Comparing decoding performance between functionally defined neural populations
2015Vol. 58, pp. 1–4
Citations Over Time
Abstract
Neurons in primary motor cortex can be divided into functional populations based on the width of their spike waveforms. These ensembles have different response properties that may subserve different roles in movement generation. Yet, how these differences impact offline decoding performance remains unknown. Here, we show that neurons exhibiting narrow spike waveforms outperform wide spiking neurons in decoding several features of movement. We further examine how decoding performance scales with respect to the number of neurons in the decoder, and show that an ensemble containing only narrow spiking units outperforms other models. These results suggest that it may be useful to consider spike waveform width when designing neural decoders.
Related Papers
- → Trainable Greedy Decoding for Neural Machine Translation(2017)55 cited
- → Model-Based Neural Decoding of Reaching Movements: A Maximum Likelihood Approach(2004)94 cited
- → Neural Decoding for Location of Macaque’s Moving Finger Using Generative Adversarial Networks(2018)3 cited
- → Comparing decoding performance between functionally defined neural populations(2015)1 cited
- → Trainable Greedy Decoding for Neural Machine Translation(2017)14 cited