Three levels.
One gesture:
intervene.
Criter Academy’s architecture brings together pedagogical archetype (macro), thirteen competencies (meso) and Intentional Labeling (micro). The macro emerges from the meso. The meso is exercised in the micro. And the other way around.
01the problem · ceibal diagnosis 2026
Why does this
framework exist?
framework exist?
Because teacher training in Latin America is broken — and the data says so. The Regional Diagnosis of Digital Transformation in Education by Fundación Ceibal and IDRC (2026) maps 9 pillars across 14 countries: teacher training is the lowest-rated of them all. Scaling up courses doesn’t solve it. We have to change the what and the how.
the study
Regional Diagnosis of Digital Transformation in Education
Fundación Ceibal · IDRC2026 · 14 Latin American countries9 maturity pillars
The study measures the digital maturity of education systems on a scale from 0 to 4, across 9 dimensions: infrastructure, connectivity, devices, content, platforms, data, regulation, management and teacher training.
The central finding: instrumental transformation is advancing (devices, connectivity), while pedagogical transformation lags behind. Teacher training comes in last. Mass courses don’t move the needle.
2.39
out of 4 — teacher training
The lowest-rated pillar in the maturity of Latin American education systems.
8/9
pillars at low levels
Only connectivity shows consistent regional progress. The rest is still lagging behind.
14
countries assessed
From Argentina to Mexico, the gap is structural — not specific to any one country.
the 9 pillars · digital maturity in Latin America
scale 0 — 4
Connectivity
3.10
Infrastructure
2.90
Devices
2.80
Platforms
2.70
Regulation
2.60
Digital content
2.50
Management
2.50
Data
2.40
Teacher training
2.39
average · 14 Latin American countries · ceibal–idrc, 2026
central thesis
“Teachers don’t consume a course: they intervene in it.”
the conceptual shift the model is built on
02architecture
Macro · Meso · Micro.
Three levels,
interlocked.
Three levels,
interlocked.
None collapses into another. The archetype can’t be reduced to a competency profile, competencies can’t be reduced to acts of labeling, and labeling can’t be reduced to a statistical proxy for the archetype. The connection runs both ways.
Level · Macro
Pedagogical archetype
who is the teacher?
An emerging pedagogical identity. A stable pattern of dominant competencies that defines the teacher’s predominant way of acting.
rationale
Teacher identity is not a static trait but an emerging pattern, reconfigurable through reflective practice and longitudinal feedback.
bibliographic anchors
2011Pratt, D. D. & Collins, J. B. — The Teaching Perspectives Inventory at 10 Years and 100,000 Respondents. Adult Education Quarterly, 61(4)↗2004Beijaard, D., Meijer, P. C. & Verloop, N. — Reconsidering research on teachers' professional identity. Teaching and Teacher Education, 20(2)↗2005Wiggins, G. & McTighe, J. — Understanding by Design (Expanded 2nd ed.). ASCD↗
Measurement · Initial SJT + longitudinal recalibration through notebook use.
Level · Meso
13 competencies
what does it activate in students?
Capacities the teacher activates — not ones they possess. Organized into 4 dimensions: Within, I Learn, I Build, I Transform.
rationale
Teacher competencies are assessed by their effect on the learner, not by self-report. This shift in focus comes from contemporary frameworks.
bibliographic anchors
2019OECD — OECD Learning Compass 2030. OECD Publishing↗2020CASEL — CASEL's SEL Framework. Collaborative for Academic, Social, and Emotional Learning↗2017Redecker, C. — European Framework for the Digital Competence of Educators: DigCompEdu. Joint Research Centre↗2006Mishra, P. & Koehler, M. J. — Technological pedagogical content knowledge: A framework for teacher knowledge. Teachers College Record, 108(6)↗
Measurement · Cumulative SJT scoring and intervention patterns.
Level · Micro
Intentional Labeling
how does it work day to day?
Categorized marking of the content + the teacher’s own note + an insight from the Signaler. Four categories: Recall, Clarify, Connect, Question.
rationale
To mark is to categorize; to categorize is to think. Experimental evidence shows active annotation outperforms passive highlighting.
bibliographic anchors
1940Adler, M. J. — How to Mark a Book. The Saturday Review of Literature↗1984Palincsar, A. S. & Brown, A. L. — Reciprocal teaching of comprehension-fostering activities. Cognition and Instruction, 1(2)↗2013Dunlosky, J. et al. — Improving students' learning with effective learning techniques. Psychological Science in the Public Interest, 14(1)↗2002Engle, R. A. & Conant, F. R. — Guiding principles for fostering productive disciplinary engagement. Cognition and Instruction, 20(4)↗
Measurement · Marking patterns, comments and mini-canvas in the notebook.
bottom-up
Micro → Meso → Macro
Every mark exercises competencies. As they accumulate, they form the stable pattern the system recognizes as an archetype.
top-down
Macro → Meso → Micro
The archetype shapes invitation post-its, suggested challenges and the language of questions. Guidance, not manipulation.
03why archetypes?
Modeling identities,
not boxes.
not boxes.
Archetyping is not a personality test — it is an evidence-based pedagogical tool. Connecting emotionally with a narrative profile increases engagement, transfer and the durability of learning. Every archetype in Criter has a reason to exist and a scientific rationale behind it.
01
Emotional engagement
Learning through a pedagogical character activates the brain’s narrative structures, which improves retention and a sense of purpose.
anchor · Bruner — Narrative Mode of Thought · Damasio · Egan
02
Grounded personalization
We adjust language, challenges and the order of content to each profile — no pseudo-AI. Differentiation rests on documented pedagogical patterns.
anchor · Pratt — Teaching Perspectives · Csikszentmihalyi — Flow
03
Identity emerges, it isn’t chosen
The archetype is not a self-report: it emerges from the pattern of answers to the SJT and is recalibrated as the notebook is used. Identity as a process, not a box.
anchor · Beijaard, Meijer & Verloop 2004 · Wenger
a concrete example · personalization in action
Two teachers, same notebook — different journeys.
archetype
Weaver
Opening
Opens with a question about connection: “Which student do you find hardest to reach?”
Tone · Challenges
Warm tone, anecdotes. Challenges in cooperative groups.
Sequence
Starts with the theory of the relationship · ends with the content
archetype
Architect
Opening
Opens with a diagram of the model: “We’re going to build this capability in 3 steps.”
Tone · Challenges
Structured tone, visual frameworks. Backward-design challenges.
Sequence
Starts with the structure · ends with practice
Same content, same rigor, same assessment. What changes is the doorway into learning.
04macro · identity · 9 archetypes
Nine pedagogical
archetypes.
archetypes.
Each archetype speaks to a specific body of research when read through its pedagogical core — not its name. Tap any archetype to see its typical radar and the three competencies that define it, based on the weights of the psychometric test.
Weaver
Core · bonds · community · belonging
Pratt — Nurturing · Noddings · Hattie
Explorer
Core · curiosity · wonder · expedition
Schwab — Inquiry · Kashdan — 5DC · Greene
Architect
Core · design · structure · scaffolding
Wiggins & McTighe — UbD · Bruner — scaffolding
Catalyst
Core · active silence · facilitation · agency
Rowe — Wait Time · Feuerstein — MLE
Narrator
Core · storytelling · meaning · resonance
Bruner — Narrative · Egan · Damasio
Alchemist
Core · transmutation · divergent thinking
Csikszentmihalyi · Robinson · de Bono
Guide
Core · autonomy · learning to learn
Knowles — andragogy · Zimmerman — SRL
Maker
Core · hands that think · prototyping · agency
Papert — Constructionism · Resnick · Dewey
Healer
Core · wellbeing as a condition of possibility
Noddings — Care · CASEL · Seligman
05meso · competencies × frameworks
Thirteen competencies.
Two anchors each.
Two anchors each.
Each competency rests on a double anchor: one framework describing the capability expected in the student, and another describing the corresponding capability of the teacher who activates it. The inversion is deliberate — these are capabilities the teacher activates, not ones they possess.
Within
the teacher builds inner life
A1
Wellbeing
a teacher who knows how to take care of themselves can take care of others — emotional self-regulation as a precondition.
+
A2
Self-confidence
acting from one’s own convictions without needing external validation.
+
A3
Curiosity
genuinely wondering about the content, the students and one’s own practice.
+
A4
Resilience
tolerating pressure and error without collapsing or denying.
+
I Learn
the teacher activates thinking
B1
Autonomy
designing so that students make real decisions about their learning.
+
B2
Judgment
building one’s own judgment from evidence before acting.
+
B3
Technological adaptability
integrating technology with pedagogical intent, not because it is trendy.
+
I Build
the teacher weaves community
C1
Bonds
cultivating genuine relationships of trust with students and peers.
+
C2
Collaboration
designing so that building together is a condition, not an option.
+
C3
Dialogue
creating conversation where every voice matters and is heard.
+
I Transform
the teacher looks outward to the world
D1
Purpose
connecting decisions to something that transcends the immediate situation.
+
D2
Agency
moving from knowledge to transformative action.
+
D3
Legacy
thinking about long-term impact — on students, on the community.
+
06micro · operation · intentional labeling
Built on 100 years
of research into
how we learn.
of research into
how we learn.
Mortimer Adler put it in 1940: “if you really read a book, you mark it.” His typology has survived 85 years because it maps cognitive functions that later research confirmed. To mark is to categorize, and to categorize is to think.
timeline · intentional labeling/20 milestones · 1940 — 2025
1940
foundation
Adler
How to Mark a Book
Defines the mark as the material trace of active thinking. 4 cognitive functions (agreement, disagreement, clarification, inquiry).
1956
categories
Bloom
Taxonomy of Educational Objectives
Systematizes a hierarchy of cognitive operations in learning. A framework for designing assessment.
1968
foundation
Ausubel
Meaningful Learning
New knowledge anchors onto prior structures. The mark activates that "anchoring" in cognition.
1972
encoding
Craik & Lockhart
Levels of Processing
Depth of processing (not repetition) determines retention. Marking is elaboration.
1979
metacog
Brown
Metacognition
Thinking about one’s own thinking can be taught. The mark as a metacognitive act.
1980
encoding
Wittrock
Generative Learning
Learning requires learners to generate relationships — not just receive. To connect is to learn.
1984
categories
Palincsar & Brown
Reciprocal Teaching
4 strategies: predict, question, clarify, summarize. Experimentally validated.
1986
practice
Rowe
Wait Time
Waiting 3 seconds after asking a question multiplies answer quality sevenfold. Silence = thinking.
1993
encoding
Marshall & Rossman
Procedural Signaling
Typographic markers direct cognitive attention. Underlining works as signaling.
1995
metacog
Schraw & Dennison
MAI · Metacognitive Awareness
A validated inventory of metacognitive awareness. Confirms it can be measured and trained.
2002
practice
Engle & Conant
Productive Disciplinary Engagement
Productive disagreement is structure, not a problem. Questioning as a legitimate cognitive category.
2006
practice
Mishra & Koehler
TPACK
Integrating technology requires specific pedagogical knowledge. A digital mark ≠ a paper mark.
2010
encoding
Roediger & Karpicke
Test-Enhanced Learning
Active recall beats rereading. The "Recall" mark is validated 70 years later.
2013
metacog
Dunlosky et al.
Effective Learning Techniques
Meta-analysis: passive highlighting has low utility. One’s own elaboration is decisive.
2016
practice
Alexander
Dialogic Teaching
Structured dialogue outperforms transmission. Mark → conversation → understanding.
2019
foundation
OECD
Learning Compass 2030
Agency, anticipation, reflection. Metacognition rises to the level of a transformative competency.
2020
encoding
Sweller et al.
Cognitive Load (4th decade)
Intrinsic + extraneous cognitive load define what can be learned. Marking reduces extraneous load.
2023
AI
Mosqueira-Rey et al.
Human-in-the-loop ML
Defines machine teaching: the human labels, the system learns. The architectural model behind the Signaler.
2024
encoding
Hattan et al.
Leveraging Prior Knowledge
Activating prior knowledge remains the strategy with the greatest effect on deep comprehension.
2025
AI
Frontiers Education
The Cognitive Paradox of AI
Receiving auto-generated answers reduces critical engagement. The fix is architecture, not good intentions.
scroll sideways →
criter’s 4 categories — a synthesis of the legacy
Recall
memory · importance · coming back
Adler 1940 — underlining · Marshall — procedural signaling
Clarify
confusion · lack of understanding
Palincsar & Brown — Reciprocal Teaching · clarifying
Connect
generativity · one’s own idea
Adler — inquiry · Hattan et al. — prior knowledge
Question
disagreement · cognitive restructuring
Adler — disagreement · Engle & Conant — productive disagreement
The teacher’s own note is the architectural keystone. On any highlight, the teacher adds a note — text or voice — explaining why they marked it. That turns the label into natural language and puts the system in a machine teaching regime, not an AI-led one. Human label first, human note second, the agent’s response third.
07signaler · dashboard
The Signaler is
a dashboard.
Not a chat.
a dashboard.
Not a chat.
A side panel built into the notebook that organizes insights about the marks the teacher has already made. It is not conversational. The interaction is exploratory: open, read, flag as useful, dismiss. The system responds only after the teacher has labeled a passage and written their note.
criteracademy.com / notebook · signaler
mockup · beta interface, May 2026
why a dashboard and not a chat
The cognitive paradox of AI.
When learners receive auto-generated answers and accept them passively, critical engagement goes down instead of up (Frontiers in Education, 2025). Chat is seductive — and that is exactly what makes it pedagogically dangerous.
Solved by architecture, not by good intentions. The dashboard requires you to have thought first: marking is a prerequisite for receiving insight. The system doesn’t start a conversation; it responds to a prior cognitive act.
grounded in:Frontiers Education 2025·Mosqueira-Rey 2023 · machine teaching
sequence · machine teaching
01Human label
the teacher marks with one of the 4 categories
02Human note
adds natural language about the mark
03Response
the Signaler returns a differentiated insight
the teacher is the operator, not the operated.