Combinatorial neural codes from a mathematical coding theory perspective

Carina Curto, Vladimir Itskov, Katherine Morrison, Zachary Roth, Judy L. Walker

Research output: Contribution to journalLetter

12 Citations (Scopus)

Abstract

Shannon's seminal 1948 work gave rise to two distinct areas of research: information theory and mathematical coding theory. While information theory has had a strong influence on theoretical neuroscience, ideas from mathematical coding theory have received considerably less attention. Here we take a new look at combinatorial neural codes from a mathematical coding theory perspective, examining the error correction capabilities of familiar receptive field codes (RF codes).We find, perhaps surprisingly, that the high levels of redundancy present in these codes do not support accurate error correction, although the error-correcting performance of receptive field codes catches up to that of random comparison codes when a small tolerance to error is introduced. However, receptive field codes are good at reflecting distances between represented stimuli, while the random comparison codes are not.We suggest that a compromise in errorcorrecting capability may be a necessary price to pay for a neural code whose structure serves not only error correction, but must also reflect relationships between stimuli.

Original languageEnglish (US)
Pages (from-to)1891-1925
Number of pages35
JournalNeural computation
Volume25
Issue number7
DOIs
StatePublished - Aug 7 2013

Fingerprint

Information Theory
Neurosciences
Research
Error Correction

All Science Journal Classification (ASJC) codes

  • Arts and Humanities (miscellaneous)
  • Cognitive Neuroscience

Cite this

Curto, Carina ; Itskov, Vladimir ; Morrison, Katherine ; Roth, Zachary ; Walker, Judy L. / Combinatorial neural codes from a mathematical coding theory perspective. In: Neural computation. 2013 ; Vol. 25, No. 7. pp. 1891-1925.
@article{e0bdac28f24e47b09d58bfe42cf1b94b,
title = "Combinatorial neural codes from a mathematical coding theory perspective",
abstract = "Shannon's seminal 1948 work gave rise to two distinct areas of research: information theory and mathematical coding theory. While information theory has had a strong influence on theoretical neuroscience, ideas from mathematical coding theory have received considerably less attention. Here we take a new look at combinatorial neural codes from a mathematical coding theory perspective, examining the error correction capabilities of familiar receptive field codes (RF codes).We find, perhaps surprisingly, that the high levels of redundancy present in these codes do not support accurate error correction, although the error-correcting performance of receptive field codes catches up to that of random comparison codes when a small tolerance to error is introduced. However, receptive field codes are good at reflecting distances between represented stimuli, while the random comparison codes are not.We suggest that a compromise in errorcorrecting capability may be a necessary price to pay for a neural code whose structure serves not only error correction, but must also reflect relationships between stimuli.",
author = "Carina Curto and Vladimir Itskov and Katherine Morrison and Zachary Roth and Walker, {Judy L.}",
year = "2013",
month = "8",
day = "7",
doi = "10.1162/NECO_a_00459",
language = "English (US)",
volume = "25",
pages = "1891--1925",
journal = "Neural Computation",
issn = "0899-7667",
publisher = "MIT Press Journals",
number = "7",

}

Combinatorial neural codes from a mathematical coding theory perspective. / Curto, Carina; Itskov, Vladimir; Morrison, Katherine; Roth, Zachary; Walker, Judy L.

In: Neural computation, Vol. 25, No. 7, 07.08.2013, p. 1891-1925.

Research output: Contribution to journalLetter

TY - JOUR

T1 - Combinatorial neural codes from a mathematical coding theory perspective

AU - Curto, Carina

AU - Itskov, Vladimir

AU - Morrison, Katherine

AU - Roth, Zachary

AU - Walker, Judy L.

PY - 2013/8/7

Y1 - 2013/8/7

N2 - Shannon's seminal 1948 work gave rise to two distinct areas of research: information theory and mathematical coding theory. While information theory has had a strong influence on theoretical neuroscience, ideas from mathematical coding theory have received considerably less attention. Here we take a new look at combinatorial neural codes from a mathematical coding theory perspective, examining the error correction capabilities of familiar receptive field codes (RF codes).We find, perhaps surprisingly, that the high levels of redundancy present in these codes do not support accurate error correction, although the error-correcting performance of receptive field codes catches up to that of random comparison codes when a small tolerance to error is introduced. However, receptive field codes are good at reflecting distances between represented stimuli, while the random comparison codes are not.We suggest that a compromise in errorcorrecting capability may be a necessary price to pay for a neural code whose structure serves not only error correction, but must also reflect relationships between stimuli.

AB - Shannon's seminal 1948 work gave rise to two distinct areas of research: information theory and mathematical coding theory. While information theory has had a strong influence on theoretical neuroscience, ideas from mathematical coding theory have received considerably less attention. Here we take a new look at combinatorial neural codes from a mathematical coding theory perspective, examining the error correction capabilities of familiar receptive field codes (RF codes).We find, perhaps surprisingly, that the high levels of redundancy present in these codes do not support accurate error correction, although the error-correcting performance of receptive field codes catches up to that of random comparison codes when a small tolerance to error is introduced. However, receptive field codes are good at reflecting distances between represented stimuli, while the random comparison codes are not.We suggest that a compromise in errorcorrecting capability may be a necessary price to pay for a neural code whose structure serves not only error correction, but must also reflect relationships between stimuli.

UR - http://www.scopus.com/inward/record.url?scp=84880976054&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84880976054&partnerID=8YFLogxK

U2 - 10.1162/NECO_a_00459

DO - 10.1162/NECO_a_00459

M3 - Letter

C2 - 23724797

AN - SCOPUS:84880976054

VL - 25

SP - 1891

EP - 1925

JO - Neural Computation

JF - Neural Computation

SN - 0899-7667

IS - 7

ER -