Neuronal tuning: To sharpen or broaden?

Kechen Zhang, Terrence J. Sejnowski

Research output: Contribution to journalArticle

Abstract

Sensory and motor variables are typically represented by a population of broadly tuned neurons. A coarser representation with broader tuning can often improve coding accuracy, but sometimes the accuracy may also improve with sharper tuning. The theoretical analysis here shows that the relationship between tuning width and accuracy depends crucially on the dimension of the encoded variable. A general rule is derived for how the Fisher information scales with the tuning width, regardless of the exact shape of the tuning function, the probability distribution of spikes, and allowing some correlated noise between neurons. These results demonstrate a universal dimensionality effect in neural population coding.

Original languageEnglish (US)
Pages (from-to)75-84
Number of pages10
JournalNeural Computation
Volume11
Issue number1
StatePublished - Jan 1 1999
Externally publishedYes

Fingerprint

Tuning
Neurons
Population
Noise
Probability distributions
Neuron

ASJC Scopus subject areas

  • Artificial Intelligence
  • Control and Systems Engineering
  • Neuroscience(all)

Cite this

Zhang, K., & Sejnowski, T. J. (1999). Neuronal tuning: To sharpen or broaden? Neural Computation, 11(1), 75-84.

Neuronal tuning : To sharpen or broaden? / Zhang, Kechen; Sejnowski, Terrence J.

In: Neural Computation, Vol. 11, No. 1, 01.01.1999, p. 75-84.

Research output: Contribution to journalArticle

Zhang, K & Sejnowski, TJ 1999, 'Neuronal tuning: To sharpen or broaden?', Neural Computation, vol. 11, no. 1, pp. 75-84.
Zhang, Kechen ; Sejnowski, Terrence J. / Neuronal tuning : To sharpen or broaden?. In: Neural Computation. 1999 ; Vol. 11, No. 1. pp. 75-84.
@article{5126fcdba1014345b9eca33bc7e25dfc,
title = "Neuronal tuning: To sharpen or broaden?",
abstract = "Sensory and motor variables are typically represented by a population of broadly tuned neurons. A coarser representation with broader tuning can often improve coding accuracy, but sometimes the accuracy may also improve with sharper tuning. The theoretical analysis here shows that the relationship between tuning width and accuracy depends crucially on the dimension of the encoded variable. A general rule is derived for how the Fisher information scales with the tuning width, regardless of the exact shape of the tuning function, the probability distribution of spikes, and allowing some correlated noise between neurons. These results demonstrate a universal dimensionality effect in neural population coding.",
author = "Kechen Zhang and Sejnowski, {Terrence J.}",
year = "1999",
month = "1",
day = "1",
language = "English (US)",
volume = "11",
pages = "75--84",
journal = "Neural Computation",
issn = "0899-7667",
publisher = "MIT Press Journals",
number = "1",

}

TY - JOUR

T1 - Neuronal tuning

T2 - To sharpen or broaden?

AU - Zhang, Kechen

AU - Sejnowski, Terrence J.

PY - 1999/1/1

Y1 - 1999/1/1

N2 - Sensory and motor variables are typically represented by a population of broadly tuned neurons. A coarser representation with broader tuning can often improve coding accuracy, but sometimes the accuracy may also improve with sharper tuning. The theoretical analysis here shows that the relationship between tuning width and accuracy depends crucially on the dimension of the encoded variable. A general rule is derived for how the Fisher information scales with the tuning width, regardless of the exact shape of the tuning function, the probability distribution of spikes, and allowing some correlated noise between neurons. These results demonstrate a universal dimensionality effect in neural population coding.

AB - Sensory and motor variables are typically represented by a population of broadly tuned neurons. A coarser representation with broader tuning can often improve coding accuracy, but sometimes the accuracy may also improve with sharper tuning. The theoretical analysis here shows that the relationship between tuning width and accuracy depends crucially on the dimension of the encoded variable. A general rule is derived for how the Fisher information scales with the tuning width, regardless of the exact shape of the tuning function, the probability distribution of spikes, and allowing some correlated noise between neurons. These results demonstrate a universal dimensionality effect in neural population coding.

UR - http://www.scopus.com/inward/record.url?scp=0032611075&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=0032611075&partnerID=8YFLogxK

M3 - Article

C2 - 9950722

AN - SCOPUS:0032611075

VL - 11

SP - 75

EP - 84

JO - Neural Computation

JF - Neural Computation

SN - 0899-7667

IS - 1

ER -