Statistical learning is lasting and consistent over time

Joanne Arciuli, Ian Craig Simpson

    Research output: Contribution to journalArticlepeer-review

    35 Citations (Scopus)


    Implicit detection of statistical regularities is thought to be a ubiquitous facet of cognition; yet, we know little about statistical learning (SL) over time. A recent study showed that visual SL can be observed at 24. h post stimulus (Kim et al., 2009 [14]). Here we sought to obtain a finer-grained picture of visual SL over time. We employed an embedded triplet paradigm and delayed presentation of the surprise test phase, in relation to the initial familiarisation phase, across five time periods: 30. min, 1. h, 2. h, 4. h and 24. h. Results revealed a significant degree of SL at each delay period. Moreover, the degree of SL was consistent across the five delay periods. These results suggest that visual SL is remarkably consistent over time. It does not appear to be fragile and does not appear to be enhanced by sleep in healthy adults. This robustness is desirable in a mechanism thought to underpin a broad range of mental activities including language processing. Future research might use the methodology we report here to examine whether similarly stable levels of SL can be observed in individuals with language impairment, such as those with SLI and dyslexia, compared with typical peers.

    Original languageEnglish
    Pages (from-to)133-135
    Number of pages3
    JournalNeuroscience letters
    Issue number2
    Publication statusPublished - 31 May 2012


    • Consolidation
    • Implicit learning
    • Sleep
    • Statistical learning (SL)
    • VSL


    Dive into the research topics of 'Statistical learning is lasting and consistent over time'. Together they form a unique fingerprint.

    Cite this