Dept. of Mechanical Engineering and Robotics, Faculty of Textile, Shinshu University
Problem Solving and Autonomous Behavior in Pill Bugs
Tohru Moriyama
Ecological Psychology 16: 287-302, 2004.
This research demonstrates an aspect of problem solving with the creation of emergent (what I have called "autonomous") behaviors in pill bugs (Armadillidium vulgare, Isopoda, Crustacean). Two pill bugs were connected back-to-back by a string and left for hours. In this situation, movement of 1 individual (Individual 1) is transmitted by the string and becomes a vibration stimulus to the other (Individual 2). Then Individual 2 tries to escape, and its movement becomes another stimulus to Individual 1. In theory, this process continues, and these individuals cannot escape from the stimulus. In this problematic situation, "mounting" and "death-feigning" behaviors (not ordinarily observed) appeared. These behaviors seem to be emergent autonomous escaping behaviors from the inescapable vibration stimulus. The spatial distribution of pairs showing these behaviors, contrasted with pairs that did not show them, appeared to obey Zipf's law, a statistical property of natural languages.
Decision-making and Anticipation in Pill Bugs (Armadillidium vulgare)
Tohru Moriyama & Masao Migita
Computing Anticipatory Systems (Dubois D M ed.), American Institute of Physics, NY, pp459-464, 2004.
In the previous study, decision-making of the direction of motion was found in pill bugs. In the present experiment, we find that they autonomously choose specific places for the decision-making. Each individual was placed in a circle track surrounded by water. Small columnar marks were placed in the center of the track at equal intervals. At first, when they encountered the marks, they moved along and passed it. After some minutes, they tended to mount on the top of the marks, stayed for a while and swung the antennae as if they searched for the direction of motion. As time went on, they sometimes traced several marks. It seemed that they anticipated the appearance of the mark in advance to use it for decision-making of the direction of the next motion. In another circle track surrounded by wall, such behaviors were not observed.
Cognitive Model of Animal Behavior to Comprehend an Aspect of Decision-Making.
Masao Migita & Tohru Moriyama
Computing Anticipatory Systems (Dubois D M ed.), American Institute of Physics, NY, pp451-458, 2004.
Most animal behaviors are considered to have been evolved through their own courses of natural selection. Since mechanisms of natural selection depend tightly on environments in which animals of interest inhabit, the environment for an animal appears a priori, and stimulus-response (S-R) relationships are stable as long as it returns constant benefit. We claim, however, no environment for an animal cannot be regarded as a priori and any animal can exhibit more elaborated behavior than S-R. In other words, every animal is more or less cognitive in terms that it may modify a meaning of stimulus. We introduce a minimal model to demonstrate the cognitive aspect of the pill bug’s turn alternation (TA) behavior. The simulated pill bug can modify its own response pattern to the stimulus of water, though stable response appears to be prerequisite to TA behavior.
Decision-making and Turn Alternation in Pill Bugs (Armadillidium vulgare)
Tohru Moriyama
International Journal of Comparative Psychology 12: 153-170, 1999
Twelve pill bugs (Armadillidium vulgare, Isopoda, Crustacean) were examined in 200 successive T-mazes. When obstacles are present, the A. vulgare tend to move by means of turn alternation, which is generally considered an innate adaptive behavior. With a decrease in air moisture, the bugs have a tendency to increase their turn alternation rate. However, in such long successive T-mazes as in this study, continued turn alternation should actually accelerate the bugs' desiccation. This fact implies that turn alternation cannot always work adaptively. In this trade-off situation, while three individuals kept turn alternation at a high rate (1) and four at a low rate; (2), the other five spontaneously increased the rate of turn alternation and then decreased it (3). This instability of turn alternation in group (3) is interpreted as resulting not from stochastic factors but rather from the bugs' own decision-making, and seems to be an escape behavior used to get out of the experimental apparatus. In order to verify the decision-making hypothesis, all animals were subsequently tested in another successive T-maze apparatus, where the ends of the chosen alleys were shut, i.e., with 50 successive blind alleys.
Autonomous Learning in Maze Solution by Octopus
Tohru Moriyama & Yukio-Pegio Gunji
Ethology 103: 499-513, 1997
Tested 7 octopuses in maze-learning experiments. Results show Ss tried to reach the goal, so as to get a reward, by using various locomotory actions in the path, and sometimes encountered obstacles. Ss selected efficient swimming actions in the path; afterwards less efficient tactile actions gradually increased, while time to detour around the obstacle was reduced. To investigate whether octopuses reduce time spent detouring around obstacles by estimating their actions in the path, a trade-off situation in which octopuses were obliged to use tactile actions even though the set-up also encouraged them to use swimming actions was devised. As a result, it was observed that they reduced the detouring time. It is asserted that octopuses look around the whole maze and estimate their actions.
Failure in Anticipation and Plasticity in Perception of Taste
Tohru Moriyama, Sari Yokokawa, Yasuo Tsukahara
Computing Anticipatory Systems (Dubois D M ed.), American Institute of Physics, NY, pp480-487, 2006.
Experiments were designed to illustrate the plasticity of gustatory processing in human. In the matching experiment, forty-two subjects were respectively given twenty-four samples, each of which consisted of a usual pair of taste and visual stimuli (ex; a cup of apple juice covered with a lid having an apple painted on it). They were required to drink the taste stimulus while looking at the visual one. Subsequently, they were required to drink a cup of water covered with a lid with an Escher’s impossible-figure painted on it and asked whether the taste was the same as that of usual one or not. In the mismatching experiment, the same subjects were respectively given twenty samples each of which consisted of an unusual pair of taste and visual stimuli (ex; a cup of soup covered with a lid having an apple painted on it). They were required to drink the taste stimulus while looking at the visual one. Subsequently, they were required to drink a cup of water covered with a lid with an Escher’s impossible-figure painted on it and asked whether the taste was the same as that of usual one or not. Eight subjects sensed taste to water (0.05) after the mismatching experiment while four did after the matching one.
A Cognitive Experiment to Generate Jamais Vu Experience
-Investigation of Emergent Property of Human Cognition-
Tohru Moriyama, Shiori Watanabe, Yasuo Tsukahara
Proceedings of the 8th World Multiconference on Systemics, Cybernetics and Informatics vol.14: 9-13, 2004
A preliminary experiment to generate jamais vu experience was executed. In the first experiment, subjects were examined to solve a maze until they could almost correctly reach to the goal. The maze was displayed on the screen of a notebook computer although the whole paths were blacked out. The subjects could see a part of the paths though a small circle window. They could move it by using arrow keys of the keyboard. In the second experiment, some arrows were put on paths just before blanching points. Subjects were confused because arrows before some blanching points pointed to different directions from those of the correct paths. At the first trial of the experiment, they sometimes stood still on the path, however, from the second trial, they seldom did. In the third experiment, all arrows were cleared off. In this normal maze, while they could move on the correct route and reach to the goal efficiently, the frequency of standing still gradually increased. In this way, the aspect of jamais vu experience; the subjects think that he / she never been in the maze before although this is not true, was experimentally produced by making subjective confidence of the maze instable.
