79

This article presents a new episodic memory task (IMA-12) whose principal interests are to take the level of French lexical knowledge into consideration and to allow assessment of the encoding, retrieval, and consolidation processes. It is ultimately aimed to be used with populations for whom French is neither their mother tongue nor their language of schooling. It uses exactly the same procedure as the GERIA-12 (itself following the same procedure as FCSRT tasks), but words are replaced by line-drawings of objects. Moreover, incorrect naming (due to level of mastery of French or another reason) during the encoding phase is accepted and becomes part of the items to be remembered. In this way, imprecise lexical knowledge should not impact recall. Normative data were collected on 393 French-speaking participants distributed in 3 age groups (60-69, 70-79, and 80-89 y.), 4 education levels, and gender. Gender affected only free recalls (women > men). Education affected only free recall on trial 2. However, age affected several scores (free recalls, total delayed recall, and consolidation indexes): the younger group (60-69) obtained better scores than the two older ones (70-79 and 80-89), which did not differ from each other. As a Bayesian analysis did not show strong evidence in favor of equivalence according to demographic variables, the normative data were split according to age, education, and gender. They were presented as cut-off scores calculated from the median and the median absolute deviation from the median (to take into consideration the non-normal distribution of all scores and their skewness). As norms were only collected on French-speaking and literate participants, the applicability of this new task to people who are illiterate or for whom French is not their mother tongue remains to be established.