The interplay between executive functions and updating predictive representations
Felipe Pedraza,Teodora Vekony,Bence Csaba Farkas,Frederic Haesebaert,Romane Phelipon,Imola Mihalecz,Karolina Janacsek,Barbara Tillmann,Royce Anders,Gaen Plancher,Dezso Nemeth
DOI: https://doi.org/10.1101/2024.12.05.626969
2024-12-07
Abstract:Modifying habits, particularly unwanted behaviors, is often challenging. Cognitive research has focused on understanding the mechanisms underlying habit formation and how habits can be rewired. A key mechanism is statistical learning, the continuous, implicit extraction of probabilistic patterns from the environment, which forms the basis of predictive processing. However, the interplay between executive functions (EF) and the rewiring - or updating - of these probabilistic representations remains largely unexplored. To address this gap, we conducted an experiment consisting of four sessions: 1) Learning Phase - acquisition of probabilistic representations, 2) Rewiring Phase - updating these probabilistic representations, 3) Retrieval Phase - accessing learned representations, and 4) EF assessment, targeting five key aspects: attentional control, inhibition, working memory, flexibility, and verbal fluency. We focused on the relationship between these EF measures and the updating of previously acquired knowledge using an interindividual differences approach. Our results revealed a positive relationship between rewiring and inhibition, suggesting that better inhibitory control may facilitate the adaptive restructuring of probabilistic predictive representations. Conversely, a negative relationship was identified between rewiring and semantic fluency, implying that certain underlying aspects of verbal fluency tasks, such as access to long-term memory representations, may hinder the updating process. We interpret this relationship through the lens of competitive memory network models. Our findings indicate that the rewiring of implicit probabilistic representations is a multifaceted cognitive process requiring both the suppression of proactive interference from prior knowledge through cognitive inhibition and a strong reliance on model-free functioning.
Neuroscience