Pattern Recognition in
Computational Thinking
This teacher training module prepares educators to confidently teach the Pattern Recognition sub-skill under Computational Thinking for Class 6, as defined in the CBSE CT & AI Curriculum 2026–27. Pattern Recognition is the ability to identify regularities, structures, and rules — a core skill that underpins algorithm design, data analysis, and AI model training.
- Understand what Pattern Recognition means across numbers, shapes, letters, and mixed contexts at Class 6 level.
- Map all 5 learning outcome areas to classroom tasks and worked examples.
- Practise 5 hands-on activities ready for immediate classroom use.
- Apply a rubric-based approach to assess pattern recognition skills fairly and consistently.
- Identify and address common student errors in pattern identification and extension.
| Module Component | Duration | Format |
|---|---|---|
| Section 1: Overview & Curriculum Context | 20 min | Presentation + Discussion |
| Section 2: Unpacking the Learning Outcomes | 30 min | Group Analysis + Examples |
| Section 3: Pedagogical Strategies | 25 min | Demonstration |
| Section 4: Classroom Activities (Hands-on) | 65 min | Workshop |
| Section 5: Assessment Strategies | 25 min | Rubric Practice |
| Section 6: Common Misconceptions & FAQs | 15 min | Q&A |
Class 6 students are expected to identify, extend, and justify complex patterns involving multiple simultaneous changes. This is a significant step up from lower classes where single-rule patterns were the norm. The five sub-areas below define the full scope of this learning outcome.
| # | LO Area | What it looks like in Class 6 | Bloom’s Level |
|---|---|---|---|
| LO 1 | Numbers with mixed operations & logical rules | Extending sequences where the rule involves two or more operations (e.g., ×2 then +3 alternating); identifying the rule from given terms | Analyse / Evaluate |
| LO 2 | Shapes/images with changing attributes (size, position, count, orientation) | A shape changes its size AND rotates AND its count increases simultaneously across a series — student must identify all three rules | Analyse |
| LO 3 | Letters & symbols with positional or alphabetical logic | A sequence like A, C, F, J, O … where the gap between letters increases by 1 each step; or symbol codes based on position-in-alphabet arithmetic | Apply / Analyse |
| LO 4 | Alternation, skipping, grouping, or cyclic behaviour | Patterns that repeat in cycles of 3 or 4; patterns with two interleaved sub-sequences; grouping patterns like (1)(2,3)(4,5,6)… | Analyse / Evaluate |
| LO 5 | Mixed patterns combining numbers, shapes & letters with dependency rules | A table/matrix where rows follow one rule, columns follow another, and the cell value depends on both — student must decode two simultaneous rules | Evaluate / Create |
Pattern Recognition is one of the four pillars of Computational Thinking. In AI, it is the literal foundation of machine learning — a model “learns” by recognising patterns in training data. When students practise identifying and justifying rules in sequences, they are building the same mental model that powers recommendation engines, image classifiers, and language models. Making this connection explicit in class dramatically increases student motivation.
Classroom Example Bank — Sample Patterns for Each LO:
Pattern recognition at Class 6 level requires students to move beyond “spot the rule” into justifying and generalising rules. The following strategies are specifically effective for this LO.
| Strategy | Description | Best for LO |
|---|---|---|
| Rule → Prediction → Verify (RPV) | Students first state a hypothesised rule, use it to predict the next 2 terms, then check. Forces explicit rule articulation before answer-seeking. | LO 1, LO 3, LO 4 |
| Attribute Decomposition | For shape/visual patterns, students list ALL changing attributes in a table (size, colour, orientation, count) before trying to find each sub-rule separately. | LO 2, LO 5 |
| Pattern Breaking (Counterexample) | Give students a pattern with a “broken” term and ask them to find it. Builds rule-checking discipline and precision. | All LOs |
| Pattern Construction (Create-Your-Own) | Students design their own 2-rule pattern for a partner to solve. Designing a pattern requires deeper understanding than solving one. | LO 4, LO 5 |
| Colour-Coding / Layering | Students use different coloured pens/markers to annotate each separate rule within a mixed pattern. Makes multi-rule patterns visually manageable. | LO 2, LO 5 |
| Connecting to AI Context | Before each activity, briefly connect the pattern type to a real AI application (e.g., number patterns → forecasting; visual patterns → image recognition). Anchors abstract skills in relevance. | All LOs |
The most impactful trainer move in this session is to present a mixed pattern (LO 5) and ask teachers to solve it collaboratively. When even experienced teachers struggle to articulate the rule precisely, it builds empathy for students and motivates the use of the Attribute Decomposition strategy. Plan for 8–10 minutes of productive struggle before revealing the approach.
Each activity targets one or more learning outcomes and is classroom-ready. All activities are unplugged or low-resource unless noted.
Students are given 6–8 number sequences where each involves two operations alternating or combining (e.g., ×3 −1, or +2 then ×2). They must (a) identify the rule, (b) write it as a formal rule using words or a formula, and (c) predict the next two terms. A “Detective Card” template guides the three steps.
- Printed Detective Card worksheet (3 difficulty levels)
- Number sequence card set (prepared)
- Calculator (optional, for verification)
- Model one sequence using the RPV strategy: “My hypothesis is ×2 +1. Let me test it on term 3…”
- Require students to write the rule in words before using symbols.
- Challenge fast learners: “Can you find a sequence where two different rules both work for the first 4 terms but diverge at term 5?”
Each group receives a strip of 5 shape cards (physical printed cards) where 2–3 attributes change simultaneously (e.g., size grows, number of sides increases, shading alternates). Card 3 or 4 is left blank — students must complete it AND write an Attribute Table documenting each changing attribute and its rule separately before deciding on the missing card.
- Printed shape card strips (3 levels: 1-attr, 2-attr, 3-attr)
- Attribute Analysis Table (printed template)
- Coloured pencils for annotation
- Blank shape cards for drawing the answer
- Insist students complete the Attribute Table BEFORE drawing the missing card.
- Ask: “How many attributes are changing? Could any attribute be staying the same on purpose?”
- Extension: Ask groups to create their own 3-attribute strip for another group to solve.
Students receive a numbered alphabet reference strip (A=1, B=2 … Z=26). They then solve three types of letter pattern tasks: (a) extend a letter sequence using position-gap logic, (b) decode a “symbol cipher” where each letter is encoded by its position + a fixed offset, and (c) find the rule in a sequence mixing letters and their alphabetical position numbers.
- Printed alphabet reference strip (A=1 to Z=26)
- Cipher worksheet (3 tasks per student)
- Optional: coloured markers for gap-annotation
- Explicitly connect to encryption: “This is exactly what Caesar’s cipher does — and modern computers use far more complex versions.”
- Have students annotate the gaps between letters (like +2, +3, +4…) above the sequence.
- For mixed letter-number sequences (e.g., A1, C3, F6, J10…), encourage students to look at numbers and letters separately first.
Students receive a long mixed sequence (30+ terms) on a strip and must first identify the cycle length, then use it to answer: (a) What is the 47th term? (b) What is the sum of the first 20 terms? (c) At what position does [a specific element] first appear for the 4th time? This task builds the critical skill of using a pattern rule to answer questions far beyond visible terms.
- Printed long sequence strips (cyclic length 3, 4, or 5)
- Cycle Diagram template (for visualising the repeating unit)
- Ruler for grouping terms visually
- Teach the “bracket grouping” strategy: draw brackets around each cycle to make structure visible.
- Model using division: “The 47th term — 47 ÷ 4 = 11 remainder 3 — so it’s the 3rd element of the cycle.”
- Connect to computing: “Cyclic patterns are like loops in programming — the code repeats every N steps.”
A 4×4 grid is presented where each cell contains a number, letter, or symbol. Students must discover that rows follow one rule and columns follow a different rule, and then use both rules simultaneously to fill in 4–5 missing cells. A final challenge cell can only be reached using both rules — students cannot guess it from a single rule alone.
- Printed Dependency Matrix worksheets (2 levels)
- Colour-coding pencils (one colour per rule)
- Rule Discovery Card (structured template for writing down each rule)
- Ask students to colour-code: “Use blue to annotate the row rule and green for the column rule.”
- Teach students to first solve only 1 complete row and 1 complete column before tackling missing cells.
- Connect to AI: “This is exactly how a spreadsheet function works — and how neural networks process rows and columns of data.”
- Extension: Students design their own 3×3 dependency matrix for a classmate to solve.
Assessment of Pattern Recognition must go beyond “right or wrong.” Partial credit should be given for correctly identifying the rule even when the extended term has a computation error. The rubric below reflects this principle.
| Criterion | Level 4 — Exceeds | Level 3 — Meets | Level 2 — Approaching | Level 1 — Beginning |
|---|---|---|---|---|
| Rule Identification | Identifies all rules in a multi-rule pattern; expresses them precisely in words and/or notation | Identifies the main rule correctly; partially describes secondary rules | Identifies a single rule in multi-rule patterns; misses additional rules | Cannot articulate the rule; relies on guessing or visual “it looks right” |
| Pattern Extension | Correctly extends pattern by 3+ terms and justifies each step using the stated rule | Extends correctly by 2 terms with minor justification | Extends 1 term correctly; errors increase in subsequent terms | Cannot extend the pattern independently |
| Justification & Generalisation | Can use the rule to answer questions about distant terms (e.g., “What is the 50th term?”) and generalise using a formula or clear verbal rule | Uses the rule for near terms; attempts but struggles with distant terms | Can verify given terms match the rule but cannot predict beyond the sequence | No evidence of rule-based reasoning; cannot verify or predict |
| Multi-Attribute Tracking | Systematically tracks all changing attributes; analyses each separately before combining | Tracks 2 attributes correctly; misses or confuses a third | Focuses on the most obvious attribute; misses subtler changes | Treats multi-attribute patterns as single-rule patterns |
| Pattern Creation | Designs an original, valid multi-rule pattern that is solvable by a peer; rule is unambiguous | Designs a valid 1–2 rule pattern; may have minor ambiguity | Attempts to create a pattern but it has errors or is unsolvable | Cannot independently design a pattern |
- Rule-in-One-Sentence: After identifying a pattern, students write the rule as one precise sentence. Teacher scans for vague language (“it keeps growing”) vs precise language (“multiply by 2 then add 1”).
- Predict Term 20: Give a cyclic or mixed pattern and ask only for the 20th term — not the next term. Forces use of generalised rule rather than step-by-step counting.
- Broken Pattern Spot: Present a 7-term sequence with one wrong term. Students circle the error and rewrite the correct term. Quick and diagnostic.
- Think-Pair-Justify: Students solve a pattern individually, then compare with a partner. If they disagree, both must justify using the rule — not just assert their answer.
| Misconception / Challenge | Why it happens | How to address it |
|---|---|---|
| “The pattern must always go up (or follow addition)” | Most exposure in earlier classes is to additive sequences; students default to addition even when the rule involves multiplication or alternation | Deliberately include sequences that decrease, oscillate, or use multiplication. Start each unit with a “surprise” decreasing or cyclic sequence to disrupt the assumption early. |
| Finding the rule from only 2 terms | Students rush to generalise too early — 2 terms can fit infinitely many rules | Make a classroom rule: “You need at least 3 terms to hypothesise a rule. You need 4 terms to be confident.” Enforce this consistently. |
| Treating multi-attribute patterns as single-rule patterns | Students see the most obvious change and stop looking, missing secondary or tertiary attribute changes | Require the Attribute Table to be completed before any answer is written. Physical colour-coding of attributes forces attention to all changes. |
| Confusing cyclic position with absolute position | For cyclic patterns, students count manually to term 47 instead of using the modulo/remainder strategy | Teach the remainder method explicitly with a worked example. Connect to clocks: “The hour hand is cyclic — 13:00 is the same position as 1:00.” |
| “I see the pattern but I can’t explain it” | Students have developed strong visual intuition but lack the mathematical vocabulary and framework to articulate rules | Provide a sentence frame: “The rule is: starting from ___, each term is obtained by ___.” Pair with vocabulary practice for terms like ‘alternating,’ ‘cyclic,’ ‘increment,’ ‘operation.’ |
| Over-generalising from one type of pattern to others | A student who cracks number patterns tries to apply numerical logic to letter patterns (e.g., treating A=1 and adding instead of looking at alphabetical gaps) | Explicitly discuss how the same abstract logic applies across domains: “The STRUCTURE is the same; the OBJECTS are different.” Use side-by-side comparison of equivalent number and letter patterns. |
- Q: How is this different from the pattern work done in Maths class?
A: Maths focuses primarily on numerical sequences and algebraic generalisation. CT Pattern Recognition explicitly includes visual, symbolic, and mixed patterns, and emphasises the process of rule extraction and justification — not just getting the right next term. The CT lens also connects each pattern type to computational and AI applications. - Q: Should students be allowed to use a calculator?
A: For LO 1 (number patterns), calculators are acceptable for verification after the rule is identified — never for finding the rule. For all other LOs, calculators are not relevant. - Q: How many periods are recommended for this LO?
A: 8–10 periods of 40 minutes each. Distribute across the unit: 2 periods per LO area, with the last 2 periods for mixed/integrated pattern tasks. - Q: What if a student finds a different valid rule for the same sequence?
A: This is a rich teaching moment. Accept it as valid and ask: “Can both rules predict term 10 correctly? Which rule is simpler?” This develops the computational thinking habit of preferring elegant, generalised solutions — a core concept in algorithm design.
Use this checklist after completing the training module to confirm readiness to teach Pattern Recognition (Class 6 CT).
- I can explain all 5 LO sub-areas of Pattern Recognition in my own words.
- I can give at least one original classroom example for each LO area.
- I can connect each LO area to a real-world CT or AI application.
- I understand why justification and generalisation are as important as identification and extension.
- I can demonstrate the Rule → Prediction → Verify (RPV) strategy with a live example.
- I can facilitate the Attribute Decomposition strategy for shape patterns.
- I can run the Dependency Matrix activity independently with a group.
- I have materials ready for at least 3 of the 5 activities in this module.
- I can use colour-coding to help students track multi-rule patterns.
- I can apply the 5-criterion rubric to a sample student response.
- I understand how to award partial credit for correct rule identification even when the extended term has errors.
- I can design a “Predict Term 20” formative question for any pattern type.
- I know how to handle a situation where a student finds a different valid rule.