, Pages 96-111, ISSN 0899-8256, https://doi.org/10.1016/j.geb.2024.10.010.