Understanding Brain Robustness and Fault Tolerance
Author Information
Author(s): Srinivasan Shyam, Stevens Charles F
Primary Institution: University of California, Irvine
Hypothesis
Granger Causality analysis may fail to accurately infer causal connections in neural circuits due to the robustness and fault tolerance of brain networks.
Conclusion
The study reveals that Granger Causality can falsely indicate connections between neurons that do not actually exist.
Supporting Evidence
- Kispersky et al. found that Granger Causality analysis claimed a significant connection between neurons that do not communicate directly.
- The study showed that Granger Causality analysis can falsely identify causal connections in simple neural circuits.
- Neural networks are designed to maintain function even when parts are perturbed, complicating analysis.
Takeaway
Brains are built to keep working even when things go wrong, which makes it hard for scientists to figure out how they work. A method called Granger Causality can sometimes get it wrong and say two neurons are connected when they aren't.
Methodology
The authors reviewed the limitations of Granger Causality analysis in neural circuits, particularly in the context of a model for the pyloric ganglion.
Limitations
The analysis may not apply to all neural circuits, especially those with redundancy and fault tolerance.
Digital Object Identifier (DOI)
Want to read the original?
Access the complete publication on the publisher's website