Neural and Behavioural Signatures of Statistical Learning in Aging: Evidence from Fast Periodic Visual Stimulation

Statistical learning (SL) enables the extraction of regularities from structured input and is thought to support visual word processing through domain-general mechanisms. Using a Fast Periodic Visual Stimulation (FPVS) paradigm, we investigated whether frequency- based regularities are implicitly learned in healthy aging and how such learning is expressed at neural and behavioural levels. Fifty-one older adults and a comparison group of young adults were tested with EEG and behavioural tasks, alongside an independent visual SL (VSL) measure. Results revealed robust neural discrimination of frequency-based categories in older adults, indicating preserved implicit sensitivity to regularities. However, neural responses were reduced in amplitude compared to young adults. Behaviourally, participants showed limited evidence of frequency-based learning, and no reliable relationship emerged between neural and behavioural indices. In contrast, VSL performance declined with age and was associated with behavioural discrimination, but not with neural responses. These findings suggest a dissociation between implicit neural sensitivity and explicit behavioural expression of SL in aging, highlighting the importance of integrating neural and behavioural measures to capture age-related changes in learning.