Neural correlates underlying statistical learning of adjacent and non-adjacent verbal sequential dependencies
Leyla Eghbalzad1, Joanne A. Deocampo1, Gretchen N.L. Smith2, Sabrina Na1, Tricia Z. King1, Christopher M. Conway1; 1Georgia State University, 2Indiana University School of Medicine
The ability to learn sequential dependencies is essential for language acquisition and other cognitive skills. Recent studies suggest there may be separate cognitive processes involved in learning adjacent (e.g., “A-B”) versus non-adjacent (e.g., “A-X-B”) dependencies, but the neural correlates accompanying such learning are under-specified. We developed a sequential learning task in which sequences of printed nonsense syllables containing both adjacent and non-adjacent dependencies were presented. After incidentally learning these grammatical sequences, eighteen healthy adults (age M=22.5, 9 females) made familiarity judgments about novel grammatical sequences and ungrammatical sequences containing violations of the adjacent or non-adjacent dependencies while in a 3T MRI scanner. Analysis of the BOLD activity showed that increased activation for adjacent dependency learning was associated with a distributed frontal –parietal and cerebellar network, whereas increased activation for non-adjacent dependency learning was associated with the anterior cingulate cortex (ACC). The frontal-parietal network is known to be associated with working memory while the ACC is proposed to be important for cognitive control, error/conflict detection, as well as allocation of attention and selection of appropriate responses. Furthermore, these networks were differentially correlated with distinct out-of-scanner cognitive measures such as working memory and processing speed. These findings provide the basis for understanding the neural underpinnings of sequential pattern learning of adjacent and non-adjacent structures and helps elucidate the possible mechanisms important for the processing and learning of language.
Topic Area: METHODS: Neuroimaging