Ethological dynamics in diorama environments

The model organism, a plasmodium of the true slime mold, is a multinucleated unicellular organism. Despite its simple structure, primitive but "intelligence"-like phenomena have been found, such as solving mazes and learning by habituation and associative memory. Although there are several hypotheses on the learning mechanism in slime mold, such as the mathematical hypothesis based on the oscillator system and the dynamical system, and the hypothesis based on the molecular mechanism of the cell, there is currently no hypothesis that provides a unified explanation for the various input stimuli. We hypothesize that the mechanism of learning lies in the network structure of the transport tubes formed in the slime molds. In this study, we investigate this mechanism by quantitatively analyzing the changes in the structure of the network of transport tubes observed during the learning process in slime molds placed in a diorama environment. Based on this, an algorithm for learning is extracted by constructing a mathematical model of the formation of the transport tube network. Through comparison with the structure of neural networks, we aim to find the algorithm of the most primitive "intelligence”.