Related Experiment Video
Updated: Jun 8, 2025

Computational Modeling of Retinal Neurons for Visual Prosthesis Research - Fundamental Approaches
Published on: June 21, 2022
An end-to-end model of active electrosensation.
Denis Turcu1,2, Abigail Zadina1, L F Abbott1,2
1The Mortimer B. Zuckerman Mind, Brain and Behavior Institute, Department of Neuroscience, Columbia University, New York, New York, United States of America.
Weakly electric fish use self-generated electric fields to sense objects. A new model and artificial neural network (ANN) can accurately predict object location, size, and electrical properties from simulated sensory data.
Area of Science:
- Neuroscience
- Biophysics
- Computational Biology
Background:
- Weakly electric fish (WE fish) navigate and identify objects using their self-generated electric organ discharge (EOD).
- WE fish electroreceptors detect minute distortions in the EOD field caused by objects, enabling them to perceive object properties like resistance and capacitance.
- Object perception is challenging due to the high dependence of EOD distortions on object distance and size.
Purpose of the Study:
- To develop a computational model simulating WE fish electroreceptor responses and electric field distortions.
- To create large, artificial datasets of simulated fish-object interactions.
- To train an artificial neural network (ANN) to interpret these simulated sensory data for object recognition.
Main Methods:
- Constructed a biophysical model of electroreceptor responses based on experimental data.
- Developed a model of electric fields generated by WE fish and their distortions caused by objects with varying electrical properties.
- Generated extensive artificial datasets simulating WE fish encountering diverse objects.
- Trained an ANN to extract 3D location, size, and electrical properties (resistance, capacitance) from the simulated data.
Main Results:
- The ANN model achieved accurate extraction of object properties from simulated sensory data.
- A two-stage ANN processing approach, first estimating object distance/size and then electrical properties, yielded optimal performance.
- The results suggest a modular organization within the WE fish electrosensory system.
Conclusions:
- End-to-end modeling provides an effective method for studying sensory processing in WE fish.
- The findings support a hypothesis of modularity in the electrosensory system, with distinct stages for processing spatial and electrical information.
- This approach facilitates experimental testing of sensory processing mechanisms and highlights the potential of AI in neuroscience research.
Related Concept Videos
Somatosensation
Sensory Perception: Organization of the Somatosensory System
The receptor level:
The receptor level is the first stage of sensation. It involves the detection of a stimulus by specialized sensory receptors. The stimulus must arrive within the receptor's receptive field. Next, the receptor converts the energy of the...
Auditory Pathway
When viewed cross-sectionally, the cochlea reveals the scala vestibuli and scala tympani flanking...
The Cochlea
Sensory Functions of the Skin
There are two main categories of receptors on the skin: capsulated and non-capsulated. The non-capsulated ones are mainly the pain receptors. The capsulated ones can be further categorized based on the...
Hair Cells

