<p>The assessment of working memory (WM) in healthy older adults has become a central topic in cognitive research due to the sustained growth of this population and the widespread use of WM tasks as indicators of cognitive functioning and training outcomes. However, the existing literature presents inconsistent evidence. The aim of this scoping review was to map the tasks used to assess working memory in healthy older adults, describe the available evidence on their psychometric properties, and examine the conditions under which near and far transfer measures have been evaluated. A scoping review was conducted following PRISMA-ScR guidelines and COSMIN criteria adapted for experimental tasks, with systematic searches in Scopus, ScienceDirect, Academic Search Complete, Psychology and Behavioral Sciences Collection, and Google Scholar. Empirical studies and reviews resulting in a total of 34 eligible studies. The results showed that complex span tasks exhibit the most consistent psychometric profile, whereas widely used paradigms such as the n-back and alternative updating tasks present limited evidence of reliability, measurement error, and responsiveness. Near transfer was strongly dependent on structural similarity between tasks, while far transfer was generally absent or weak. Overall, the findings reveal marked methodological and psychometric heterogeneity in the assessment of working memory in older adults, highlighting the need to improve psychometric rigor and the interpretation of gains observed in training and assessment contexts.</p>