Communication can improve or worsen group decision accuracy
When Does Communication Help? Beyond Spectral Descriptions of Collective Intelligence
Multiagent Systems
Summary
Sometimes, talking and sharing information helps a group make better decisions, but other times it can make their choices worse. The authors show that simple summaries of communication effects miss important details like the direction messages are sent. They developed a method that predicts when communication will help or harm accuracy, tested on both made-up and real data. Their work also finds that even if most people get better at a task, some groups can be hurt by the shared information.
What this means in practice
- •For distributed system engineers: Improve design of communication protocols by predicting when messaging enhances or harms accuracy in connected agents.
- •For machine learning developers: Use calibrated prediction methods to evaluate when collaborative inference improves performance on classification tasks like digit recognition.
Authors
Xuening Wu
Abstract
Communication can bring agents into agreement while making their decisions worse. We identify two limits of aggregate descriptions of communication gain in distributed inference. First, stable linear systems with fixed evidence, network and readout can have interaction and finite-time state operators with identical eigenvalue and singular-value spectra, yet produce gains of opposite sign. Changing only message orientation raises accuracy from 72.6% to 91.2% or lowers it to 65.9%. A standard task-projected local-response approximation retains the directional information missing from spectral summaries. Using labeled calibration data separate from the test set, it predicts multi-round gains in small trained nonlinear agents with a root-mean-square error of 0.45 percentage points on two synthetic tasks; tests with natural edge changes and handwritten digits extend the evaluation. Second, under community-shared bias, higher mean individual accuracy can coexist with harm to unaffected communities or lower global-vote accuracy. At fixed communication rounds, calibration constraints reduce observed community harm while retaining much of the mean benefit, but do not guarantee protection. Full direct calibration performs similarly. The results connect spectral insufficiency, task-aware prediction and the distribution of communication benefits, while leaving broad transfer and practical superiority open.