A model for the neuronal substrate of dead reckoning and memory in arthropods: a comparative computational and behavioral study

• Published in: Theory in biosciences = Theorie in den Biowissenschaften Vol. 127; no. 2; pp. 163 - 175
• Main Authors: Bernardet, Ulysses, Bermúdez i Badia, Sergi, Verschure, Paul FMJ
• Format: Journal Article
• Language: English
• Published: Berlin/Heidelberg Springer-Verlag 01.06.2008; Springer Nature B.V
• Subjects: Animals; Arthropods - physiology; Behavior; Behavior, Animal - physiology; Bioinformatics; Biology; Biomedical and Life Sciences; Complex Systems; Computational neuroscience; Computer Simulation; Decision making; Evolutionary Biology; Information processing; Integration; Life Sciences; Locomotion - physiology; Mathematical and Computational Biology; Memory - physiology; Models, Neurological; Nerve Net - physiology; Neurons - physiology; Orientation - physiology; Original Paper; Philosophy of Biology; Robotics - methods; Theoretical Ecology/Statistics; Dead reckoning; Population code; Arthropod; Neuronal model; Path integration; Vector decoding; Path memory
• ISSN: 1431-7613; 1611-7530
• DOI: 10.1007/s12064-008-0038-8

Returning to the point of departure after exploring the environment is a key capability for most animals. In the absence of landmarks, this task will be solved by integrating direction and distance traveled over time. This is referred to as path integration or dead reckoning. An important question is how the nervous systems of navigating animals such as the 1 mm 3 brain of ants can integrate local information in order to make global decision. In this article we propose a neurobiologically plausible system of storing and retrieving direction and distance information. The path memory of our model builds on the well established concept of population codes, moreover our system does not rely on trigonometric functions or other complex non-linear operations such as multiplication, but only uses biologically plausible operations such as integration and thresholding. We test our model in two paradigms; in the first paradigm the system receives input from a simulated compass, in the second paradigm, the model is tested against behavioral data recorded from 17 ants. We were able to show that our path memory system was able to reliably encode and compute the angle of the vector pointing to the start location, and that the system stores the total length of the trajectory in a dependable way. From the structure and behavior of our model, we derive testable predictions both at the level of observable behavior as well as on the anatomy and physiology of its underlying neuronal substrate.