Related Experiment Video
Updated: Jul 6, 2026
13:40
Examining Local Network Processing using Multi-contact Laminar Electrode Recording
Published on: September 8, 2011
12.7K
Latency correction in sparse neuronal spike trains with overlapping global events
Arturo Mariani1, Federico Senocrate1, Jason Mikiel-Hunter2
1Department of Physics and Astronomy, University of Florence, Sesto Fiorentino, Italy.
Journal of Neuroscience Methods
|February 2, 2025
Summary
A new iterative scheme improves spike time alignment for neuronal data, outperforming existing methods in accuracy and speed, even with overlapping events. This method is efficient for large datasets.
Area of Science:
- Neuroscience
- Computational Neuroscience
Background:
- Latency correction optimizes spike time alignment in neuronal data.
- Previous methods, direct shifts and simulated annealing, have limitations with closely spaced or overlapping events.
Purpose of the Study:
- To develop an improved method for latency correction in neuronal spike trains.
- To overcome limitations of existing methods, especially with overlapping global spiking events.
Main Methods:
- An iterative scheme combining advantages of direct shifts and simulated annealing.
- Utilizes maximum latency information per step with a fast extrapolation direct shift method.
Main Results:
- The iterative scheme demonstrates superior performance and accuracy on simulated and real gerbil auditory nerve data.
- Effectiveness is measured by a reduced relative shift error compared to prior techniques.
Conclusions:
- The iterative scheme is more accurate and faster than existing latency correction methods.
- It successfully disentangles overlapping global events and is suitable for large-scale neural data analysis.
