MedLex-25: Diagnosing the recognition-production gap in medical English vocabulary

This study introduces MedLex-25, a compact diagnostic instrument designed to quantify recognition-production asymmetries in medical English vocabulary. Second-year medical students at Medical University Varna, Bulgaria completed nine parallel test forms (50 targets per form; 225 unique lexical targets sampled from a coursebook corpus) under supervised conditions; usable data from 186 students were analysed, with a total of 9300 questions answered. Each form paired a four-option receptive item with a sentence-production task for the same targets; production responses were binary-scored (correct/incorrect) with tolerance for non-meaning-preventing surface errors. Results suggest a systematic recognition-production gap across five lexical categories (nouns, verbs, adjectives, adverbs, two-word multiword items): receptive accuracy consistently exceeded productive accuracy, and no item achieved perfect production. Counts of perfectly recognized items were: nouns 33/45, adjectives 27/45, verbs 23/45, adverbs 25/45, and multiword items 28/45. Verbs, adjectives and, most strikingly, adverbs exhibited the largest productive weaknesses (lowest production for erroneously = 0.071). Item-level diagnostics identify concrete teaching targets. The MedLex-25 test can guide focused instruction and recommend follow-up studies that expand item coverage, refine production scoring, and investigate etymology and cross-disciplinary exposure as mediating factors.