Domain-general mechanism relationship with non-canonical sentence comprehension

Purpose: Canonical sentence structures are the most frequently used in a given language. Less frequent or non-canonical sentences tend to be more challenging to process and to induce a higher cognitive load. To deal with this complexity several authors suggest that not only linguistic but also non-linguistic (domain-general) mechanisms are involved. In this study, we were interested in evaluating the relationship between non-canonical oral sentence comprehension and individual cognitive control abilities. Method: Participants were instructed to perform a sentence-picture verification task with canonical and non-canonical sentences. Sentence structures (i.e., active or passive) and sentence types (i.e., affirmative or negative) were manipulated. Furthermore, each participant performed four cognitive control tasks measuring inhibitory processes, updating in working memory, flexibility and sustained attention. We hypothesized that more complex sentence structures would induce a cognitive cost reflecting involvement of additional processes, and also that this additional cost should be related with cognitive control performances. Results: Results showed better performances for canonical sentences compared to non-canonical ones supporting previous work on passive and negative sentence processing. Correlation results suggest a close relationship between cognitive control mechanisms and non-canonical sentence processing. Conclusion: This study adds evidence for the hypothesis of a domain-general mechanism implication during oral language comprehension and highlights the importance of taking task demands into consideration when exploring language comprehension mechanisms.