Related Experiment Videos
Why does the single neuron activity change from trial to trial during sensory-motor task?
M Akamatsu1, A Terao, T Hasbroucq
1Neuroinformatics Laboratory, National Institute of Bioscience and Human Technology, MITI, Ibaraki, Japan. akamatsu@nibh.go.jp
Methods of Information in Medicine
|February 21, 1998
Summary
Neural network models reveal that parallel organization within units causes significant trial-to-trial variation in peak neural activity timing during sensory-motor tasks.
Area of Science:
- Neuroscience
- Computational Neuroscience
- Systems Neuroscience
Background:
- Single neuron recordings during sensory-motor tasks reveal trial-to-trial variability.
- Variability in neural activity timing exceeds behavioral reaction time variability.
Purpose of the Study:
- Investigate how neural connection structures influence peak activity timing variations.
- Model the effects of neural network architecture on neural response dynamics.
Main Methods:
- Recorded single neuron activities from monkey cortical areas during a choice reaction time task.
- Developed and simulated a multi-unit dynamic neural network model.
- Performed quantitative trial-by-trial analysis of neural data and model outputs.
Main Results:
- Peak neural activity timing showed substantial trial-to-trial variation.
- Model simulations demonstrated that parallel neural organization within units leads to wide variations in peak activity timing.
- Cascade connections alone did not fully explain the observed variability.
Conclusions:
- Parallel organization within neural network units is a key factor in generating significant trial-to-trial variability in neuronal firing.
- Understanding neural network structure is crucial for explaining neural response variability in sensory-motor tasks.