A New Test for the Assessment of Working Memory in Clinical Settings: Validation and Norming of a Month Ordering Task

• Published in: Psychological assessment 2018-10, Vol.30 (10), p.1367-1381
• Main Authors: Buekenhout, Imke, Leitão, José, Gomes, Ana A
• Format: Article
• Language: English
• Subjects: Adolescent; Adult; Aged; Aged, 80 and over; Cognition & reasoning; Female; Human; Humans; Male; Memory; Memory, Short-Term; Middle Aged; Psychometrics; Quantitative psychology; Reference Values; Reproducibility of Results; Short Term Memory; Task; Test Norms; Test Reliability; Test Validity; Validity; Young Adult
• ISSN: 1040-3590; 1939-134X
• DOI: 10.1037/pas0000585

Month ordering tasks have been used in experimental settings to obtain measures of working memory (WM) capacity in older/clinical groups based solely on their face validity. We sought to assess the appropriateness of using a month ordering task in other contexts, including clinical settings, as a psychometrically sound WM assessment. To this end, we constructed a month ordering task (ucMOT), studied its reliability (internal consistency and temporal stability), and gathered construct-related and criterion-related validity evidence for its use as a WM assessment. The ucMOT proved to be internally consistent and temporally stable, and analyses of the criterion-related validity evidence revealed that its scores predicted the efficiency of language comprehension processes known to depend crucially on WM resources, namely, processes involved in pronoun interpretation. Furthermore, all ucMOT items discriminated between younger and older age groups; the global scores were significantly correlated with scores on well-established WM tasks and presented lower correlations with instruments that evaluate different (although related) processes, namely, inhibition and processing speed. We conclude that the ucMOT possesses solid psychometric properties. Accordingly, we acquired normative data for the Portuguese population, which we present as a regression-based algorithm that yields z scores adjusted for age, gender, and years of formal education. Public Significance Statement We discuss the ucMOT, a new instrument for the assessment of working memory, and provide validation and reliability results as well as regression-based norms for the Portuguese population. In the ucMOT, increasingly longer lists of months out of calendar order are read to the test taker, who repeats them in the correct sequence. The task is suitable for patients with significant cognitive impairments and/or individuals with very low educational attainment.