Mathematical model of neural network for the insect brain

Hidetoshi Ikeno

Goal of this study is to understand the functional mechanisms for sensory signal processing and behavioral control in a small brain of honeybee. Many investigations on honeybee behavior and neural mechanisms have been done during this century. Mathematical models for cellular, neural network and behavioral mechanisms will be presented in this study. These models will be implemented on the simulation server and can be accessed through the web browser.

We are constructing models of honeybee in harmony with bottom-up (cellar to network) and top-down (behavior to network) approaches.

1. behavioral changes of conditioning by using associated visual or olfactory stimulus would be a suitable paradigm to investigate the neural mechanisms for acquisition of sensory signals and controlling honeybee behavior. in recent years, ionic current properties of neurons in the insect brain have been measured by the patch-clamp technique on an isolated cell. on the other hand, spatio-temporal properties of neural network responses have also been shown by the optical recording and multi-channel recording methods. in order to reveal the mechanisms and functions of neuron and neural network of honeybee, we are integrating results and knowledge from these experiments, and building a mathematical model.
   29k PNG image Click to zoom up
   
2. In the hive, honeybee's behavior is changed by her age and and other sensory inputs. Worker, queen and drone's behaviors in the hive have been observed and recorded by single-individual trailing (SIT) method. Behavioral patterns of honeybee are categorized in about 150 based on the position and movement of body parts. We describe the behavioral sequence in SIT data as a mathematical model. It is expected to get the neural network model for generating honeybee behavior, because behavior would be reflecting the changing of activity in the nervous system.