Introduction. Il existe à l’heure actuelle un large panel de tests cliniques de mémoire, parmi lesquels le Hopkins Verbal Learning Test (HVLT), particulièrement approprié au suivi longitudinal des patients, puisque comportant six formes parallèles. N’existant que dans sa version originale anglo-saxonne, nous avons décidé de l’adapter en français. Méthodes. Cent quatre-vingts sujets adultes sains ont été répartis en six groupes, afin de passer chacun une des six formes du HVLT. Chaque forme est constituée d’une liste de 12 mots, contrôlés en fonction de leurs qualités lexico-sémantiques. La procédure comprend trois tâches : a) un rappel libre d’une liste de 12 mots lus oralement aux sujets (trois essais consécutifs) ; b) un rappel différé 20 minutes après (occupées par des tâches intercurrentes) ; c) puis une tâche de reconnaissance des 12 mots parmi 24. Résultats. Les sujets obtiennent des performances équivalentes aux six formes, quel que soit le type de tâche (hormis pour le rappel immédiat de la forme 3, qui est exclue du test). Conclusion. L’adaptation française du HVLT fournit un outil indispensable pour l’évaluation longitudinale de la mémoire. Les prochaines étapes seront sa validation (en déterminant la sensibilité et spécificité du test), et sa standardisation (en fournissant des normes par tranches d’âges et niveaux socioculturels).

Introduction. A number of tests are currently used in clinical and research settings for the assessment of patients with memory deficits. Among them, the Hopkins Verbal Learning Test (HVLT) is particularly appropriate for the longitudinal follow-up of patients with memory disorders because it exists in six parallel forms, and therefore avoids the risk of learning effect at retest. Since a test with these characteristics is not available in French, we decided to adapt a French version of the HVLT. Methods. 180 normal subjects participated in the study. Their mean age was 41 years (SD=11), and they had had on average 12 years of schooling (SD=3). The subjects were randomly divided into 6 groups of 30 subjects. One of the six forms of the French version of the HVLT was administered orally to each group of subjects. Each form consisted of a list of 12 words belonging to 3 different semantic categories. For the construction of the French version of the HVLT, we adopted the same procedure as used in the original version of the test taking into account the French lexical and semantic characteristics of the items. In the first part of the test, the list was administered three times to the subjects. Following each administration, subjects were asked for an immediate free recall. Twenty minutes later, used for intercurrent tasks, subjects were asked for a delayed free recall, which was immediately followed by a recognition memory task. In this task, subjects listened to a list of 24 words, 12 belonging to the studied list and 12 were distractors; the subjects were asked to recognize the 12 studied words. Results. The subjects’ performance was equivalent in the six forms of the test, except for the immediate recall of Form 3 (which was excluded from the test). No significant difference emerged in free recall, delayed free recall, and recognition across the five remaining forms of the test. Conclusion. Our study provides a useful tool for the longitudinal evaluation of patients with memory impairment and may become the test of reference in European longitudinal clinical trials. The French adaptation of the HVLT represents only a first step, because it needs to be standardized, in order to provide norms, and validated, in order to provide values of sensitivity and specificity.