What question did this study set out to answer?

This work investigates whether internal feedback can be maintained in chaotic systems without collapses.

May 25, 2026Open Access

Self-referential feedback in chaotic reservoirs with preserved dynamics: a minimal architecture and the role of feedback type

View Full Paper

Ask AI

Bookmark

This work investigates whether internal feedback can be maintained in chaotic systems without collapses.
Developed a minimal architecture called new anima based on reservoir computing.
Utilized two specialized reservoirs with different spectral radii coupled to a dual-head self-model.
Conducted experiments on ten random seeds to analyze the effects of content and parametric feedback.
The content loop significantly increased prediction match from 0.33 to 0.66 (Cohen's d about 2.0, 10/10 seeds positive).
The Lyapunov exponent remained stable and positive at 0.020 ± 0.002 across different modes.
Ablation studies confirmed that the content loop is sufficient for maintaining self-consistency, while the parametric loop was less effective.

No takes yet. Share an insight, caveat, or question.

Randomized trial explores chaos and feedback's role in predictive coding dynamics, indicating significant effects on self-consistency.

Alexander Vasil’ev (2026) studied this question.

View Full Paper

Ask AI

Bookmark