TopperLens Junior Basics mastered. Future unlocked.

For parents

What we measure, why it matters, and how practice moves it

Three tests, each built on a different body of research. This page explains what each one is actually looking at, why that skill matters for your child's school years, and exactly how the practice engine is designed to improve it — including the parts we deliberately don't claim.

01 · Cognitive Aptitude Test

How your child thinks — across six separate abilities

CHC theory Three-stratum model Compared only to themselves

The theory it's built on

The test is structured around CHC theory — Cattell–Horn–Carroll — the most extensively researched model of how human cognitive ability is organised. It began with Raymond Cattell's distinction between fluid reasoning (solving something new) and crystallised knowledge (what you've already learned), was extended by John Horn, and was consolidated by John Carroll's landmark 1993 re-analysis of more than 460 datasets gathered over six decades.

Its central finding is the one that shapes this whole test: cognitive ability is not one thing. It resolves into distinct abilities that can sit at very different levels inside the same child. This is why modern assessment batteries report a profile rather than a single number — and why we do too.

Why a profile beats a single score

A child with strong reasoning but a narrow working memory looks "average overall" on a single score — and that average hides both the talent and the obstacle. The profile is what makes the result actionable: it tells you which ability to support, not just how your child ranks.

Every domain is scored ipsatively — measured against your child's own average across the six, never against other children. A "growth area" means it's lower for them, which is exactly the comparison that tells you where help goes furthest.

The six domains — and why each one matters at school

Logical thinking Gf — fluid reasoning

Working out something they've never been taught. This is the ability that carries a child through unfamiliar material, and it matters most in the years when new concepts arrive faster than they can be memorised.

Verbal & language Gc — crystallised knowledge

Vocabulary, comprehension, and knowledge already acquired. It underpins reading across every subject — including the word problems in maths — and it's the domain that responds most directly to reading and conversation at home.

Working memory Gwm

Holding information in mind while doing something with it. This is what a child uses to follow multi-step instructions, carry a number during mental arithmetic, or take notes while still listening. When it's narrow, a child who genuinely understands can still lose the thread halfway through.

Quick thinking Gs — processing speed

How quickly routine work gets done accurately. Low processing speed is one of the most commonly misread signals in school: homework takes twice as long, and it gets attributed to laziness when the understanding was never the problem.

Spatial thinking Gv — visual processing

Reasoning about shape, position and space. It shows up in geometry, diagrams, maps and graphs, and it's the ability that later underlies design, engineering and the sciences.

Memory & recall Glr — long-term retrieval

Storing something today and getting it back weeks later. Exams don't test what a child understood in the lesson; they test what can still be retrieved months on.

How practice is linked

The report names the lowest domain relative to your child's own average, and the recommendation links straight into practice targeted at that domain — not a general set. Retaking the test later re-measures the same six domains, so the change is measured on the same ruler rather than estimated.

02 · Spelling Test

Hear it, type it — because recognising a word isn't knowing it

Orthographic mapping Self-teaching hypothesis Transcription fluency

The theory it's built on

Skilled spelling isn't visual memory for word shapes. Research on orthographic mapping — most associated with Linnea Ehri — describes how children bond the sounds in a spoken word to the letters that represent them. Once that bond forms, the word is stored for good and can be both read and spelled without effort.

David Share's self-teaching hypothesis explains why practice compounds: each time a child successfully works a word out, that attempt itself lays down the stored spelling. The child teaches themselves, one word at a time — which only happens if they're actually attempting words.

Why we use dictation, not multiple choice

Choosing the right spelling from four options only requires recognition. Hearing a word and typing it requires production — translating sound to letters with no cue on screen. That's a harder task, and deliberately so: it's the one that reveals whether the mapping has actually formed.

It's also why the test records how often a word was replayed and how much it was edited. A word spelled correctly on the fourth listen with six corrections is not the same as one typed straight through, and the report treats them differently.

Why spelling matters more than it looks

It frees the mind to write

Virginia Berninger's work on transcription fluency found that when getting words onto the page is effortful, the effort is paid for out of the same limited working memory the child needs for composing ideas. A child struggling to spell doesn't write simpler sentences because they think simply — they write simpler sentences because there's no capacity left. Automatic spelling gives that capacity back.

It shares a store with reading

Spelling and reading draw on the same underlying knowledge of how words are built. Strengthening the letter-sound patterns behind spelling tends to support reading fluency too — which is why the seven skills tested here are patterns and rules, not an arbitrary word list.

How practice is linked

The test reports accuracy across seven distinct skills — sight words, phonics, silent letters, double consonants, prefixes and suffixes, context spelling, and listening dictation — so a weakness is a pattern, not a list of missed words. Practice targets the specific pattern, which is what generalises to words your child has never seen.

03 · Maths Speed Test

Why speed on the basics decides how hard the hard problems feel

Cognitive load theory Spaced repetition Interleaving The 85% rule

The theory it's built on

John Sweller's cognitive load theory starts from a hard limit: working memory holds very little at once. So when a child meets a multi-step problem, every basic fact they have to work out rather than recall consumes part of that limited space — and it's taken from the same pool needed to follow the actual method.

This is the single most useful idea on this page. A child who computes 7 × 8 each time isn't bad at maths; they're solving the problem with less capacity than a classmate who simply knows it. Making the basics automatic doesn't just save seconds — it hands back the room to think.

Why the test is timed and mixed

Accuracy alone can't tell the difference between recalling a fact and reconstructing it — both end in the right answer. Speed is what separates them, which is why every answer is timed against an age-appropriate fluency allowance.

The problems are also interleaved rather than grouped by type. Rohrer and Taylor's research found that mixing problem types makes practice feel harder and look worse in the moment, while producing markedly better retention later — because the child must work out which method applies, not just apply the one from the top of the page.

What the practice engine actually does

Not a general description — this is the mechanism running behind your child's maths practice.

1

Every fact gets its own memory schedule

Each number fact is tracked separately through six stages. A fact your child knows cold and one they keep missing are never treated the same way, because they aren't the same problem.

2

Review is spaced at widening intervals

Once a fact is holding, it comes back after 1 day, then 3, then 7, then 21. The spacing effect — one of the oldest and most replicated findings in memory research — shows that reviews spread over time produce far more durable recall than the same practice packed together.

3

Correct isn't enough — it has to be fast

A fact only advances when it's answered correctly and inside the fluency allowance for your child's age. Correct but slow holds its place rather than moving up, because slow-and-correct means it's still being computed, not recalled. A mistake sends it back to the start.

4

Difficulty is tuned to about 85% success

The session tracks rolling accuracy and adjusts. Work that's too easy teaches nothing; work that's too hard causes children to give up. Research on optimal challenge points to roughly 85% success as the fastest place to learn — hard enough to matter, successful enough to keep going.

5

Facts are retired, not drilled forever

Once a fact is answered fluently several times across different sessions — not several times in one lucky run — it's retired from active practice and only reappears as an occasional warm-up. Practice time keeps moving to what isn't mastered yet.

04 · General Knowledge Test

The one test here with no age standard — and why we say so

Curriculum-referenced Retrieval practice The spacing effect No norm claimed

Why there is no “should know” here

The other three tests compare a child to a standard we can defend. This one cannot, and we would rather say it than dress it up. CBSE prescribes no General Knowledge syllabus — no NCERT textbook, no learning outcomes. What a school teaches under that name comes from whichever private publisher it happens to buy.

Nor is there national data to fall back on. The last time this content was measured at Class 3 was the National Achievement Survey of 2021; its successor, PARAKH Rashtriya Sarvekshan, dropped it in 2024, and now assesses nothing of the kind until Class 6. Anyone quoting you an age percentile for general knowledge has invented it.

What we measure instead

Every question carries the NCERT class its content is introduced in. That turns a vague subject into a checkable one: instead of a mark out of ten, the report says how much of the ground already covered by your child’s own class came back.

Questions from later classes are included too, but scored separately as reach. Missing something a child has not been taught yet is not a gap, and counting it as one would penalise a younger child for the test having headroom.

Why testing beats re-reading

The strongest finding in the memory literature is that retrieving something strengthens it far more than reviewing it. Roediger and Karpicke’s work on the testing effect found students who were tested on material remembered dramatically more a week later than students who simply studied it again — even though the re-readers felt more confident.

That is what practice here is for. Each question is an act of retrieval, and the answer with its one-line explanation arrives immediately, which is the condition under which retrieval practice works best.

The honest caveat about exposure

General knowledge tracks a child’s environment more closely than anything else we measure: books in the house, travel, which languages are spoken at dinner, time online. A child who has been read to about volcanoes will answer the volcano question.

So a lower score here is frequently a fact about circumstances rather than about a child, and the report says so on its first page. We would rather a parent read the number correctly than be impressed by it.

Why short and regular beats long and rare

This isn't encouragement — it's how the method is built. Three of the four principles below only work if your child comes back.

Spacing needs a return visit

The review intervals are 1, 3, 7 and 21 days. A fact scheduled to come back on day 3 can only be strengthened if there's a session on day 3. Without return visits the schedule never runs, and facts slide back to the beginning — the engine is doing nothing while nobody is there.

Retrieving is what builds memory

The testing effect — Roediger and Karpicke's work is the best-known demonstration — shows that pulling an answer out of memory strengthens it far more than reviewing it does. Reading the report teaches your child nothing. Answering twenty questions does.

Twenty minutes beats two hours

The same total practice produces more durable learning split across several short sessions than concentrated into one long one. Three fifteen-minute sessions in a week are worth more than a forty-five-minute Sunday — and they're far easier to actually keep.

The report gets sharper with use

Observations need evidence: we won't call something a strength or a weakness on fewer than three questions. More sessions mean more per-question detail, which means the patterns we can show you — where they rush, where they slow down — get more specific and more reliable.

What we don't claim

A page about scientific grounding should be equally clear about its limits. Everything above describes the research these tests are designed around — that is not the same as a claim that this is a clinical instrument.

  • This is not an IQ test, and it doesn't produce an IQ score. CHC theory is the framework it borrows its structure from, not a certification.
  • It is not a diagnosis. Nothing here can identify dyslexia, dyscalculia, ADHD or any other condition. If something in a report concerns you, or your instincts already have, please talk to a qualified professional — this is not a substitute for that conversation.
  • We don't rank your child. There is no percentile against a national sample and no comparison to other children anywhere in the product. Every score compares your child to their own earlier attempts, on purpose.
  • A score describes a test, on a day. Tiredness, a rushed start and a noisy room all show up in results. That's exactly why we look at trends across attempts rather than treating one number as a verdict.
  • We won't invent a conversion rate. You'll never see us promise that a set number of practice questions raises a score by a set number of points. We don't have the data to support a claim like that, so we don't make it — projections in your reports are arithmetic on your own child's latest attempt, and they say so.

Ready to put this to work?

Head back to the portal — the recommendations tab turns everything on this page into the next specific thing to do.

Back to my portal