Collective Benefits and Efficiency in Coupled Echo State Networks
Abstract
Simple aggregation of independent responses ("wisdom of crowds") generally improves collective performance through statistical facilitation, but interaction among members can sometimes hinder group outcomes. However, under certain circumstances, coordination studies show that interacting groups can outperform their individual members and nominal groups, and interaction can also be more efficient than independence. We present a reservoir computing model to investigate how interaction can enhance joint performance in terms of accuracy and efficiency. Echo state networks were trained independently to identify Japanese speakers from vowel recordings and tested jointly with and without coupled feedback. Model pairs reproduced the dyadic advantage reported by Bahrami et al. (2010) with interaction via pooled-feedback, but not when independent responses were aggregated together. Model dynamics revealed that interaction was beneficial whereas aggregated outputs were prone to diverging from runaway feedback over time. In terms of efficiency, larger groups of smaller interacting models outperformed smaller groups of larger models while equating for total amount of group computation. Our results offer a formal framework for explaining when and why interaction enhances collective intelligence.
Related
- Neural networks reveal emergent properties of collective learning in democratic but not despotic groups
- Vimentin promotes collective cell migration through collagen networks via increased matrix remodeling and spheroid fluidity
- Fundamental Bounds on the Oscillation Efficiency of Markov Networks
- Efficiency and robustness partition the solution space for memory in recurrent networks
- Mixed-timestep spiking neural networks: a temporal alignment framework for high-efficiency neuromorphic computing
BibTeX
@misc{bruna2026collective,
title = {{Collective Benefits and Efficiency in Coupled Echo State Networks}},
author = {Polyphony Bruna and Rick Dale and Michael J. Spivey and Christopher T. Kello},
howpublished = {arXiv},
year = {2026},
month = oct,
doi = {10.48550/arxiv.2610.17506},
eprint = {2610.17506},
archivePrefix = {arXiv},
url = {https://arxiv.org/abs/2610.17506},
}
RIS
TY - UNPB
TI - Collective Benefits and Efficiency in Coupled Echo State Networks
AU - Polyphony Bruna
AU - Rick Dale
AU - Michael J. Spivey
AU - Christopher T. Kello
PY - 2026
DA - 2026/10/13
PB - arXiv
DO - 10.48550/arxiv.2610.17506
UR - https://arxiv.org/abs/2610.17506
AB - Simple aggregation of independent responses ("wisdom of crowds") generally improves collective performance through statistical facilitation, but interaction among members can sometimes hinder group outcomes. However, under certain circumstances, coordination studies show that interacting groups can outperform their individual members and nominal groups, and interaction can also be more efficient than independence. We present a reservoir computing model to investigate how interaction can enhance joint performance in terms of accuracy and efficiency. Echo state networks were trained independently to identify Japanese speakers from vowel recordings and tested jointly with and without coupled feedback. Model pairs reproduced the dyadic advantage reported by Bahrami et al. (2010) with interaction via pooled-feedback, but not when independent responses were aggregated together. Model dynamics revealed that interaction was beneficial whereas aggregated outputs were prone to diverging from runaway feedback over time. In terms of efficiency, larger groups of smaller interacting models outperformed smaller groups of larger models while equating for total amount of group computation. Our results offer a formal framework for explaining when and why interaction enhances collective intelligence.
ER -