Reconciling two computational models of working memory in aging

Violette Hoareau (1) , Benoit Lemaire (1) , Sophie Portrat (1) , Gaën Plancher (2)

1 LPNC - Laboratoire de Psychologie et NeuroCognition
2 EMC - Laboratoire d'Etude des Mécanismes Cognitifs

It is well known that working memory performance changes with age. Two recent computational models of working memory , TBRS* and SOB-CS, corresponding to two distinct causes of forgetting, namely time-based decay and interference, are applied on a set of complex span data produced by young and older adults. As expected, these models are unable to account for the older adult data. An investigation on the effect of the main parameters of these models showed that the poorer performance of older adult does not come from a weaker encoding of items, or even a longer time spent on distractors, but rather on difficulties during the free time that immediately follows each distractor, as well as a higher level of confusion between items. These results are discussed with respect to the current theories of working memory and aging.

Mots clés: SOB-CS; TBRS; aging; working memory; computational modeling

Domaines: Sciences cognitives

Violette Hoareau, Benoit Lemaire, Sophie Portrat, Gaën Plancher. Reconciling two computational models of working memory in aging. 13th International Conference on Cognitive Modeling, N. Taatgen; M. van Vugt; J. Borst; K. Mehlhorn, Apr 2015, Groningen, Netherlands. pp.130-135. ⟨hal-01144141⟩