← All method guides

Useful comparisons

What Metkagram Can Claim — and What Still Needs Testing

A language-learning product can cite good research and still make a bad scientific claim. The missing step is often attribution: evidence for retrieval practice, spacing or textual enhancement is quietly presented as evidence for the entire product that combines them. Metkagram deliberately keeps four layers separate — what the method does, why its design is plausible, what we predict, and what direct experiments have actually shown.

Method description is the easiest layer

We can state directly what Metkagram does: it keeps meaningful sentences visible, places reviewed Marks inside examples, extracts language-specific Frames, links them to communicative Moves, builds Contrasts and Choices, and supports reuse and reviewed cross-language Bridges.

Those are observable properties of the system. They do not require a learning-outcome claim and should not be dressed up as one.

Design rationale connects components to external research

Research on textual enhancement informs the decision to make selected form visually salient. Formulaic-language research provides context for treating recurring multiword structures and frames as meaningful units. Retrieval-practice and spacing research inform the decision to require recall and to return to material over time.

These connections make the design more reasoned than random. They still do not establish that the full Mark–Frame workflow produces a particular improvement, because the combined system is a new intervention with its own implementation details.

Hypotheses should be specific enough to fail

A useful hypothesis is not “Metkagram makes the brain learn better.” It can ask whether inline Marks change attention to a target structure, whether a Contrast plus Choice improves later discrimination, or whether spaced retrieval of Frames improves delayed reuse compared with rereading under a defined task.

Specific hypotheses tell us what data to collect and what result would count against the idea. That is much more informative than attaching the word science to a feature list.

Direct evidence requires direct testing

Metkagram’s public evidence policy currently declares no efficacy claim for the complete method. That is not a weakness to hide; it is the correct state until comparative studies support stronger wording. The research section can publish protocols, pilot results and limitations as the evidence grows.

The credibility advantage is practical: when a claim eventually changes, the project can point to the study that changed it instead of relying on marketing language that was stronger than the data from the start.