[PS-3.10] Investigating the link between statistical learning and language in older healthy adults and in stroke patients with aphasia

Schevenels, K. ¹ , De Smedt, B. . ² , Zink, I. ¹ & Vandermosten, M. ¹

1 Experimental Oto-Rhino-Laryngology, Department of Neurosciences, KU Leuven, Leuven, Belgium
2 Parenting and Special Education Research Unit, Faculty of Psychology and Educational Sciences, KU Leuven, Leuven, Belgium

Statistical learning (SL) is crucial for language acquisition, however, it is unknown whether SL is important for relearning language in aphasia after stroke, which typically affects elderly. Current evidence suggests that SL abilities decline with age and that SL behaves independently in different modalities and types of information. Therefore, we first investigated which of five learning paradigms was sensitive enough to tap the SL ability of elderly. We tested 25 young and 25 older adults on two types of SL (conditional and distributional) in two modalities (auditory and visual) and an implicit sequence-learning task. Second, we used the selected test(s) to examine whether SL is related to language recovery in persons with aphasia (N = 15). Healthy persons did not show SL for the distributional tasks, whereas the conditional tasks elicited learning at an individual level in the younger group and at group level in elderly. The sequence-learning task evoked learning without effects of age. Consequently, we simplified the conditional and sequence-learning tasks to achieve learning at an individual level in elderly. Data collection in persons with aphasia is ongoing, but we will relate their SL results to language measures collected at 1 week and 3-4 months post-stroke.