Spontaneous neuronal activity is correlated to statistical learning performance: computation of ALFF and fALFF indices on resting-state fMRI

Abstract

Statistical learning (SL) is a mechanism that enables us to detect and learn probabilistic regularities and patterns from the environment. Previous studies have explored the role of SL in resting-state functional connectivity, but none of them has focused on spontaneous neuronal activity (SNA) and whether it can predict performance at a word segmentation task. Here we compute the functional segregation indices, ALFF and fALFF, on resting-state functional MRI (rs-fMRI) data and correlate them to statistical learning performance after listening to an artificial language stream. Our results show that there is a significant negative correlation between fALFF index and SL performance after a 4-minute exposure at bilateral temporo-occipital junction. This region seems to play a role in auditory attention and speech perception and, according to our results, is relevant for statistical learning when SNA is taken into account.