• Top page
  • Timetable
  • Per session
  • Per presentation
  • How to
  • Meeting Planner



Neural Network Modeling

開催日 2014/9/13
時間 18:10 - 19:10
会場 Room G(303)
Chairperson(s) 酒井 裕 / Yutaka Sakai (玉川大学脳科学研究所 / Brain Science Institute, Tamagawa University, Japan)
濱口 航介 / Kosuke Hamaguchi (京都大学大学院医学研究科生体情報科学講座 / Department of Biological Sciences, Graduate School of Medicine, Kyoto University, Japan)

Computer simulation of superior colliculus dynamics using spiking neural circuit models

  • O3-G-2-3
  • Richard Veale:1,2 Tadashi Isa:2,3 Masatoshi Yoshida:2,3 
  • 1:Cognitive Science Program, Indiana University, Bloomington, USA 2:Dept. Developmental Physiology, National Institute for Physiological Sciences, Okazaki, Japan 3:School of Life Sciences, Graduate University for Advanced Studies, Hayama, Japan 

The superior colliculus (SC) is a layered midbrain structure with both visual (SGS) and motor layers (SGI) which are retinotopic and overlaid. The superior colliculus plays an important role in visual orienting behavior, and has been proposed as a central structure for saliency computation. We performed spiking neural circuit simulations of horizontal slices of SC based on previously reported data from acute slices of mouse SC (Phongphanphanee et al, 2014). The purpose of these simulations is to understand how variations of the spatial (axonal and dendritic) and temporal (synaptic dynamics) properties of inhibitory and excitatory neuron populations in the SC can lead to different dynamics and functions in different layers. The visual layers had different intrinsic lateral interactions and dynamics than the motor layers. Specifically, the visual layers had local excitation and distal inhibition (mexican hat). Furthermore, in the visual layers, additional inputs summed super-linearly when they were close to each other. In contrast, in the motor layers, excitation completely masked inhibition over the entire circuit, and multiple inputs summed linearly regardless of their distance from one another. We used differential evolution monte carlo to estimate the anatomical spreads and synaptic dynamics parameters of the inhibitory and excitatory neuron populations in each region. In conclusion, using only one inhibitory and excitatory neuron population, we were able to find good fits for both the visual and motor regions of SC. Future physiology and modelling studies will be able to identify the key circuit properties that differentiate the functions of the visual and the motor regions and lead to their particular contributions to attention and saliency computation in awake and behaving animals.

Copyright © Neuroscience2014. All Right Reserved.