CoChat

Artificial Astrocytes Improve Neural Network Performance

Open access

The DOI matches its Crossref record; 19 of the 19 references it cites were found in a registry or an index.

Abstract

Compelling evidence indicates the existence of bidirectional communication between astrocytes and neurons. Astrocytes, a type of glial cells classically considered to be passive supportive cells, have been recently demonstrated to be actively involved in the processing and regulation of synaptic information, suggesting that brain function arises from the activity of neuron-glia networks. However, the actual impact of astrocytes in neural network function is largely unknown and its application in artificial intelligence remains untested. We have investigated the consequences of including artificial astrocytes, which present the biologically defined properties involved in astrocyte-neuron communication, on artificial neural network performance. Using connectionist systems and evolutionary algorithms, we have compared the performance of artificial neural networks (NN) and artificial neuron-glia networks (NGN) to solve classification problems. We show that the degree of success of NGN is superior to NN. Analysis of performances of NN with different number of neurons or different architectures indicate that the effects of NGN cannot be accounted for an increased number of network elements, but rather they are specifically due to astrocytes. Furthermore, the relative efficacy of NGN vs. NN increases as the complexity of the network increases. These results indicate that artificial astrocytes improve neural network performance, and established the concept of Artificial Neuron-Glia Networks, which represents a novel concept in Artificial Intelligence with implications in computational science as well as in the understanding of brain function.

Overview

The claims and statistics of this paper haven’t been checked yet.

Free: it reads the open-access text and checks the first 5 claims that cite a source we can check. Create a free account to check every claim.

A few free checks a day · papers already verified open instantly · results are public

Integrity signals

Registry

Crossref · 10.1371/journal.pone.0019109

Cites

19 of 19 references verified0 citation errors · 0 not checkable

Statistics

Figures

Figure 1 Artificial astrocytes enhance neural network performance. (A) Schematic drawing representing the design of artificial neural networks without (left) and with artificial astrocytes (red stars; right) designed to solve the Ionosphere (IS) problem. (B) Representative example (left) and mean training accuracy (n = 100) vs. time for the (NN) and (NGN) solving the IS problem. (C) Representative example (left) and mean test accuracy (n = 100) vs. time for the NN and NGN solving the IS problem. (D) Mean steady training and test accuracies (left and right, respectively; n = 100) of NN and NGN solving the four problems tested. (E) Mean training and test times (left and right, respectively; n = 100) of NN and NGN solving the four problems tested. *P<0.05, **P<0.01 and ***P<0.001. Values represent mean ± S.E.M.
Figure 1 Artificial astrocytes enhance neural network performance. (A) Schematic drawing representing the design of artificial neural networks without (left) and with artificial astrocytes (red stars; right) designed to solve the Ionosphere (IS) problem. (B) Representative example (left) and mean training accuracy (n = 100) vs. time for the (NN) and (NGN) solving the IS problem. (C) Representative example (left) and mean test accuracy (n = 100) vs. time for the NN and NGN solving the IS problem. (D) Mean steady training and test accuracies (left and right, respectively; n = 100) of NN and NGN solving the four problems tested. (E) Mean training and test times (left and right, respectively; n = 100) of NN and NGN solving the four problems tested. *P<0.05, **P<0.01 and ***P<0.001. Values represent mean ± S.E.M.CC BY · PubMed Central
Figure 2 Neural network performance does not depend on the number of neurons or the architecture of the network. (A) Schematic drawing representing the design of three artificial neural networks with different number of neurons and different architectures. (B and C) Mean steady training and test accuracies, respectively (n = 100) of each NN for each problem tested. (D) Mean training and test times (left and right, respectively; n = 100) of each NN for each problem tested. *P<0.05, **P<0.01 and ***P<0.001. Values represent mean ± S.E.M.
Figure 2 Neural network performance does not depend on the number of neurons or the architecture of the network. (A) Schematic drawing representing the design of three artificial neural networks with different number of neurons and different architectures. (B and C) Mean steady training and test accuracies, respectively (n = 100) of each NN for each problem tested. (D) Mean training and test times (left and right, respectively; n = 100) of each NN for each problem tested. *P<0.05, **P<0.01 and ***P<0.001. Values represent mean ± S.E.M.CC BY · PubMed Central
Figure 3 Network performance improvement by artificial astrocytes increases as the network complexity increases. (A) Mean steady training and test accuracies (left and right, respectively; n = 100) of NN and NGN with 1, 2 or 3 hidden layers to solve the four problems tested. (B) Performance indexes (i.e., NGN values relative to NN values) of the training and test accuracies (left and right, respectively). Red symbols represent the corresponding averaged values (n = 16). *P<0.05, **P<0.01 and ***P<0.001. Values represent mean ± S.E.M.
Figure 3 Network performance improvement by artificial astrocytes increases as the network complexity increases. (A) Mean steady training and test accuracies (left and right, respectively; n = 100) of NN and NGN with 1, 2 or 3 hidden layers to solve the four problems tested. (B) Performance indexes (i.e., NGN values relative to NN values) of the training and test accuracies (left and right, respectively). Red symbols represent the corresponding averaged values (n = 16). *P<0.05, **P<0.01 and ***P<0.001. Values represent mean ± S.E.M.CC BY · PubMed Central
Figure 4 Relative network performance improvement by artificial astrocytes depends on the problem tested. (A) Mean steady training and test accuracies (left and right, respectively; n = 100) of NN and NGN with 1, 2 or 3 hidden layers to solve the four problems tested. (B) Performance indexes (i.e., NGN values relative to NN values) of the training and test accuracies (left and right, respectively). Red symbols represent the corresponding averaged values (n = 12). (C) Mean performance indexes of the training and test accuracies (left and right, respectively; n = 100) for each problem tested when artificial astrocytes were stimulated by different patterns of neuronal connection activity. The notation n,m indicates that artificial astrocytes were stimulated when the associated neuronal connections were active for at least n out of m iterations. (D) Mean performance indexes of the training and test accuracies (left and right, respectively; n = 100) for each problem of NGN with non-selected (black bars) or with specifically selected neuron-glia interaction parameters (red bars). *P<0.05, **P<0.01 and ***P<0.001. Values represent mean ± S.E.M.
Figure 4 Relative network performance improvement by artificial astrocytes depends on the problem tested. (A) Mean steady training and test accuracies (left and right, respectively; n = 100) of NN and NGN with 1, 2 or 3 hidden layers to solve the four problems tested. (B) Performance indexes (i.e., NGN values relative to NN values) of the training and test accuracies (left and right, respectively). Red symbols represent the corresponding averaged values (n = 12). (C) Mean performance indexes of the training and test accuracies (left and right, respectively; n = 100) for each problem tested when artificial astrocytes were stimulated by different patterns of neuronal connection activity. The notation n,m indicates that artificial astrocytes were stimulated when the associated neuronal connections were active for at least n out of m iterations. (D) Mean performance indexes of the training and test accuracies (left and right, respectively; n = 100) for each problem of NGN with non-selected (black bars) or with specifically selected neuron-glia interaction parameters (red bars). *P<0.05, **P<0.01 and ***P<0.001. Values represent mean ± S.E.M.CC BY · PubMed Central

Verification

Claims the paper makes

The claims and statistics of this paper haven’t been checked yet.

Free: it reads the open-access text and checks the first 5 claims that cite a source we can check. Create a free account to check every claim.

A few free checks a day · papers already verified open instantly · results are public

Claims other papers make about this one

Each is checked against this paper’s own text: does it say what the citing paper says it does?

Self-check

Percentages, sums and reported differences re-computed from the paper’s own numbers.

The self-check needs the paper’s full text.

Citations

Cites

Show all 19 references

Cited by

See all 89

Metadata

Authors
Porto-Pazos, A. B., Veiguela, N., Mesejo, P., Navarrete, M., Alvarellos, A., Ibáñez, O., Pazos, A., & Araque, A.
Affiliations
  • Universidade da Coruña (ES)
  • Consejo Superior de Investigaciones Científicas (ES)
Corresponding author
Porto-Pazos, Ana B.
Venue
PLoS ONE
Published
2011-04-19
Volume
6
Issue
4
Pages
e19109
ISSN
1932-6203 (ISSN-L)
OpenAlex
W1985943544
Cited by
89
References
19
Citations by year
2026: 4 · 2025: 3 · 2024: 3 · 2023: 7 · 2022: 3 · 2021: 2 · 2020: 6 · 2019: 13 · 2018: 6 · 2017: 5 · 2016: 8 · 2015: 11 · 2014: 7 · 2013: 4 · 2012: 6
FWCI
5.29
Type
journal article
Language
en
Open access
Locations
  • PLoS ONE (publishedVersion, cc-by)
  • Europe PMC (PubMed Central) (publishedVersion, cc-by)
  • PubMed Central (publishedVersion, cc-by)
  • PubMed (publishedVersion)
  • HAL (submittedVersion)
  • DIGITAL.CSIC (Spanish National Research Council (CSIC)) (submittedVersion)
  • DOAJ (DOAJ: Directory of Open Access Journals) (submittedVersion, cc-by-sa)
Funders
  • European Commission (202167)
  • Ministerio de Ciencia e Innovación
  • CYTED Ciencia y Tecnología para el Desarrollo (08SIN010105PR, 209RT0366, REGICC 2009/58)
  • Xunta de Galicia
  • Norwegian Biodiversity Information Centre
Topics
  • Physical Sciences › Computer Science › Artificial Intelligence › Neural Networks and Applications
  • Life Sciences › Neuroscience › Cognitive Neuroscience › Neural dynamics and brain function
  • Physical Sciences › Engineering › Electrical and Electronic Engineering › Advanced Memory and Neural Computing
Keywords
astrocyte-neuron communication, artificial neural networks, classification, network architecture, network complexity, evolutionary algorithms, connectionist networks, brain function
MeSH terms
Astrocytes, Neurons, Neural Networks, Computer, Artificial Cells

About these checks

Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.

Statistics and arithmetic are recomputed, not judged.

Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.

What counts as a claim ›How we measure confidence ›How we check this ›

Are you an author, or do you see a problem with this page? Request a review

How to cite

Cite CoChat for the checks on this page, and the paper for its own findings.

This page
CoChat. (2026). Citation check: Artificial Astrocytes Improve Neural Network Performance. https://app.cochat.ai/papers/10.1371/journal.pone.0019109
The paper
Porto-Pazos, A. B., Veiguela, N., Mesejo, P., Navarrete, M., Alvarellos, A., Ibáñez, O., Pazos, A., & Araque, A. (2011). Artificial Astrocytes Improve Neural Network Performance. PLoS ONE. https://doi.org/10.1371/journal.pone.0019109

Data sources

Bibliographic data: Semantic Scholar, OpenAlex, Crossref.

Contains information from the Semantic Scholar Academic Graph Datasets, made available under the ODC Attribution License (ODC-BY 1.0).

Semantic Scholar’s data is described in: Kinney, R., et al. (2023). The Semantic Scholar Open Data Platform. arXiv:2301.10140.