Validate content comprehensibility using calibrated Student Simulator prompt-personas (alto rendimiento, promedio, necesidades especiales) for age-appropriate readability. Trigger on: comprensibilidad, legibilidad, student simulator, carga cognitiva, validar contenido, readability, age-appropriate, pre-filtro, perfiles estudiante, simulacion lectora, accesibilidad contenido. Runs 3-profile simulation scoring legibility, cognitive load, and engagement per section with quantitative thresholds and fragment-level flags.
If a 10-year-old cannot understand it, the content has not done its job. Student Simulator personas detect comprehensibility problems before human review — catching dense sentences, unfamiliar vocabulary, and cognitive overload that authors are blind to because they already understand the material. Test with all three profiles; the "promedio" student is the gatekeeper.
Validate content comprehensibility by running calibrated Student Simulator prompt-personas against generated editorial content. Three profiles (alto rendimiento, promedio, necesidades especiales) simulate the reading experience of the target student population, scoring legibility, cognitive load, and engagement per section. The global score determines whether content advances to human review (Gate G2) or returns for revision.
references/priming-rag/rag-student-simulator-personas.md, because each profile is calibrated to a specific cognitive profile (reading speed, vocabulary range, attention span).| Name | Source | Required |
|---|---|---|
| Generated editorial unit | Output of generate-unit | Yes |
| Student Simulator profiles | references/priming-rag/rag-student-simulator-personas.md | Yes |
| Cognitive load reference | references/priming-rag/rag-carga-cognitiva.md | Yes |
| Target grado | Inferred from unit DZ codes or user request | Yes |
| Artifact | Format | Destination |
|---|---|---|
| Comprehensibility report | YAML | output/reporte-comprensibilidad-<UNIT-ID>.yaml |
| Flagged fragments list | Table within report | Embedded |
| Simplification suggestions | Per-fragment within report | Embedded |
| Decision (APROBADO/OBSERVACIONES/BLOQUEO) | Header field in report | Embedded |
reporte_comprensibilidad:
unidad: "U-LENG-5-01"
fecha: "2026-03-18"
grado: 5
asignatura: "LENG"
perfiles_evaluados: ["AR", "PR", "NEE"]
score_global: 87
decision: "APROBADO"
scores_por_seccion:
apertura:
legibilidad: 92
carga_cognitiva: "baja"
enganche: "alto"
desarrollo_tema1:
legibilidad: 84
carga_cognitiva: "media"
enganche: "medio"
desarrollo_tema2:
legibilidad: 88
carga_cognitiva: "media"
enganche: "medio-alto"
cierre:
legibilidad: 90
carga_cognitiva: "baja"
enganche: "alto"
evaluacion:
legibilidad: 85
carga_cognitiva: "media"
enganche: "medio"
scores_por_perfil:
AR:
comprensibilidad: 96
enganche: "alto"
carga: "baja"
PR:
comprensibilidad: 87
enganche: "medio-alto"
carga: "media"
NEE:
comprensibilidad: 74
enganche: "medio"
carga: "alta (sin adaptaciones), media (con DUA)"
fragmentos_problematicos:
- seccion: "desarrollo_tema1"
parrafo: 3
texto_extracto: "[first 10 words of problematic paragraph]..."
problema: "Oracion de 28 palabras con 2 clausulas subordinadas"
severidad: "media"
perfil_afectado: "PR, NEE"
sugerencia: "Dividir en 2 oraciones: separar clausula causal con punto seguido"
- seccion: "desarrollo_tema1"
parrafo: 5
texto_extracto: "[first 10 words]..."
problema: "3 conceptos nuevos en un solo parrafo sin definicion contextual"
severidad: "alta"
perfil_afectado: "NEE"
sugerencia: "Definir cada termino en primera aparicion. Distribuir en 2 parrafos."
resumen:
total_fragmentos_flagged: 2
severidad_alta: 1
severidad_media: 1
severidad_baja: 0
recomendacion: "Corregir fragmento de severidad alta antes de avanzar a Gate G2"
[INFERENCIA].references/priming-rag/rag-student-simulator-personas.mdreferences/priming-rag/rag-carga-cognitiva.mdreferences/priming-rag/rag-dua-inclusion.mdreferences/comprehensibility-metrics.md (local)