Find a task to explore further
Remembering locations, ordered items and digit strings calls for different approaches. A stronger task result is a useful starting point, not proof of a fixed talent, personality or complete memory ability.
Memory: Understand Your Three Task Results
Positions, sequences and digits are different challenges. Discover which task feels strongest for you.
Loading dimension detail data…
UNDERSTAND YOUR RESULTS
What do the three memory tasks show?
The memory report brings together visual, sequence and number tasks. Their units and rules differ, so interpret each result and its history separately. Levels, sequence lengths and digit spans should not be averaged as raw numbers.
Latest and best results both matter
The latest record describes your recent attempt; a personal best preserves a previous peak. Compare multiple matching sessions rather than treating one peak as your stable level.
Interpret tasks before combining them
Each task adds a different observation. Even after mapping scores to a common reference scale, their average does not automatically become a calibrated population percentile. Read estimates alongside raw results and methods.
How the memory tasks differ
| Test | Main task | How to read the result |
|---|---|---|
| Visual memory | Recall the positions of highlighted tiles | Review the level reached against matching history |
| Sequence memory | Reproduce items in their displayed order | Review the sequence length reproduced successfully |
| Number memory | Recall and reproduce a digit string | Review the number of digits reproduced successfully |
Memory score questions
Do all three tests measure the same memory ability?
Not exactly. Locations, order and digit material impose different demands, and results can also reflect attention, strategy and interaction conditions. Together they offer several task observations.
Can I read a report after only one test?
Yes. Review existing records and add the other tasks later. Coverage tells you which tasks are present; it does not invalidate a single result or establish a validated composite assessment.
Why cannot I directly compare a sequence of 17 with 15 digits?
Materials, rules and difficulty differ, so the numbers are not the same scale. Compare matching tasks first. A global composite rank requires an applicable sample and combination method.
How can I keep retests comparable?
Keep device, difficulty and protocol consistent, and avoid notes or screenshots. Consider changes in strategy. One change does not establish a permanent gain or loss in memory ability.
Where do scores and references come from?
Start with recorded scores. Historical coverage and comparable results may count different records.
Read scoring and record notes
Scores come from saved attempts
Milliseconds, levels and item counts come from test records saved in this browser. Latest and personal-best scores are different observations, not fixed abilities.
References are not measured global ranks
Current reference scales use provisional project rules, not validated population norms. Estimated Top does not describe an observed population proportion; combining reference scores does not establish an overall population percentile.
Why can overview and detail counts differ?
The ordinary Profile overview and dimension details use the same production record-eligibility rules. Historical records remain visible, but records without comparable conditions do not contribute to reference scores. Saved history is not automatically a valid same-condition result. Distinguish all linked tests from the core tests contributing to a dimension.
Match conditions before comparing
Prefer the same task, mode, version and input method. Similar dates or scores alone cannot establish comparability when these conditions were not saved.
Personal records stay in the browser
These notes do not read personal scores. Public HTML and page structured data contain no personal records. Sharing the page URL does not transfer your locally saved results.
Current calculation and interpretation scope by dimension
Memory
Keep the existing task reference scores and combination rule. Averaging references does not create a memory percentile calibrated against a joint population distribution.
Notes reviewed: 2026-10-21 · public-method-v1 (notes version, not a calculation model version)