Best AI for Mathematics and Problem-Solving Education in 2026

Quick Answer: AI for mathematics and problem-solving generates Polya four-phase problem-solving coaching sequences (understand the problem; devise a plan; carry out the plan; look back) with Schoenfeld metacognitive prompts at each phase; NCTM five-process-standards-aligned lesson designs that develop problem solving, reasoning, communication, connections, and representation simultaneously; Boaler low-floor high-ceiling open mathematical tasks accessible to all learners; Van Hiele level-targeted geometry investigations; Lesh-Doerr model-eliciting activities in real-world contexts; and mathematical discourse facilitation tools for classroom discussions. EduGenius (edugenius.app) supports mathematical problem-solving development from Grade KG-9 with research-grounded, thinking-centered mathematics tools.

Mathematics is, in a fundamental sense, the science of patterns — the discipline through which humans develop and use rigorously defined structures to identify, analyze, and extend patterns in quantity, shape, space, change, and uncertainty. This characterization of mathematics as pattern science is very different from the image of mathematics that most students encounter: a collection of procedures and algorithms to be memorized and applied; a source of right-or-wrong answers arrived at through prescribed methods; a subject in which the fastest students are the best students.

The transformation of mathematics education over the past forty years has been driven by a growing recognition — from mathematicians, mathematics educators, cognitive scientists, and neuroscientists — that the procedural, algorithm-focused model of mathematics instruction produces students who can follow procedures but cannot think mathematically; who cannot solve novel problems; who cannot reason and justify their conclusions; and who frequently hate mathematics.

The research foundation for a richer vision of mathematics education — built on genuine mathematical problem-solving; reasoning and justification; communication of mathematical ideas; connections within and across mathematical domains; and flexible use of representations — is robust, and the frameworks of Polya; NCTM; Schoenfeld; Boaler; Lesh and Doerr; and Van Hiele provide the most powerful and most practical guides to what mathematical thinking development requires.

Research Foundations of Mathematics and Problem-Solving Education

George Polya: How to Solve It

George Polya (Stanford University), in How to Solve It (1945) — one of the most widely read and most influential books in the history of mathematics education — developed the most enduring framework for understanding mathematical problem solving as a process that can be taught, practiced, and improved:

The Problem-Solving Heuristic: Polya describes a four-phase heuristic for approaching mathematical problems:

1. Understanding the problem: Before attempting to solve a problem, a solver must genuinely understand what is being asked. This phase involves: reading the problem carefully; identifying what is known (given information) and what is unknown (what needs to be found); representing the problem in a useful way (drawing a diagram; making a table; rephrasing in one's own words); and checking whether the problem is well-defined (do I have enough information? is the question clear?). Polya's questions for this phase: "What is the unknown? What are the data? What is the condition? Is it possible to satisfy the condition?"

2. Devising a plan: Most students' mathematical education gives them no tools for this phase — they know how to execute procedures but not how to select among approaches when the procedure isn't obvious. Polya identifies a range of problem-solving strategies (heuristics) useful at this phase: looking for a pattern; working backwards; solving a simpler problem; drawing a diagram; making a list or table; using symmetry; considering special cases; constructing an analogy; decomposing the problem into sub-problems; reframing the problem in a different representation; trying an extreme case. The most important meta-level skill is recognizing which heuristic might apply to a given problem — a recognition that develops through experience with a wide variety of problem types.

3. Carrying out the plan: Executing the chosen approach, while checking each step for validity. Polya's questions: "Can you see clearly that the step is correct? Can you prove that it is correct?"

4. Looking back: Checking the answer (does it make sense?); checking the argument (is each step valid?); considering whether the result can be obtained differently; considering whether the result or method could be applied to other problems; generalizing the result when possible. The "looking back" phase is the most educationally important and the most commonly skipped: students who get an answer and move on miss the deepest mathematical learning — understanding why the approach worked and what it might generalize to.

Heuristics as Teachable Strategies: Polya's central pedagogical claim is that problem-solving strategies (heuristics) can be explicitly taught — that teachers who model and discuss the strategies they use when solving problems, and who give students practice with those strategies in explicit ways, develop more capable mathematical problem-solvers than teachers who assign problems without attending to the processes students use to solve them.

NCTM: Principles and Standards for School Mathematics

The National Council of Teachers of Mathematics (NCTM), in Principles and Standards for School Mathematics (2000) and subsequent policy documents, established the most influential standards framework for mathematics education in the United States — and one of the most widely cited internationally:

Five Process Standards: NCTM identifies five process standards that describe mathematical competence:

1. Problem Solving: Building mathematical knowledge through problems; solving problems that arise in mathematics and other contexts; applying and adapting a variety of strategies; monitoring and reflecting on mathematical problem-solving. Problem solving is both the means and an end of mathematics education — the primary way students develop mathematical understanding and the primary competency that schooling should develop.

2. Reasoning and Proof: Recognizing reasoning and proof as fundamental to mathematics; making and investigating mathematical conjectures; developing and evaluating mathematical arguments; selecting and using various types of reasoning and methods of proof. Reasoning and proof connects mathematics to its epistemic foundation: mathematical truth is not established by authority or by the fact that an answer appears in the back of the textbook but by logical argument.

3. Communication: Organizing and consolidating mathematical thinking through communication; communicating thinking coherently to peers, teachers, and others; analyzing and evaluating mathematical thinking and strategies; using mathematical language precisely. Mathematical communication is both a vehicle for mathematical learning (explaining one's thinking develops and deepens it) and a mathematical competency in itself.

4. Connections: Recognizing and using connections among mathematical ideas; understanding how mathematical ideas build on one another; recognizing and applying mathematics in contexts outside mathematics. Mathematics is not a collection of isolated topics — fractions and division are connected; geometry and algebra are connected; probability and statistics are connected; mathematics and science are deeply connected.

5. Representation: Creating and using representations to organize, record, and communicate mathematical ideas; selecting, applying, and translating among mathematical representations; using representations to model and interpret physical, social, and mathematical phenomena. Multiple representations (concrete; pictorial; symbolic; numerical; verbal) are not interchangeable redundancies but distinct windows into mathematical ideas, each highlighting different features.

The Equity Principle: NCTM also articulates the Equity Principle — that "Excellence in mathematics education requires equity — high expectations and strong support for all students." This principle challenges the common assumption that mathematical ability is a fixed trait distributed unequally across students and groups, and calls instead for instruction designed to develop all students' mathematical competence rather than to identify and sort the "naturally" mathematically gifted.

Alan Schoenfeld: Mathematical Problem Solving

Alan Schoenfeld (University of California, Berkeley), in Mathematical Problem Solving (1985) and subsequent research, developed the most comprehensive empirical and theoretical account of what differentiates expert from novice mathematical problem solvers:

Four Components of Mathematical Competence: Schoenfeld identifies four components that together determine a solver's effectiveness:

1. Resources: Mathematical facts, procedures, and knowledge available to the solver. This is what most mathematics education focuses on — building students' repertoire of mathematical content knowledge. But resources alone are insufficient: expert solvers frequently fail not because they lack the relevant resources but because they fail to access or apply them appropriately.

2. Heuristics: Problem-solving strategies like those Polya describes — working backwards; considering special cases; exploiting analogies; decomposing problems. These are meta-level strategies for directing the problem-solving process rather than object-level mathematical knowledge.

3. Control (metacognition): The ability to monitor, evaluate, and regulate one's own problem-solving process — asking "Is this approach working? Should I try something different? Have I checked this step?" Schoenfeld's most striking research finding is that novice problem-solvers (including undergraduates) typically persist in an unproductive approach until time runs out, while expert problem-solvers frequently abandon unproductive approaches, reflect on what has not worked, and try a different strategy. Control is the most important differentiator of expert and novice problem-solvers — and the component most systematically absent from typical mathematics instruction.

4. Beliefs and Practices: A solver's beliefs about mathematics and about themselves as mathematicians profoundly affect their problem-solving behavior. Beliefs that limit mathematical performance include: "Mathematics problems have exactly one correct method"; "If you can't solve a problem in five minutes, you probably never will"; "Doing mathematics means following procedures the teacher showed you." These beliefs — which are not irrational given what most students have experienced in mathematics classrooms — produce behavior (giving up after a few minutes; searching for the "right" procedure; being unwilling to try an uncertain approach) that prevents effective problem-solving.

The Teaching Implication: Schoenfeld's research implies that mathematics instruction must attend to all four components — not only building resources (content knowledge) but also explicitly developing heuristic strategies; practicing metacognitive monitoring; and challenging limiting beliefs about the nature of mathematics and mathematical ability.

Jo Boaler: Mathematical Mindset and Open Tasks

Jo Boaler (Stanford University), in What's Math Got to Do with It? (2008); Mathematical Mindsets (2016); and research at YouCubed (Stanford), developed the most widely read practitioner-facing account of how mathematical mindset, task design, and classroom culture interact to produce (or undermine) mathematical learning:

Mathematical Mindset: Boaler draws on Dweck's growth mindset research to argue that mathematical ability is profoundly malleable — that the brain's capacity for mathematical thinking grows through engagement with challenging mathematical work, and that the common experience of students concluding they are "not math people" reflects the limitations of how they were taught, not their actual mathematical potential. This claim has significant equity implications: the students most likely to conclude they are "not math people" are disproportionately female; from lower-income backgrounds; and from minority groups — suggesting that beliefs about mathematical ability are shaped by social factors as much as by genuine differences in capacity.

Open Mathematical Tasks: Boaler argues that the most important change teachers can make is moving from closed tasks (problems with a single correct answer achieved by a prescribed method) to open tasks (problems that multiple students can approach in multiple ways at multiple levels of mathematical sophistication). She developed the concept of "low floor, high ceiling" tasks: mathematical problems that are accessible to all students (low floor — anyone can start) while remaining challenging and generative for the most advanced students (high ceiling — no one runs out of interesting mathematics to do).

The Status Problem in Collaborative Mathematics: Boaler's research on collaborative mathematical learning identifies what she calls the "status problem" — the classroom dynamic in which some students are perceived (by themselves and others) as mathematically competent while others are perceived as mathematically incompetent, and where this status hierarchy shapes participation in mathematical work. Students with low mathematical status participate less; contribute less; and receive less acknowledgment for their mathematical contributions — with the result that collaborative mathematical work can reproduce and reinforce existing inequities rather than challenging them. She proposes "complex instruction" (based on Cohen and Lotan's work) as an approach to disrupting this status problem through intentional celebration of multiple mathematical competencies.

Richard Lesh and Helen Doerr: Model-Eliciting Activities

Richard Lesh (Indiana University) and Helen Doerr (Syracuse University), in Beyond Constructivism: Models and Modeling Perspectives on Mathematics Problem Solving, Learning, and Teaching (2003), developed the framework of model-eliciting activities (MEAs) as a distinctive approach to mathematical problem-solving in real-world contexts:

Models and Modeling: Lesh and Doerr define mathematical models as systems of structured thinking used to describe, explain, or predict something in the world — quantitative relationships; spatial structures; logical rules; patterns; probability distributions. Mathematical modeling involves constructing, testing, revising, and communicating mathematical models for real-world situations — a mathematical practice that is both highly authentic (this is what mathematicians and engineers actually do) and highly developmental (constructing models requires students to think about what mathematical structure captures a real-world situation, and why).

Model-Eliciting Activities: MEAs are structured problem-solving activities that require students to create, test, and refine mathematical models for real situations. An MEA typically: (1) presents a realistic, complex situation that requires mathematical analysis; (2) asks students to create a shareable tool, procedure, or system that others could use to solve similar problems; (3) requires documentation and communication of the mathematical reasoning; (4) leads to a product that can be evaluated for quality and refined. The key feature of an MEA is that it requires students to make explicit the mathematical structure they are using — not just to compute an answer but to articulate the model.

Six Principles for MEA Design: Lesh and Doerr identify six principles guiding MEA design:

1. Reality principle: The situation should be one students can imagine as real.
2. Model construction principle: Students must build a mathematical model, not just apply a known procedure.
3. Self-assessment principle: Students must be able to evaluate the quality of their model themselves — does it work? is it reasonable?
4. Documentation principle: Students must document and communicate their mathematical reasoning.
5. Generalizability principle: The model must be applicable beyond the specific case — a general solution procedure, not a one-off answer.
6. Effective prototype principle: The activity should yield a deep, generalizable mathematical understanding that students can revisit and extend.

Pierre van Hiele: Levels of Geometric Thinking

Pierre van Hiele (Dutch educator), in Structure and Insight: A Theory of Mathematics Education (1986), based on earlier work with his wife Dieke van Hiele-Geldof, developed the most influential theory of children's geometric reasoning development:

Five Van Hiele Levels:

1. Level 0 — Visualization: Students recognize geometric shapes holistically — by their appearance as a whole — without analyzing their properties. A student at this level recognizes a square because it "looks like a square," not because it has four equal sides and four right angles.
2. Level 1 — Analysis: Students begin to notice and name properties of shapes — a square has four equal sides; opposite sides of a rectangle are parallel; a triangle has three sides. But they do not yet understand the relationships between properties.
3. Level 2 — Abstraction/Ordering: Students begin to understand the relationships between properties — if a quadrilateral has four equal sides and four right angles, it must be both a square and a rectangle; therefore squares are special rectangles. Logical reasoning about the relationships between definitions begins.
4. Level 3 — Formal Deduction: Students can construct and understand formal mathematical proofs — they understand why properties follow from definitions and can construct valid logical arguments.
5. Level 4 — Rigor: Students can reason about geometry in formal axiomatic systems — understanding different geometries and the relationships between axiom systems.

Instructional Implications: Van Hiele's most important claim is that progress through the levels cannot be rushed by direct instruction at a higher level — students must develop through each level in sequence, and instruction pitched above a student's current level will be incomprehensible. The implication for curriculum and instruction is that geometric instruction must be calibrated to students' actual level of geometric thinking — beginning with rich visual experience (Level 0-1) before moving to formal properties (Level 2) and eventually to proof (Level 3).

AI Applications in Mathematics and Problem-Solving Education

Polya-Schoenfeld Problem-Solving Coaching

"Design a complete mathematical problem-solving coaching framework for Grades 5-9 — 'Thinking Like a Mathematician: A Polya-Schoenfeld Problem-Solving Coaching System' — grounded in Polya's four phases, Schoenfeld's four-component model (resources; heuristics; control; beliefs), and Boaler's mathematical mindset principles. This framework teaches mathematical problem-solving as a learnable, practiceable skill — not as something students either have or don't have. System Structure: The framework consists of: explicit heuristic instruction; metacognitive prompting at each phase of problem solving; growth-mindset framing; and collaborative problem-solving protocols. Phase 1: Understanding the Problem — Coaching Protocol. When students first encounter a problem, coaches (teachers or trained peers) ask: 'Can you tell me in your own words what this problem is asking you to find?' 'What information does the problem give you? Let's write it out.' 'Is there a way to represent this situation — a diagram? a table? a number line?' 'Have you seen a problem like this before? How was it similar? How was it different?' 'What would the answer look like? What type of quantity will it be?' Common mistakes at this phase: jumping to calculation before understanding the problem; misreading what is given vs. what is unknown. Teacher move: require students to write a sentence stating what they are looking for before beginning any calculation. Phase 2: Devising a Plan — Heuristic Instruction. Explicitly teach and practice 8 key problem-solving heuristics, one per week in the first 8 weeks: Week 1 — Draw a diagram. When to use: problems involving spatial relationships; quantities connected to each other; changing quantities over time. Practice problems: three progressively harder problems where diagram drawing is the key. Week 2 — Make a table or list. When to use: problems involving patterns; combinations; organized counting. Week 3 — Find a pattern and generalize. When to use: problems involving sequences; number patterns; repeating structures. Connects to algebraic thinking. Week 4 — Work backwards. When to use: problems where the final state is known and you must find the starting state. Week 5 — Solve a simpler related problem. When to use: complex problems where a simpler version reveals the structure. Week 6 — Use logical reasoning / process of elimination. When to use: problems with constraints that allow systematic elimination of possibilities. Week 7 — Consider extreme cases or special values. When to use: problems where the general case is hard but extreme cases are clear. Week 8 — Reframe in a different representation. When to use: problems where the given representation is difficult but a different representation (algebraic; geometric; numeric; graphical) reveals the structure. Each week: introduce the heuristic with examples; practice with scaffolded problems; apply to a new unseen problem; reflect 'When would you use this? How did you know?' Phase 3: Carrying Out the Plan — Metacognitive Monitoring (Schoenfeld). During solution, coach with the 'inner voice' protocol: teach students to ask themselves: 'Is what I am doing moving me toward the answer?' 'Have I checked this step? Can I verify it?' 'I'm stuck — should I keep trying this approach or try something else?' 'It's been 5 minutes on this approach — what's a different way I could think about this?' Explicitly teach the monitoring skill that Schoenfeld identifies as the most important differentiator of expert and novice solvers: the ability to step back, evaluate whether your current approach is working, and switch approaches if it isn't. Phase 4: Looking Back — The Mathematical Gold. Looking back protocol, required for every problem: 'Does my answer make sense? Does it answer the question that was asked?' 'Can I check my answer a different way?' 'Is there a faster or more elegant method?' 'What mathematical pattern or structure does this problem reveal?' 'What would happen if I changed one of the conditions — would the answer change in a predictable way?' 'Where else have I seen this structure?' Looking back is where the deepest mathematical learning occurs — where students develop the pattern-recognition and connection-making that characterizes mathematical maturity. Beliefs Work: At the start and end of each unit, discuss and challenge limiting mathematical beliefs. 'Mathematics problems should be solved in a few minutes — if you're taking longer, you're doing it wrong.' Challenge: 'Professional mathematicians work on single problems for years. In a 45-minute class period, 15 minutes on a single problem is serious mathematical work.' Full system with: weekly heuristic lesson plans; metacognitive monitoring poster for classroom display; problem-solving journal template; growth-mindset reflection prompts; partner problem-solving protocol; self-assessment rubric for each Schoenfeld component; bank of 40 rich problems organized by primary heuristic."

Open Mathematical Task Design

"Design a bank of 15 open mathematical investigations — 'Open Mathematics: Low-Floor High-Ceiling Tasks for Inclusive Problem-Solving' — grounded in Boaler's open task principles, NCTM's five process standards, and Lesh-Doerr model-eliciting activity principles. Each task is designed to be accessible to all students in a mixed-ability class while remaining genuinely challenging for the most advanced learners. TASK DESIGN PRINCIPLES (Boaler): Low floor: every student can begin engaging meaningfully from their first day in the class — no prior mastery of specific content required to access the task. High ceiling: there is always more to explore, more to generalize, more to prove — the task is not exhausted when a first answer is found. Multiple representations: the task can be approached numerically; algebraically; geometrically; or verbally. Collaborative: the task generates genuine mathematical conversation — different students bring different insights that together exceed what any individual could produce. Valued multiple competencies: different students' strengths (pattern recognition; spatial reasoning; systematic organization; algebraic fluency; verbal explanation) are all valuable in the task. TASK 1: The Square Numbers Investigation (Grades 3-8). Opening: 'What do you notice about the sequence 1, 4, 9, 16, 25, 36...? Explore as many patterns as you can find.' Low floor: a student who knows multiplication can generate the sequence; a student who doesn't can build it from square tiles. High ceiling: the differences between consecutive squares are odd numbers (1, 3, 5, 7...) — why? Can you prove it algebraically? How does this relate to the formula for the sum of the first n odd numbers? What are perfect squares in other bases? TASK 2: The Rectangle Investigation (Grades 3-7). Opening: 'Using exactly 24 square tiles, how many different rectangles can you make? What do you notice?' Low floor: students can use actual tiles. High ceiling: the different rectangles correspond to factor pairs of 24 — what does this tell us about prime numbers? What happens with prime numbers of tiles? What happens with numbers that are perfect squares? How many rectangles can you make from n tiles for any n? TASK 3: The Growing Patterns Investigation (Grades 4-9). Opening: Present a visual pattern (e.g., a sequence of figures where figure n has n² + n squares arranged in an L shape). Questions: 'How many squares are in Figure 1? Figure 2? Figure 10? Figure n?' Low floor: count the squares in Figure 1 and Figure 2. High ceiling: derive the general formula algebraically; connect to the visual structure; prove the formula using the visual representation; connect to quadratic functions. TASK 4: The Fraction Between Two Fractions Investigation (Grades 5-9). Opening: 'Can you always find a fraction between any two fractions? How many fractions are there between 1/2 and 2/3? Between any two fractions?' Low floor: find one fraction between 1/2 and 2/3 by finding common denominators. High ceiling: there are infinitely many fractions between any two fractions (density of rationals); the Stern-Brocot tree; mediant fractions. TASK 5: The Sum of Angles Investigation (Grades 4-7). Opening: 'What is the sum of the interior angles of a triangle? A quadrilateral? A pentagon? A polygon with n sides? How do you know?' Low floor: measure angles with a protractor; cut and rearrange triangle angles. High ceiling: prove the formula algebraically using triangulation; extend to exterior angles; connect to the formula for regular polygons and their angle measures. Full task bank with: all 15 investigations; teacher facilitation guide for each task (common approaches; conversation launching questions; extending questions; connections to curriculum); student recording sheet; assessment rubric evaluating problem solving; reasoning and proof; communication; connections; representation; adaptation guide for different grade levels and ability levels; home/family investigation versions."

Van Hiele-Aligned Geometry Investigation Sequence

"Design a Van Hiele-aligned geometry investigation sequence for Grades 3-8 — 'Geometry from the Inside Out: A Van Hiele Developmental Investigation Sequence' — grounded in Van Hiele's five levels, NCTM's process standards, and Polya's looking-back habits. This sequence develops geometric understanding through rich investigation at each Van Hiele level before progressing to the next, ensuring that instruction is appropriately pitched rather than formally premature. Level 0-1 Investigations (Grades 3-4): Geometric shape recognition and property identification. Investigation 1 — Shape Sort: Provide students with a large set of physical and pictorial geometric shapes (varied sizes, orientations, and colors). Students sort them in their own ways; describe their sorting rules; re-sort with different criteria. Discussion: 'How did you decide? What makes a shape a [name]? Can you find a shape that could belong to two groups?' Level 0 → 1 transition: Students move from 'it looks like a rectangle' to 'it has four sides and four right angles.' Investigation 2 — Property Hunt: For each shape category, create a 'property detective' worksheet. Students must discover and record all properties of each shape: sides (count; length; parallel pairs); angles (count; size; right?); symmetry (fold lines; rotational); diagonals (count; bisecting?; perpendicular?). Level 1-2 Investigations (Grades 4-6): Property relationships and classification. Investigation 3 — Classification Tree: 'All squares are rectangles. True or false? All rectangles are squares. True or false?' Students create a Venn diagram or hierarchy tree showing the relationships among quadrilaterals: parallelograms; rectangles; rhombuses; squares; trapezoids. The key Level 2 insight: shapes can belong to multiple categories, and broader categories have fewer properties while narrower categories have more. Investigation 4 — The Triangle Family: Classify triangles by sides (equilateral; isosceles; scalene) and by angles (acute; right; obtuse). Create a two-dimensional grid: 'Can you have a right equilateral triangle? An obtuse equilateral triangle? A right scalene triangle?' Which cells are possible and which are not — and why? Level 2-3 Investigations (Grades 6-8): Logical deduction and proof. Investigation 5 — The Triangle Angle Sum: 'What is the sum of the interior angles of a triangle? Always? How do you know?' Students: (a) Measure and add angles of many triangles — inductive evidence. (b) Tear off the corners of a triangle and arrange them to form a straight line — visual-intuitive justification. (c) Construct a logical argument using parallel line angle properties — formal deductive proof. Discussion: What is the difference between these three levels of knowing? When is each sufficient? Investigation 6 — Exterior Angle Theorem: 'An exterior angle of a triangle equals the sum of the two non-adjacent interior angles. Can you prove it?' Two approaches: using the triangle angle sum (indirect); constructing a parallel line (direct). Students choose and present their proof. Full sequence with: investigation guides for all six investigations and additional extension investigations; Van Hiele level indicator tool for assessing student's current level; teacher facilitation guide at each level; common misconceptions at each level and how to address them; connection to formal curriculum at each grade level."

Classroom Scenario: Ms. Fernández's Mathematics at Ceuta

The mathematical problem-solving program at CEIP Murillo in Ceuta — a Spanish autonomous city and enclave of approximately 85,000 inhabitants on the northern coast of Morocco, separated from Spain by the Strait of Gibraltar and sharing a land border with Morocco — is shaped by the cultural and mathematical vision of Silvia Fernández-Manzano, a primary and lower secondary mathematics specialist who has adapted the Boaler mathematical mindset approach to the multilingual context of one of Spain's most diverse cities.

Ceuta's Unique Context: Ceuta is one of two Spanish autonomous cities on the African continent (the other being Melilla). It has been under Spanish sovereignty since 1668 and is a full member of the European Union, despite being geographically located in North Africa separated from the Iberian Peninsula by the Strait of Gibraltar (approximately 14 kilometers wide). The city's approximately 85,000 inhabitants are roughly 50% of Christian background and approximately 40-45% of Muslim background (principally descended from Moroccan Amazigh and Arab families), with the remainder of diverse origins including Jewish and Hindu communities — the latter predominantly descendants of Sindhi merchants. This cultural and religious diversity is reflected in the schools, where students speak Spanish as the primary language of instruction alongside Arabic, Darija (Moroccan Arabic), Tamazight (Berber languages), and various heritage languages. Ceuta is also notable for its unique position as a border city: the Ceuta-Morocco border is one of the EU's land borders in Africa, and the city's economic and social life is deeply shaped by cross-border activity.

Silvia's Mathematical Context: The diverse multilingual context creates specific mathematical challenges: many students have strong mathematical intuitions and family mathematical traditions (mental arithmetic practices in Arabic; spatial reasoning from geometric decoration traditions in Islamic art) that the Spanish curriculum does not draw on. Silvia uses EduGenius (edugenius.app) to generate Boaler-style open mathematical tasks that are genuinely accessible across the varied mathematical backgrounds of her students — low-floor tasks that students can approach from different mathematical cultural traditions — while simultaneously meeting the Spanish national curriculum competencies for mathematics. She uses NCTM's five process standards as her organizing framework for assessment: tracking whether students are developing mathematical problem-solving, reasoning, communication, connections-making, and representational fluency — not merely computational accuracy. Her collaboration with the school's Arabic language teacher to integrate Islamic geometric patterns (which encode sophisticated geometric understanding: tessellations; symmetry; ratio; proportionality) into her Van Hiele-aligned geometry investigations has produced some of her richest mathematical classroom experiences.

Internal Resources

• Best AI for STEM and Science Inquiry Education
• Best AI for Project-Based Learning
• Best AI for Assessment and Feedback in Education
• Best AI for Early Literacy Phonics and Decoding
• Best AI for Differentiated Instruction and Universal Design for Learning