# 1. SRS Configuration

User Type: **Super Administrator**
Source: *Mi Digital Academy - Education CRM Features Document*

---

## 1. SRS Configuration

### 1.1 Review Interval Settings
**What it does:** Configures the SRS review intervals: the optimal intervals for review (1 day, 3 days, 1 week, 2 weeks, 1 month), the interval progression per recall outcome, and per-content-type interval overrides.

**Sub-features:**
- Base review intervals: 1 day, 3 days, 1 week, 2 weeks, 1 month
- Interval progression mapping (recall outcome → next interval)
- Per-content-type interval overrides (flashcards, quiz questions, videos)
- Interval cap (maximum interval before graduation)
- Interval floor (minimum interval between reviews)
- Available on web
- Event logging (action performed)
- Audit logging of SRS interval configuration

**Super Admin User Journey:**
1. Super Admin opens SRS Management → Configuration and sets the base review intervals (1 day, 3 days, 1 week, 2 weeks, 1 month).
2. Maps the interval progression per recall outcome (e.g., "Hard" → 1 day, "Good" → 3 days, "Easy" → 1 week).
3. Sets per-content-type overrides where a content type needs different intervals.
4. Sets the interval cap and floor.
5. Saves; the intervals apply to all new SRS scheduling.
6. Each configuration change is event-logged and audit-logged.

**Rules & Edge Cases:**
- A recall outcome always maps to exactly one next interval; there is no ambiguous mapping.
- The interval floor prevents reviews closer together than the minimum; the cap prevents intervals beyond the maximum.
- A configuration change applies to newly scheduled reviews; already-scheduled reviews keep their interval until re-evaluated.
- Configuration changes are audit-logged with the setting, the change, and the timestamp.

### 1.2 Ebbinghaus Curve Parameters
**What it does:** Configures the Ebbinghaus forgetting curve parameters that drive the SRS: the memory decay rate, the initial retention target, and the curve shape parameters.

**Sub-features:**
- Memory decay rate parameter
- Initial retention target (e.g., 90% recall at first review)
- Curve shape parameters (decay exponent, stability factor)
- Per-grade-band parameter sets (age-appropriate decay)
- Parameter preview (predicted retention over time for a given setting)
- Available on web
- Event logging (action performed)
- Audit logging of Ebbinghaus curve parameters

**Super Admin User Journey:**
1. Super Admin opens SRS Management → Ebbinghaus Parameters.
2. Sets the memory decay rate and the initial retention target.
3. Adjusts the curve shape parameters (decay exponent, stability factor).
4. Defines per-grade-band parameter sets where decay differs by age.
5. Uses the parameter preview to see the predicted retention over time before saving.
6. Saves; the parameters drive the SRS scheduling.
7. Each configuration change is event-logged and audit-logged.

**Rules & Edge Cases:**
- Parameters must produce a monotonically decaying retention curve; invalid combinations are rejected with the reason.
- A per-grade-band parameter set overrides the global set for students in that band.
- The parameter preview is computed from the same model used for scheduling (no divergence).
- Configuration changes are audit-logged with the parameter, the change, and the timestamp.

### 1.3 Recall Success Adjustment Rules
**What it does:** Defines how recall success adjusts the SRS schedule: how a successful or failed recall moves the next review interval, the consecutive-success bonus, and the failure reset behavior.

**Sub-features:**
- Success adjustment: interval extension factor per successful recall
- Failure reset: interval reset behavior on a failed recall
- Consecutive-success bonus (accelerated progression)
- Confidence rating input (Too Easy / Good / Hard) mapping to adjustment
- Adjustment caps (max extension, min reset)
- Available on web
- Event logging (action performed)
- Audit logging of recall adjustment rules

**Super Admin User Journey:**
1. Super Admin opens SRS Management → Recall Adjustment Rules.
2. Sets the success adjustment (interval extension factor per successful recall).
3. Defines the failure reset behavior (e.g., reset to 1 day) and the consecutive-success bonus.
4. Maps the confidence ratings (Too Easy / Good / Hard) to the adjustment outcomes.
5. Sets the adjustment caps (max extension, min reset).
6. Saves; the rules apply to all recall events.
7. Each configuration change is event-logged and audit-logged.

**Rules & Edge Cases:**
- A failed recall always moves the item to an earlier review (never later); the reset is bounded by the min reset.
- The consecutive-success bonus accelerates progression but is bounded by the max extension and the interval cap.
- Each recall event produces exactly one schedule adjustment; there is no compounding within a single review.
- Configuration changes are audit-logged with the rule, the change, and the timestamp.