# Continue the pattern questions for visual reasoning

**How do I write a question that asks what comes next in a pattern?** A continue the pattern question shows a sequence or a grid built on a rule and asks what comes next. Choose one or two rules, show each at least twice, write wrong options that each break one rule, and check that only one option fits. More rules make a harder question, and practice with the rule types raises scores.

Published 2026-10-10 by EdProcto. https://edprocto.com/guides/continue-the-pattern-questions-for-visual-reasoning

A continue the pattern question shows a row of shapes, a run of numbers or a grid built on a rule, and asks what comes next. It is a standard question in non-verbal reasoning tests, in early mathematics, and, as number series and figure matrices, in aptitude batteries.

This guide is published by EdProcto, where the shape is called [Continue the pattern](https://edprocto.com/templates/pattern_cont): the pattern is typed in the prompt, the options are typed or given as pictures, and the one right answer is marked by rule.

The short version: one or two rules, each shown at least twice; wrong options that each break one rule; a hunt for a second answer; and the same practice for everyone before scores are compared.

## What does a continue the pattern question measure?

It measures inductive reasoning: finding the rule in the examples shown and applying it to a case not yet seen.

Carpenter et al. (1990) studied how college students solved a widely used matrix test, from what they said aloud, where they looked and their errors. What separated higher scorers from lower was the ability to induce abstract relations and to manage a large set of goals in working memory.

The reasoning can be taught. Klauer and Phye (2008) pooled 74 experiments in which nearly 3,600 children were trained in a procedure for making comparisons: scores on tests of fluid reasoning improved, and the learning of school subject matter improved more.

| Where it is used | The pattern | In EdProcto |
| --- | --- | --- |
| Early mathematics | A repeating or growing row of shapes | [Build the pattern](https://edprocto.com/templates/build_pattern), on large buttons |
| Non-verbal reasoning tests | A row or grid of figures, one missing | Continue the pattern |
| Aptitude batteries | Number series, letter series, matrices | The reasoning [kits](https://edprocto.com/kits#hiring) |

### Patterning in early mathematics

England's non-statutory early years guidance has three- and four-year-olds learning to extend and make ABAB patterns and to put right a mistake in a repeating pattern, and reception children continuing, copying and creating repeating patterns, with rules such as AB, ABB and ABBC (Department for Education, 2023).

Wijns et al. (2021) assessed 410 children each year at four, five and six. From five to six, skill with repeating patterns and with growing patterns each predicted later number ability, and not the reverse.

Kidd et al. (2013) went past prediction. First-graders who scored lowest on a patterning test had 15-minute lessons three times a week for six and a half months. Those taught patterning outscored those taught mathematics, reading or social studies on one mathematics concepts scale, and not on applied problems or reading.

Burgoyne et al. (2017), reviewing such studies, call much of the evidence weak and the causal claim tentative. For a child who cannot read yet, see [quizzes with pictures and sounds](https://edprocto.com/guides/quizzes-for-children-who-cannot-read-yet-pictures-and-sounds).

## What makes a pattern question hard?

The number of rules first, then the kind of rule.

Koch et al. (2022) built 220 matrix items from six rules, one to five rules to an item, and gave them to more than 2,500 applicants to German medical schools. The more rules an item held, the harder it was (r = 0.53). Which rules an item used explained 34% of the differences in difficulty: a third, not the whole.

The kind matters too. The studies Krautter et al. (2021) cite treat two rules as easier: the third cell is the first two added together, or the first less the second. Two are harder: the third cell keeps only what the first two share, or only what they do not share.

For a class, use one rule for a first item or a young child and two for a hard one.

## Do practice and coaching change scores on pattern tests?

Yes, and quickly: being taught the rules raises scores.

- Krautter et al. (2021) randomly assigned 287 adults to a brief training in none of four rules, the two easy ones, the two difficult ones or all four, before a matrix test. The group taught only the difficult rules solved a median 95% of the rules, against 88% with no training and 88% with the easy rules alone.
- Schneider and Sparfeldt (2021) showed 58 people a video tutorial on number series and 52 an unrelated one. The first group scored about half a standard deviation higher (d = 0.48).
- GL Assessment (n.d.), a publisher of 11+ tests for school admissions, gives away familiarisation papers, saying they are particularly useful for verbal and non-verbal reasoning because children may not have met such questions before.

So a score on pattern questions mixes reasoning with familiarity. Before scores are compared, give everyone the same worked examples and practice set. Figures for repeated sittings are in [aptitude test practice](https://edprocto.com/guides/aptitude-test-practice-numerical-verbal-and-logical-reasoning) and [aptitude tests for hiring](https://edprocto.com/guides/aptitude-tests-for-hiring-what-they-measure-and-how-to-use-them).

## Are pattern questions culture-fair?

Not on the evidence, though they were long treated as culture-fair tests because they use no words.

Rosselli and Ardila (2003) reviewed work on non-verbal tests, which, they note, have often been considered fair across cultures and levels of schooling, or at least fairer than verbal tests. They report a strong association between level of education and non-verbal scores, and significant differences between cultural groups.

Lohman et al. (2008) gave three non-verbal tests to 1,198 elementary school children, about 40% of them still learning English. The English learners scored 0.5 to 0.67 of a standard deviation lower on all three.

A page without words still draws on schooling and on familiarity with the task. Give the instructions in the student's own language and a worked example before the first item, and never read the score as a measure of intelligence.

### What we could not confirm

We found no study of teacher-written pattern questions in a class quiz, and none on the best number of options. The rule-counting figures come from able adults applying to medical school. We read Koch et al. (2022), Krautter et al. (2021) and the government and publisher pages in full; every other study we read as an abstract.

## How do you write a continue the pattern question?

Write the rule first, then the pattern, then the options, and finish by hunting for a second answer.

1. **Say the rule in one sentence.** "Each gap is one bigger than the last." If the sentence needs "and", the item has two rules.
2. **Show each rule at least twice.** A repeating unit twice in full before the gap (▲ ● ● ▲ ● ● ▲ ?), a number series of at least four terms, a grid with two complete rows.
3. **Type it in characters.** Shapes, arrows, letters or digits, copied from this page or your device's symbol keyboard, with a space between items and a question mark for the gap. Type a grid one row to a line: the prompt keeps its line breaks and its spaces on the student's screen and on paper.
4. **Write the right answer, then the wrong ones.** Each wrong option breaks one rule and keeps the rest, so the choice shows which rule was missed. See [how to write distractors](https://edprocto.com/guides/how-to-write-distractors-that-work).
5. **Balance the set.** With two rules, add the option that breaks both, so every shape and every number appears equally often. Otherwise the right answer is the option most like the rest, and a student can spot it without the pattern.
6. **Hunt for a second answer.** Try every simple rule on the terms shown, and read a grid by columns as well as by rows.
7. **Name no position.** Options are shuffled by default, so never write "the second one".
8. **Put the rule in the explanation,** build the question in EdProcto as [Continue the pattern](https://edprocto.com/templates/pattern_cont), and sit it on a phone to see that every symbol shows.

### Why exactly one answer needs a check

A short pattern usually fits more than one rule. Take 2, 4, 8, ?: doubling gives 16, and gaps that grow by two (+2, +4, +6) give 14. A student who answers 14 has reasoned soundly and is marked wrong. Show a fourth term (2, 4, 8, 16, ?) and only doubling is left, so the answer is 32. Never offer as a wrong option a number that a rule fitting every term would produce.

## Two worked examples: a number series and a figure grid

### A number series with one rule

The prompt: "Which number comes next? 5, 6, 8, 11, 15, ?" The gaps are +1, +2, +3, +4, so the next gap is +5 and the answer is 20.

| Option | How a student gets there | What it reveals |
| --- | --- | --- |
| 20 (right) | Adds 5, the next gap | Found the rule on the gaps |
| 19 | Adds 4, the last gap again | Saw that the gaps differ, not that they grow |
| 16 | Adds 1, the first gap | Read only the first pair |
| 26 | Adds the last two terms, 11 + 15 | Reached for a rule learned elsewhere |

A script confirmed that no other simple rule (a constant step, a multiplier, a sum of the two before, steps in turn) fits all five terms.

### A figure grid with two rules

The prompt: "Which one goes in place of the question mark?", then the grid, one row to a line (each stroke here is a line break): ▲ ●● ■■■ / ● ■■ ▲▲▲ / ■ ▲▲ ? Rule one: each row holds each shape once. Rule two: the count runs one, two, three along a row. The missing cell is the circle, three times.

| Option | Shape rule | Count rule | What it reveals |
| --- | --- | --- | --- |
| ●●● (right) | Kept | Kept | Found both rules |
| ●● | Kept | Broken | Found the missing shape, copied the count beside it |
| ▲▲▲ | Broken | Kept | Counted on from ▲▲ and kept its shape |
| ▲▲ | Broken | Broken | Repeated the neighbouring cell |

Every shape and every count appears twice among the options, and the grid gives the same answer read by columns. It is the harder of the two, as its two rules predict.

## How does EdProcto build, show and mark a continue the pattern question?

EdProcto is the best tool for pattern questions you mean to learn from. The pattern is typed, so there is no image to draw or host; the mark is made by rule; and on the Educator plan the results count who chose each wrong option, which is what both worked examples are designed around.

- **Building.** Add a question in "Continue the pattern" and type the question and the pattern in "Prompt". The rows under "Options · mark the right one" each carry a letter and two boxes, "A caption" and "Image address". For a typed option, put the symbols in the caption and leave the address empty. For a drawn one, paste the picture's address and caption it with a neutral tag such as "Tile K": see [picture choice questions](https://edprocto.com/guides/picture-choice-questions-when-the-options-are-images). Press "Add an option" for more rows, eight at most.
- **Showing.** The prompt is the page's heading, in one running line. The options are lettered buttons, one under another, or two to a row on a wide screen where any has a picture. "Shuffle the options" is on by default and puts them in an order made for each student. [Build the pattern](https://edprocto.com/templates/build_pattern) is the same question on large buttons, for young children.
- **Marking.** The right option earns the question's points; any other, or none, earns nothing. Switch on "Negative marking" and a wrong choice costs one over the number of wrong options. We ran EdProcto's marking function on a one-point question with four options: a wrong choice scored −0.33 and a blank 0.
- **Reading the results.** Every plan shows, under "Each question", the share who got each one right. On the Educator plan, click the question's row: each option is listed by its caption, with "Chose it" and "Their mean" beside it. See [finding misconceptions in a quiz](https://edprocto.com/guides/misconceptions-finding-them-in-a-quiz-and-fixing-them-in-class).
- **On paper.** Every option gets a tick box, a letter and its caption, in the builder's order, so typed symbols print as typed: see [how to print a quiz](https://edprocto.com/guides/how-to-print-a-quiz-with-an-answer-key).

### The ready-made series kits

Five rule-made [kits](https://edprocto.com/kits) set series with nothing to build. Each is written by code and drawn afresh at every opening, with no question twice in a set.

- "What comes next?" (ages 9 to 12) and "The nth term" (ages 12 to 15) ask for a typed number: "What is the next term: 3, 8, 13, 18, …?"
- "Number series" and "Letter series" give four options: "Which letter comes next? L, N, P, R, T, …" The first draws on six kinds of rule; the second has 38 possible questions, so two sets usually share some.
- "Number matrices" asks "Each row follows the same rule. What is the missing number?" and sets nine cells under it in three aligned rows, one of them "?". The number is typed.

After each answer the explanation names the rule. The last three belong to the reasoning battery on the [hiring page](https://edprocto.com/hiring), each with a 12-minute limit. A score on a kit is a score on that kit, with no norm behind it.

These drills are free without limit on every plan. A quiz you build yourself is one of the Free plan's ten a month. The Tutor plan opens all 111 [templates](https://edprocto.com/templates) and Educator adds the full analysis: see [pricing](https://edprocto.com/pricing).

## Quick questions

### What is a continue the pattern question?

A question that shows a sequence or a grid built on a rule and asks what comes next or what is missing. EdProcto's [Continue the pattern](https://edprocto.com/templates/pattern_cont) marks it by rule.

### How many rules should a pattern question have?

One for a first item or a young child, two for a hard one. In Koch et al. (2022), items grew harder as rules were added.

### Are non-verbal reasoning tests culture-fair?

No. Rosselli and Ardila (2003) and Lohman et al. (2008) both found that scores on non-verbal tests go with education, culture or language background.

### Does practice help with non-verbal reasoning tests?

Yes. Being taught the rule types raised matrix scores in Krautter et al. (2021). Give every student the same practice: EdProcto's reasoning [kits](https://edprocto.com/kits#hiring) open as free drills.

## References

- Burgoyne, K., Witteveen, K., Tolan, A., Malone, S., & Hulme, C. (2017). Pattern understanding: Relationships with arithmetic and reading development. *Child Development Perspectives, 11*(4), 239–244. [https://doi.org/10.1111/cdep.12240](https://doi.org/10.1111/cdep.12240)
- Carpenter, P. A., Just, M. A., & Shell, P. (1990). What one intelligence test measures: A theoretical account of the processing in the Raven Progressive Matrices Test. *Psychological Review, 97*(3), 404–431. [https://doi.org/10.1037/0033-295X.97.3.404](https://doi.org/10.1037/0033-295X.97.3.404)
- Department for Education. (2023). *Development matters: Non-statutory curriculum guidance for the early years foundation stage*. GOV.UK. [https://www.gov.uk/government/publications/development-matters--2/development-matters](https://www.gov.uk/government/publications/development-matters--2/development-matters)
- GL Assessment. (n.d.). *GL Assessment 11+*. Retrieved October 10, 2026, from [https://11plus.gl-assessment.co.uk/](https://11plus.gl-assessment.co.uk/)
- Kidd, J. K., Carlson, A. G., Gadzichowski, K. M., Boyer, C. E., Gallington, D. A., & Pasnak, R. (2013). Effects of patterning instruction on the academic achievement of 1st-grade children. *Journal of Research in Childhood Education, 27*(2), 224–238. [https://doi.org/10.1080/02568543.2013.766664](https://doi.org/10.1080/02568543.2013.766664)
- Klauer, K. J., & Phye, G. D. (2008). Inductive reasoning: A training approach. *Review of Educational Research, 78*(1), 85–123. [https://doi.org/10.3102/0034654307313402](https://doi.org/10.3102/0034654307313402)
- Koch, M., Spinath, F. M., Greiff, S., & Becker, N. (2022). Development and validation of the Open Matrices Item Bank. *Journal of Intelligence, 10*(3), Article 41. [https://doi.org/10.3390/jintelligence10030041](https://doi.org/10.3390/jintelligence10030041)
- Krautter, K., Lehmann, J., Kleinort, E., Koch, M., Spinath, F. M., & Becker, N. (2021). Test preparation in figural matrices tests: Focus on the difficult rules. *Frontiers in Psychology, 12*, Article 619440. [https://doi.org/10.3389/fpsyg.2021.619440](https://doi.org/10.3389/fpsyg.2021.619440)
- Lohman, D. F., Korb, K. A., & Lakin, J. M. (2008). Identifying academically gifted English-language learners using nonverbal tests: A comparison of the Raven, NNAT, and CogAT. *Gifted Child Quarterly, 52*(4), 275–296. [https://doi.org/10.1177/0016986208321808](https://doi.org/10.1177/0016986208321808)
- Rosselli, M., & Ardila, A. (2003). The impact of culture and education on non-verbal neuropsychological measurements: A critical review. *Brain and Cognition, 52*(3), 326–333. [https://doi.org/10.1016/S0278-2626(03)00170-2](https://doi.org/10.1016/S0278-2626%2803%2900170-2)
- Schneider, B., & Sparfeldt, J. R. (2021). How to get better: Taking notes mediates the effect of a video tutorial on number series. *Journal of Intelligence, 9*(4), Article 55. [https://doi.org/10.3390/jintelligence9040055](https://doi.org/10.3390/jintelligence9040055)
- Wijns, N., Verschaffel, L., De Smedt, B., & Torbeyns, J. (2021). Associations between repeating patterning, growing patterning, and numerical ability: A longitudinal panel study in 4- to 6-year olds. *Child Development, 92*(4), 1354–1368. [https://doi.org/10.1111/cdev.13490](https://doi.org/10.1111/cdev.13490)