Q1 — examination intelligence closed loop

Mathematics

1. Ŝ = αA + βT + γD + δR

   Predicted score from accuracy, time efficiency, difficulty solved, and revision consistency.

2. L(t) = L₀ (1 − e^{−kt})

   Learning curve: mastery approaches L₀ at rate k.

3. DNA_{t+1} = U(DNA_t, sitting_t)

   Knowledge DNA is rewritten after every sitting — not a progress bar.

4. u* = Pathway(DNA, goal)

   Next learn / practice unit is a function of mastery and exam goal.

5. t_response < 1.5 s

   Design latency bound from the monograph — not a measured SLA.

Instruction

1. 01Sit

   Student enters Experience: Atlas pathway, learn, practice, mock. LectureCognis is educator-only.

2. 02Observe

   Assessment writes (A, T, D, R) into the sitting log. This is the evidence vector.

3. 03Update

   DNA layer applies U(·). Sequence, difficulty, and revision intervals move with evidence.

4. 04Adapt

   Pathway returns u*. Assessment writes the next lesson; the lesson writes the next assessment.

5. 05Predict

   Score model Ŝ and rank band fire before the real sitting. Learn → Practice → Analyze → Adapt → Improve → Simulate.