01
SciLoop Flow
Structure the learning loop
result
Open portal Discovery + Feedback Loop
14 nodes · 14 connections SCILOOP / UNIFIED CONCLUSION
From four engines to one understanding system.
This is the main SciLoop surface. Each portal contributes one job, then the synthesis layer compresses the result into an implementation decision you can act on.
Current learner signal
Ready for the first learner test
Main conclusion
Build around prediction before explanation.
The strongest common result is a prediction-first, feedback-driven loop: give the learner a concrete experience, require a guess, show the consequence, expose the cause, and let the next challenge adapt to the evidence.
What to implementOne shared learning grammar
Experience → Prediction → Simulation → Observation → Rule → Transfer
What must adaptThe next experience
Errors request contrast and explanation; successful predictions earn a harder or broader test.
Shared basis
ExperiencePredictionStateRelationshipTimeCausalityErrorTransfer
These are the smallest practical pieces shared by the four portals. Subject matter can change; the learning grammar stays stable.
The four jobs
Each portal contributes one layer
02
Cognitive Lab
Test whether the flow supports understanding
result
Open portal Prediction → Simulation → Observation → Feedback
reasoning + transfer protocol03
Visual Engine
Turn meaning into visible structure
result
Open portal Object · Change · Relationship · Time · Error
shared visual basis04
Evolving Engine
Use learner evidence to change the next test
result
Open portal Contrast after error · challenge after success
local learner memory Implementation decision
1. Start with a world
Concrete objects and visible changes come before equations or terminology.
2. Ask for a commitment
The learner must predict before the system explains.
3. End with transfer
The final test asks whether the learner can use the rule somewhere new.