Executive functioning in adults with down syndrome: Machine-learning-based prediction of inhibitory capacity

Mario Fernando Jojoa-Acosta, Sara Signo-Miguel, Maria Begoña Garcia-Zapirain, Mercè Gimeno-Santos, Amaia Méndez-Zorrilla, Chandan J. Vaidya, Marta Molins-Sauri, Myriam Guerra-Balic, Olga Bruna-Rabassa

Producció científica: Article en revista indexadaArticleAvaluat per experts

10 Cites (Scopus)

Resum

The study of executive function decline in adults with Down syndrome (DS) is important, because it supports independent functioning in real-world settings. Inhibitory control is posited to be essential for self-regulation and adaptation to daily life activities. However, cognitive domains that most predict the capacity for inhibition in adults with DS have not been identified. The aim of this study was to identify cognitive domains that predict the capacity for inhibition, using novel data-driven techniques in a sample of adults with DS (n = 188; 49.47% men; 33.6 ± 8.8 years old), with low and moderate levels of intellectual disability. Neuropsychological tests, including assessment of memory, attention, language, executive functions, and praxis, were submitted to Random Forest, support vector machine, and logistic regression algorithms for the purpose of predicting inhibition capacity, assessed with the Cats-and-Dogs test. Convergent results from the three algorithms show that the best predictors for inhibition capacity were constructive praxis, verbal memory, immediate memory, planning, and written verbal comprehension. These results suggest the minimum set of neuropsychological assessments and potential intervention targets for individuals with DS and ID, which may optimize potential for independent living.

Idioma originalAnglès
Número d’article10785
RevistaInternational Journal of Environmental Research and Public Health
Volum18
Número20
DOIs
Estat de la publicacióPublicada - 1 d’oct. 2021

Fingerprint

Navegar pels temes de recerca de 'Executive functioning in adults with down syndrome: Machine-learning-based prediction of inhibitory capacity'. Junts formen un fingerprint únic.

Com citar-ho