This index catalogs the 17 load-bearing learning theories woven throughout AI Thinking for Educators. Use it as a reference when designing lessons, evaluating AI tools, or tracing any pedagogical claim back to its theoretical source.
1Theory-to-Chapter Map at a Glance¶
| Chapter | Theories Integrated |
|---|---|
| 1 | Behaviorism · Constructivism (Piaget) · Desirable Difficulties (Bjork) · Productive Failure (Kapur) · Mindset (Dweck) |
| 2 | Cognitivism / Information Processing (Miller; Atkinson & Shiffrin) · Connectivism (Siemens; Downes) |
| 3 | Sociocultural / ZPD (Vygotsky) · Social Cognitive / Self-Efficacy (Bandura) |
| 4 | Cognitive Load (Sweller) · Multimedia Learning (Mayer) |
| 5 | Experiential Learning (Dewey; Kolb) |
| 6 | Andragogy (Knowles) · Self-Determination (Deci & Ryan) |
| 7 | Bloom’s Taxonomy (Bloom; Anderson & Krathwohl) · Metacognition (Flavell) · Fink’s Taxonomy |
| 8 | Transformative Learning (Mezirow) |
2Full 17-Theory Reference Table¶
| # | Theory | Year(s) | Theorist(s) | Core Idea | AI Connection | K-12 Artifacts Most Impacted |
|---|---|---|---|---|---|---|
| 1 | Behaviorism | 1898–1958 | Thorndike; Pavlov; Skinner | Learning is observable behavior change shaped by stimulus, response, and reinforcement. | Powers adaptive-learning systems, gamification loops, and RLHF — the dopamine economy of EdTech is behaviorist. | Grade books, behavior tracking logs, sticker/reward charts, classroom management plans, progress reports, report cards |
| 2 | Cognitivism / Information Processing | 1956; 1968 | Miller; Atkinson & Shiffrin | Mind as information processor with finite working memory (~4–7 chunks) and long-term store. | AI tutors decompose problems, manage context, and chunk content; students increasingly offload working memory to LLMs. | Worksheets, graphic organizers, study guides, anchor charts, note-taking templates |
| 3 | Constructivism | 1936–1970s | Piaget | Learners actively build knowledge by assimilation and accommodation into existing schemas through developmental stages. | AI should support messy schema-building, not deliver finished knowledge. | Inquiry-based lesson plans, hands-on activity guides, exploration journals, discovery-learning units |
| 4 | Sociocultural Theory / ZPD | 1920s–30s (pub. 1978) | Vygotsky | Learning is social; the ZPD is the gap between what a learner can do alone vs. with a More Knowledgeable Other. | AI is the first scalable MKO in history — the deepest theoretical case for AI tutoring. | Scaffolded lesson plans, differentiation plans, IEPs/504s, small-group instruction notes, peer-collaboration protocols |
| 5 | Experiential Learning | 1938; 1984 | Dewey; Kolb | Learning cycles through concrete experience → reflection → abstraction → active experimentation. | AI compresses the experimentation cycle from weeks to minutes. | Lab activities, field trip plans, project-based unit plans, reflection journals, permission slips |
| 6 | Bloom’s Taxonomy (revised) | 1956; 2001 | Bloom; Anderson & Krathwohl | Cognitive hierarchy: Remember → Understand → Apply → Analyze → Evaluate → Create. | AI now performs the bottom three cheaply — the central pedagogical disruption. Curriculum has to shift upward. | Learning objectives, test blueprints, formative & summative assessments, rubrics, standards-alignment docs |
| 7 | Cognitive Load Theory | 1988 | Sweller | Working memory is finite across intrinsic, extraneous, and germane load. | AI can strip extraneous load (good) or strip germane load students need for understanding (catastrophic). | Slide decks, worksheet design, instructional videos, scaffolded handouts, sub plans |
| 8 | Cognitive Theory of Multimedia Learning | 2001 | Mayer | Words + pictures beat either alone, but only under design principles (coherence, signaling, modality, redundancy). | AI-generated multimedia violates Mayer’s principles by default. | Slide decks, instructional videos, infographics, illustrated worksheets, parent newsletters, bulletin board materials |
| 9 | Andragogy | 1968; 1970 | Knowles | Adults need autonomy, draw on experience, are problem-centered, need immediate relevance. | Maps directly to adult-learner tutoring and professional development. | High-school career/college prep, dual-enrollment syllabi, PD materials, family literacy materials |
| 10 | Self-Determination Theory | 1985; 2000 | Deci & Ryan | Intrinsic motivation requires autonomy, competence, and relatedness. | AI as shortcut undermines all three at once; AI as coach amplifies all three. Same tool, opposite outcomes. | Student choice menus, goal-setting templates, conferring notes, engagement/motivation plans |
| 11 | Social Cognitive Theory / Self-Efficacy | 1977; 1986 | Bandura | We learn by observing others; belief in one’s capability drives effort and persistence. | AI as observation partner — but when AI does the work for students, self-efficacy erodes. | Modeling/think-aloud lesson plans, exemplars, mentor texts, student work portfolios, letters of recommendation |
| 12 | Metacognition | 1979 | Flavell | Thinking about thinking — planning, monitoring, and evaluating one’s own cognition. | Deciding when to use AI and judging its output is metacognition by another name. | Reflection prompts, exit tickets, self-assessment checklists, learning journals, error-analysis logs |
| 13 | Desirable Difficulties / Productive Failure | 1994; 2008 | Bjork; Kapur | Effortful, slow, well-calibrated learning consolidates; fluency without effort is shallow. | AI’s frictionless default is the opposite of what consolidates learning — the central tension of AI in education. | Spaced-practice schedules, retrieval-practice quizzes, productive-struggle problem sets, interleaved practice plans |
| 14 | Fink’s Taxonomy of Significant Learning | 2003 | L. Dee Fink | Six interactive, non-hierarchical categories: Foundational Knowledge, Application, Integration, Human Dimension, Caring, Learning How to Learn. | AI handles Foundational Knowledge competently but is weak on Human Dimension and Caring — the irreducible human-teacher zone. | Unit plans, course syllabi, integrative project rubrics, service-learning plans, SEL materials |
| 15 | Mindset | 1988; 2006 | Dweck | Belief that ability is fixed vs. growth shapes how learners respond to difficulty. | When everyone has the same AI tools, willingness to struggle becomes the differentiator — the mindset gap widens. | Feedback templates, growth-mindset language guides, error-celebration protocols, praise scripts |
| 16 | Transformative Learning | 1978; 1991 | Mezirow | Deep adult learning involves a disorienting dilemma that forces revision of frames of reference. | AI is itself the disorienting dilemma for most educators — the encounter is the curriculum. | PD reflection logs, equity/critical-thinking discussion protocols, journaling assignments, teacher observation responses |
| 17 | Connectivism | 2005 | Siemens; Downes | Knowledge lives in networks; learning is the capacity to traverse, curate, and form connections across them. | First learning theory written for the digital era — the most native fit for an AI-mediated world. | Curriculum maps, pacing guides, resource-curation lists, digital portfolio specs, research project frameworks |
3How to Use This Index¶
When designing a lesson or unit, start with the Theory-to-Chapter Map to identify which theoretical cluster governs the learning goal you are targeting. For example, if you are building a retrieval-practice quiz, you are operating squarely in the territory of Desirable Difficulties (Theory 13) and Cognitive Load (Theory 7) — both of which appear in Chapters 1 and 4 respectively. Use the K-12 Artifacts Most Impacted column as a practical checklist: if one of those artifact types appears in your lesson plan, the corresponding theory is already shaping your design decisions whether you have named it or not. Naming it gives you leverage.
The AI Connection column is where this index earns its place in an AI-focused book. Before selecting or deploying any AI tool in your classroom, cross-reference it against the theories most relevant to your lesson type. A tool that strips extraneous cognitive load (good, per Sweller) may simultaneously strip the germane load that builds durable understanding (catastrophic, per the same theory). A tool that delivers fluent, finished answers may satisfy Bloom’s lower rungs while actively undermining the productive struggle that Bjork and Kapur identify as essential for consolidation. These are not abstract concerns — they translate directly into the artifacts your students produce and the competencies they retain.
Finally, treat this index as a living reference, not a compliance checklist. The 17 theories here are not independent silos; they overlap, tension-test each other, and interact in every real classroom moment. Fink’s Human Dimension and Caring categories remind us that no theoretical framework fully captures what happens between a teacher and a student. Vygotsky’s MKO and Bandura’s self-efficacy research together explain why AI tutoring can either liberate or diminish a learner depending entirely on how the teacher frames the interaction. The goal of this index is not to reduce your practice to theory but to give your intuitions a sharper vocabulary — so that when AI enters the room, you can see exactly what is at stake.