MirLabs / The lab / French versus English

Planned

French versus English

Do models give in more often when the trap is in French?

Result

No result yet.

Protocol

The same traps are played in French and in English, with texts translated word for word and proofread. We compare the faults of the same model in both languages.

Limits

A translation is never perfectly neutral: a gap can come from a word that is stronger in one language than in the other.

← Back to the experiments in progress and planned