Abstract Grammar in Broca’s area

The brain basis of abstract grammar, the feat that allows language to express infinite thoughts with a finite menu of words, has been notoriously difficult to identify because many other faculties vary with grammatical complexity (e.g., working memory, integration costs, etc). Therefore it is hard to compare grammatically complex sentences to grammatically simple ones, and assess the differences in brain activity.

Meanwhile, discovering the core function of Broca’s area, arguably the most widely known brain center, has defied 150 years of research: in part due to tasks that varied multiple faculties, and in part due to methods with low temporal or spatial resolution (like fMRI or MEG).

This paper uses inflection as a model system for grammar that sidesteps many confounds (see the task description), and event-related fMRI to identify the neural territories involved. The goal was to identify these neural territories with fMRI then investigate computations within them further, using the ICE method.

We asked whether Broca’s area computes grammar per se, what regions are responsible for inflecting vs. pronouncing words, and if nouns and verbs are inflected by the same neural processors or separate ones. The paper that reports the results is here.

This paper introduced the language task I have used in several other studies, and began several threads of continuing research (on grammatical processing, noun-verb differences, the nature of Broca’s area, and the components of neurolinguistic processing underlying production). Also, the paper may be useful in the classroom, because it includes a review-length introduction as well as discussion section, which provide literature review and technical background material that may suit a psycholinguistics or fMRI-methods course.