Phoebe
Science Library

Phoebe's Neuro-Edge is peer-reviewed.

The aggregation, calibration, and deliberation mechanics that power the Intelligence Engine sit on a foundation of published cognitive science from CSO Bahador Bahrami and collaborators.

Four operating principles
01

Calibrated certainty over binary signal

Every prediction is graded on conviction. Forecasters who say 90% must be right 90% of the time — calibration becomes part of merit weighting.

02

Structured deliberation over unstructured crowds

Phoebe enforces blind and semi-blind phases inside social circles. Forecasters commit before peers reveal — neutralising bandwagon contagion.

03

Merit weighting over equal voice

Historical accuracy, calibration quality, and argument density compound into a per-forecaster weight applied to each contribution.

04

Atomic arguments over opaque scores

Every aggregated forecast is reconstructible from its atomic arguments — making Phoebe's outputs auditable in a way market scores never are.

Peer-reviewed papers
Paper · 01·Nature Human Behaviour·2018

Aggregated knowledge from a small number of debates outperforms the wisdom of large crowds.

A core finding underwriting Phoebe: small, structured debates aggregated to consensus systematically outperform large unstructured crowds. The cognitive bandwidth of deliberation beats the raw sample size of polling.

Navajas J, Niella T, Garbulsky G, Bahrami B, Sigman M. Nat Hum Behav 2, 126–132 (2018).
Paper · 02·JEPA·2022

Democratic forecast: Small groups predict the future better than individuals and crowds.

Replicates and extends the small-group advantage across forecast horizons: aggregated micro-deliberations beat both lone experts and unstructured crowds. The basis for Phoebe's circle topology.

Dezecache, G., Dockendorff, M., Ferreiro, D. N., Deroy, O., & Bahrami, B. (2022). Journal of Experimental Psychology: Applied, 28(3), 525–537.