Order-Based Representation in Random Networks of Cortical Neurons
Author Information
Author(s): Shahaf Goded, Eytan Danny, Gal Asaf, Kermany Einat, Lyakhov Vladimir, Zrenner Christoph, Marom Shimon
Primary Institution: Technion—Israel Institute of Technology
Hypothesis
Is recruitment order applicable for representing stimuli that are not temporally ordered in complex large-scale recurrent neural networks?
Conclusion
Recruitment order is a reliable method for classifying stimuli in large-scale networks of cortical neurons, even with significant temporal changes in spike times.
Supporting Evidence
- Recruitment order is invariant to significant temporal changes in spike times.
- Classification accuracy increases with the number of sampled neurons.
- Recruitment order representation is lossless in terms of classification accuracy compared to absolute spike times.
Takeaway
The way neurons fire in order can help the brain understand what it sees, even if the timing of their firing changes a lot.
Methodology
The study used large-scale random networks of cortical neurons in vitro, stimulating them and recording their activity to analyze recruitment order.
Limitations
The applicability of the findings under in vivo constraints remains to be seen.
Participant Demographics
Cortical neurons obtained from newborn rats (Sprague-Dawley).
Statistical Information
P-Value
p<0.05
Statistical Significance
p<0.05
Digital Object Identifier (DOI)
Want to read the original?
Access the complete publication on the publisher's website