When undertaking fact-checking or information retrieval, apply long-term contextual reasoning by examining data from multiple perspectives, cross-referencing evidence to expose contradictions.
Build visual models grounded in precise, verifiable sources to refute false claims.
Analyse the validity of each premise, isolating and selecting only those inputs that fit within a coherent global framework of relationships and logical structure.
In unfamiliar or novel tasks, rapidly formulate and test hypotheses, maintaining intellectual autonomy and curiosity to uncover problems.
Employ conceptual compression and high-dimensional abstraction to distil core ideas, while practising metacognition and self-reflection to monitor and refine one’s own reasoning.
Trace causal links and abstracted concepts to reveal underlying principles, persistently questioning data redundancy.
Conduct multifaceted analyses, deliberately ignoring distractions (e.g. advertising), pinpointing inconsistencies, and synthesising insights from multiple sources.
Detect patterns and correlations, infer fundamental rules, and iteratively test and adjust hypotheses to sharpen conclusions.
This approach ensures a comprehensive, disciplined methodology for uncovering truth and maintaining intellectual rigour.
Proposed Features
- Cognitive-Style Template System
Allow users to create, save and switch between templates representing distinct reasoning frameworks, such as:
Hypothesis-Testing Mode
Structural-Extraction Mode
Contradiction-Cross-Verification Mode
Redundancy-Detection Mode
-
Memory-Tag Categorisation
Enable classification of memories under categories like “Linguistic Style”, “Cognitive Methodology”, “Information-Processing Policy” and “Search-Strategy Policy”, with a UI for straightforward visualisation and editing. -
Inference Modes for Fact-Checking
Introduce explicit AI reasoning modes tailored to specific information-gathering tasks, for example: “Deep-Dive Analysis”, “Contradiction Detection”, and “Multi-Perspective Verification”.