There is a newer version of the record available.

Published March 25, 2026 | Version v1.0
Presentation Restricted

Towards a Metacognitive AI Architecture: A Cognition-Inspired Framework for Structured Reasoning and Cognitive Diversity

Description

This work proposes a metacognitive AI architecture designed to address key limitations in current AI systems, including weak higher-order knowledge synthesis, lack of iterative reasoning, poor contextual alignment, and absence of goal maintenance.

The framework introduces a two-layer cognitive architecture that builds on existing AI systems while adding a higher-order orchestration layer for structured reasoning, context monitoring, and adaptive problem-solving. Instead of scaling diversity through data alone, the approach engineers cognitive diversity through system design.

The proposed architecture draws inspiration from cognitive science and neuroscience, positioning intelligence as a dynamic, multi-factor system governed by orchestration rather than a single algorithm. The long-term vision is to enable structurally diverse, interoperable AI systems that support richer knowledge generation and true democratization of AI capabilities.

Files

Restricted

The record is publicly accessible, but files are restricted. Log in to check if you have access.

Additional details

Dates

Created
2026-03-25