Editorial policy

How the guides get chosen, checked, and kept honest.

Hearkle's Learn section exists to make the games make sense. This page sets out how those guides are put together and where the standards sit, so you can judge for yourself how much to trust what you read here.

How topics are chosen

Almost every guide is paired to a game or a skill the games train. If a game asks you to repeat a tone from memory, there's a guide on how that kind of memory works and how to practise it. The topics aren't picked for search traffic first. They're picked because someone playing Hearkle would reasonably want to understand what they just did, or how to get better at it. When a concept keeps coming up across several games, like just-noticeable differences or intervals, it earns its own deeper piece.

What sources we prefer

The order of preference is deliberate. Primary and peer-reviewed research comes first: the actual studies, plus recognised standards and reference works in hearing and music cognition. Reputable general references, like well-maintained encyclopaedia entries and established educational sites, come next, mostly to frame or summarise. Secondary blogs and opinion pieces are a last resort, used only when nothing better covers the point, and flagged as the lighter-weight source they are. Where a guide leans on a specific claim, we try to point you at something you can check yourself.

How claims are checked and kept modest

Hearing and pitch-perception research is real, but it's rarely as tidy as a headline wants it to be. So the writing hedges on purpose. Research "suggests" or "tends to show"; it doesn't "prove" you'll gain a fixed amount in a fixed time. We avoid guaranteed-result claims, round numbers dressed up as certainties, and any promise that practising with Hearkle will fix a specific problem. If the evidence is mixed or thin, the guide says so. Overclaiming would be easy and it would be dishonest, so we'd rather tell you what's likely and where the uncertainty lives.

Using AI

Some guides are drafted with help from AI tools. That's worth being open about. AI can speed up a first pass, but it also invents things, so nothing goes live on the strength of a draft alone. Every guide is read, fact-checked, and edited by a person for accuracy and for voice, claims are traced back to real sources, and anything that can't be supported is cut. The person behind Hearkle is accountable for what's published, whatever tool helped produce the first version. If you find a mistake, it's a human who owns it and fixes it.

Translation

Hearkle's guides start in English, and that English version is the source of truth. Localized versions are produced from it and then reviewed rather than published raw, so a translation that drifted from the meaning gets caught and corrected. Where a claim or a nuance is hard to carry across a language, we favour keeping the meaning honest over keeping it word-for-word. Any correction to the English source is meant to flow through to the translations too.

Corrections

Mistakes happen, and we fix them in the open rather than quietly. If you spot an error, the corrections page explains how to report it and how substantive fixes get logged with a date. Reader reports are one of the better ways problems get caught, so they're welcome.

How often content is updated

There's no fixed publishing calendar. Guides are revised when the underlying research or our own understanding moves, when a game they describe changes, or when a reader flags something wrong or unclear. Older guides get a fresh read from time to time so they don't quietly go stale. The aim isn't churn for its own sake; it's keeping what's already here accurate.

More about the project