Towards a causal role of Broca’s area in language: A TMS-EEG study on syntactic prediction

Abstract

Categorical predictions have been proposed as the key mechanism supporting the fast pace of syntactic composition in human language. Accordingly, grammar-based expectations facilitate the analysis of incoming syntactic information - e.g., hearing the determiner 'the' enhances the prediction of a noun - which is then checked against a single or few other word categories. Previous functional neuroimaging studies point towards Broca's area in the left inferior frontal gyrus (IFG) as one fundamental cortical region involved in categorical prediction during on-line language processing. Causal evidence for this hypothesis is however still missing. In this study, we combined Electroencephalography (EEG) and Transcranial Magnetic Stimulation (TMS) to test whether Broca's area is functionally relevant in predictive mechanisms for language. Specifically, we transiently perturbed Broca's area during the categorical prediction phase in two-word constructions, while simultaneously measuring the Event-Related Potential (ERP) correlates of syntactic composition. We reasoned that if Broca's area is involved in predictive mechanisms for syntax, disruptive TMS during the processing of the first word (determiner/pronoun) would mitigate the difference in ERP responses for predicted and unpredicted categories when composing basic phrases and sentences. Contrary to our hypothesis, perturbation of Broca's area at the predictive stage did not affect the ERP correlates of basic composition. The correlation strength between the electrical field induced by TMS and the magnitude of the EEG response on the scalp further confirmed this pattern. We discuss the present results in light of an alternative account of the role of Broca's area in syntactic composition, namely the bottom-up integration of words into constituents.

Syntax through the looking glass: A review on two-word linguistic processing across behavioral, neuroimaging and neurostimulation studies

“…The ESN was elicited both when participants performed a grammaticality judgement task (Experiment 1) and when their attention was diverted away from the stimuli (Experiment 2), supporting the automaticity analysis of syntactic analysis. The presence of the ESN was replicated in two experiments with agreement (Jakuszeit et al, 2013) and categorical violations (Maran et al, 2021). Of note, Jakuszeit et al (2013) failed to replicate the ESN effect for categorical violations, possibly due to reduced statistical power.…”

“…First, while the effect was first reported in the 100-300 ms time-window under attentive conditions (Hasting & Kotz, 2008), follow-up studies showed a prolonged negativity lasting approximately until 500 ms (Jakuszeit et al, 2013;Maran et al, 2021), possibly reflecting the overlap of an ESN and a subsequent N400. This second negativity, reminiscent of the effect reported by Carreiras (2003, 2005), might reflected additional processes related to processing a mismatching suffix, as it was observed in conditions in which syntactic violations were realised with overt marking (Jakuszeit et al, 2013;Maran et al, 2021)). Second, an increased late positivity starting approximately at 500 ms has been reported for agreement (Hasting & Kotz, 2008) and categorical (Jakuszeit et al, 2013;Maran et al, 2021) violations.…”

“…Techniques such as the crosssplicing further allowed to control for phonetic and acoustic features up to the time-locking point (Hasting & Kotz, 2008;Hasting et al, 2007;Shtyrov et al, 2003). An interesting aspect of ESN studies (Hasting & Kotz, 2008;Jakuszeit et al, 2013;Maran et al, 2021) is that their design, in combination with the cross-splicing technique, allows to investigate gram-matical effects which are orthogonal to the identity and the category of both words and to low-level of acoustic features up to the time-locking point (see Steinhauer and Drury, 2012 for a related discussion).…”

“…• Isolating top-down and bottom-up processes: The twoword level allows to isolate the relative contribution of top-down and bottom-up processes in language comprehension. In particular, it is possible to employ experimental condition which restrict prediction to the first word and integration to the second one, manipulating selectively one of the two key processes (Maran et al, 2021;Matar et al, 2021). This line of research could provide crucial insights into the neural basis of parsing (Abney & Johnson, 1991;Resnik, 1992), which at present is tested by comparing topdown, left-corner and bottom-up models of language comprehension in the context of narratives or longer stimuli (Bhattasali et al, 2019;Brennan & Pylkkänen, 2017;Hale et al, 2018;Nelson et al, 2017).…”

Cognitive performance and specific aspects of language processing are associated with Oldowan-like chert flaking and retouch

“…While neural and cognitive correlates of syntactic transitivity are not well understood, some studies which investigated the processing of more elaborate transitive coding frames (such as those involving dependent object clauses) have found increased metabolic activity and/or structural integrity of the left pars triangularis of Broca's area (Tyler et al 2011;cf. Casado et al 2020;Maran et al 2021;van Dam & Desai 2016). Some have also hypothesized that because the supposed roles of the pars opercularis include, among others, representing actions at the conceptual level and representing the sequential and hierarchical organization of action concepts [i.e., agent-action(-instrument, patient, etc.…”