Presentation
Search Abstracts | Symposia | Slide Sessions | Poster Sessions | Lightning Talks
Broca is alive and well: an articulation-selective area in the left inferior frontal gyrus, distinct from nearby language and Multiple Demand areas.
Poster D9 in Poster Session D, Wednesday, October 25, 4:45 - 6:30 pm CEST, Espace Vieux-Port
Agata Wolna1, Jakub Szewczyk2, Michele Diaz3, Aleksandra Domagalik4, Marcin Szwed1, Zofia Wodniecka1, Evelina Fedorenko5; 1Institute of Psychology, Jagiellonian University, Kraków, Poland, 2Donders Institute for Brain Cognition and Behaviour, Radboud University, Nijmegen, The Netherlands; Max Planck Institute for Psycholinguistics, Nijmegen, The Netherlands, 3Social, Life, and Engineering Sciences Imaging Center, the Pennsylvania State University, Pennsylvania, United States, 4Centre for Brain Research, Jagiellonian University, Kraków, Poland, 5McGovern Institute for Brain Research and Brain and Cognitive Sciences Department, Massachusetts Institute of Technology, Cambridge, United States
Introduction: In 1861, Broca made a strong empirical claim: that an area in the left inferior frontal gyrus (LIFG) selectively supports speech articulation. Broca asserted selectivity relative to both high-level language processing and general intelligence. Over the years, Broca’s claim received a lot of criticism. However, many critics a) have relied on analytic approaches that average brains in a common space, which often leads to a misleading picture because of inter-individual variability in functional topographies, and b) have relied on paradigms that conflate speech and language. fMRI studies that rely on individual-subject analyses (‘precision fMRI’) have shown that the left IFG contains language-selective and domain-general Multiple Demand (MD) areas (Fedorenko et al., 2012; Blank et al, 2014; Fedorenko & Blank, 2020). But these language-selective areas support high-level comprehension and production (Fedorenko et al., 2010; Meneti et al., 2011; Hu, Small et al., 2022), not speech motor control. So how does Broca’s claim of an articulation-selective area fit into the picture? Using precision fMRI, we first replicate the language/MD dissociation in the LIFG in a large dataset (Experiment 1) and then present evidence of articulation-selective areas distinct from both the language and MD areas (Experiment 2). Experiment 1: Participants (n=489) completed two ‘localizer’ tasks (extensively validated paradigms that work at the individual-subject level): (1) a language network localizer, based on the contrast of sentence vs. nonword reading (Fedorenko et al., 2010); and (2) a Multiple Demand network localizer, based on a harder vs. easier spatial working memory (WM) task (Fedorenko et al., 2013; Assem et al., 2020). Within an LIFG mask, we defined language and MD functional regions of interest (fROIs) in each individual using a portion of the data from each task and then examined these fROIs’s responses in independent data. We robustly replicated the language vs. MD dissociation: language fROIs responded strongly to sentence comprehension, but did not respond to the spatial WM task; MD fROIs responded strongly during spatial WM (more strongly during the harder condition) but showed little response during sentence comprehension. Furthermore, language and MD fROIs showed almost no overlap at the voxel level. Experiment 2: Participants (n=40) each completed the localizers from Experiment 1 and additionally, an articulation localizer based on the contrast of overt syllable repetition vs. finger tapping. We defined putative articulation fROIs in each individual using a portion of the data from the articulation localizer (articulation>finger-tapping). These areas i) responded robustly to articulation in independent runs, and ii) showed little or no response during language comprehension (during listening or reading) or during the spatial WM task (see also Basilakos et al., 2018). Language and articulation fROIs showed a small amount of overlap at the voxel level, but the majority of voxels responded selectively to language vs. articulation. Conclusions: Our results strongly support Broca’s original claim (also, Hillis et al., 2004) of the existence of an articulation-selective area in the LIFG, which does not play a role in high-level language processing or general intelligence.
Topic Areas: Language Production,