Functional specificity for high-level linguistic processing in the human brain

“Faces (Kanwisher et al., 1999), body parts (Downing et al., 2001), scenes (Epstein and Kanwisher, 1998), language (Fedorenko et al., 2011), and letters (Cohen and Dehaene, 2004) are further examples of stimuli that were claimed to be selectively processed by a specific brain region.”

“The activation areas for the third segment are localized in the left temporal and frontal lobes, potentially representing high-level stages of language processing, including sentence construction, semantic processing, and language expression 48 .”

“These regions neatly matched the language network [48,49] with the absence of the triangular part of the IFG and the 55b region in the premotor cortex [63] (see the overlap in S5 Fig), in line with previous studies suggesting these two dorsal regions play a non-semantic role in language processing…”

“, 2017; for a review, see Fedorenko, 2014) but not to nonlinguistic stimuli such as music, mathematical expressions, or computer code (Fedorenko et al., 2011; Ivanova et al., 2020; Monti et al., 2012).”

“…words, while two different cognitive functions may activate neighboring but distinct cortical regions within Broca’s area in every subject individually, group analyses might blur them into overlapping activations of a broadly defined Broca’s area, as shown by Fedorenko, Behr, and Kanwisher (2011).”

“In particular, the fronto-temporal language network should be more similar between the two hemispheres in its responses to language during early development compared to later childhood and adulthood.”

“Last, the functional data were smoothed spatially using spatial convolution with a 4 mm FWHM Gaussian kernel.”

“Therefore, it is also possible that the evaluation of pseudoword-shape associations might activate regions in the left hemisphere lexical-semantic network, e.g. as identified by Fedorenko et al. (2010, 2011) using language localizers contrasting sentences with strings of pseudowords.”

“The former regions, identified with a passive reading task, show a stronger response to sentences than lists of nonwords (Fedorenko et al., 2011); the latter regions, identified with a working-memory task that requires keeping track of several spatial locations, are more engaged during trials that are harder (8 locati…”

“Our key hypothesis is that reasoning and language processing can be explicitly disentangled , allowing reasoning abili-ties—once acquired in a high-resource language—to transfer more broadly across languages.”

“The networks are the dorsal and ventral attention networks, default mode network, salience network, and most importantly the frontoparietal cognitive control network ( Cole et al., 2014 ; Dixon et al., 2018 ; Fedorenko et al., 2011 ; Marek and Dosenbach, 2018 ; Yeo et al., 2011 ).”

“Occasionally, activity can also be seen in the rostrolateral prefrontal cortex (RPFC) (Duncan, 2010; Fedorenko et al., 2011).”

“Linguistic knowledge is most likely stored within the fronto-temporal language network (Fedorenko et al. 2011), and these are exactly the regions that became massively expanded in humans, along with some additional association zones in the parietal lobes (Buckner and Krienen 2013).”

“…activation related to language (both written and spoken) was localized successfully compared to the known literature on language related activation (Fedorenko et al., 2010, 2011, 2012), evoking activity in well-known language related structures, such as bilateral middle temporal gyri, inferior…”

“…mathematical calculation, spatial working memory, stroop effects and music processing) do not overlap those in linguistic tasks (Saxe et al., 2006; Fedorenko et al., 2011), whereas others illustrate that frontal or prefrontal brain regions become activated in both linguistic and nonlinguistic…”

“And second, many studies have relied on paradigms that conflate language processing and general task demands, which recruit distinct brain networks: the language-selective network (Fedorenko et al., 2011) and the domain-general Multiple Demand network (e.g., Duncan, 2010, 2013), respectively.”

“the patterns of letters and sounds that capture words) and semantic information about word meaning (Binder and Desai 2011; Fedorenko et al. 2011; Albouy et al. 2020; Braga et al. 2020; Hamilton et al. 2021).”

“It has been used for localizing the language network 24,26,29,41 , examining the linguistic functions of the brain networks defined by resting-state fMRI data processing 64 and revealing the recruitment of the occipital cortex of congenitally blind individuals in linguistic processing 65 .”

“Bach’s compositions, in particular, are associated with high-level linguistic processing in the brain and rhythmic processing, indicating the remarkable overlap between language and nonlinguistic tasks [21].”

“Neural studies have pinpointed specific areas involved in language and music processing (Fedorenko et al. 2011).”

“Neuroimaging studies have consistently shown that the language network – a set of brain regions responsible for language comprehension and production – remains largely inactive during various reasoning tasks (Amalric and Dehaene, 2019; Monti et al., 2012, 2007, 2009; Fedorenko et al., 2011).”

“These include the Pereira fMRI dataset, where participants read short passages [8]; the Fedorenko electrocorticography (ECoG) dataset, where participants read isolated sentences [19]; and the Blank fMRI dataset, where participants listened to short stories [20].”

“However, the human brain does not need to reformulate an entire functional network to learn a single new task [ Cole et al. , 2013; Fedorenko et al. , 2011 ] .”

“…like facial expressions or gestures [e.g., Amalric and Dehaene, 2019, Benn et al., 2021, Blank et al., 2014, Chen et al., 2021, Deen et al., 2015, Fedorenko et al., 2011, Ivanova et al., 2020, Jouravlev et al., 2019, Liu et al., 2020, Monti et al., 2007, 2009] [Monti et al., 2012, Paunov et…”

“This localizer identifies the fronto-temporal language network that selectively (Fedorenko et al., 2011) supports high-level language comprehension and production, including the processing of word meanings and combinatorial syntactic/semantic processing (Bautista & Wilson, 2016; Fedorenko et al.,…”

“Although temporal, frontal, and parietal regions encompass language-dedicated areas (Hickok and Poeppel, 2007; Fedorenko et al., 2011; Silbert et al., 2014), occipital and cerebellar regions are selectively involved when comprehension is challenging (Erb et al., 2013; Guediche et al., 2014).”

“Fedorenko et al. (2011) localized the brain regions involved in linguistic and nonlinguistic tasks, and found that regions activated by linguistic tasks (e.g., left inferior frontal gyrus, left temporal structures) showed little response to nonlinguistic functions.”

“In contrast, we take inspiration from functional specificity in neuroscience, which suggests that certain brain regions are specialized for particular tasks [20, 37].”

“Due to anatomical differences between individuals, the language system is defined via a functional localizer that contrasts syntactically and semantically valid sentences against a perceptually matched control, such as strings of nonwords (Fedorenko et al., 2011).”

“…is sensitive to integration across a broad array of tasks, it is not strongly engaged by basic language processes (e.g., sentence comprehension 23,31 ) Since L2 is typically less automatized than L1, the sensitivity of the MD network to task automatization, makes it a suitable candidate for…”

“Plus, an individualized data integration approach is herein developed, keeping in mind that an average brain may not be similar to an individual brain (Fedorenko et al., 2011; Fedorenko, Duncan, & Kanwisher, 2012).”

“Bearing such differences in mind, a particularly interesting result is the deactivation observed during the hard version of the verbal working memory task, which mirrors the patterns previously observed for the neocortical language regions (55).”

“It preferentially responds to language tasks when compared to non-linguistic tasks of similar complexity (Fedorenko et al., 2011).”

“…of left-hemisphere frontal and temporal regions are involved in processing language (Bates et al., 2001; Binder et al., 1997; Dronkers et al., 2004; Fedorenko et al., 2010, 2011; Friederici, 2011; Friederici & Gierhan, 2013; Price, 2010, 2012), robustly responding to language whether it is spoken…”

“…has shown that the human language network — a set of brain areas that are selectively and robustly activated when humans process language (e.g., Fedorenko et al., 2011; Fedorenko & Thompson-Schill, 2014; Lipkin et al., 2022; Regev et al., 2013) — exhibits the same amount of activation to…”

“Nevertheless, cognitive neuroscientists have documented rather distinct neural circuits for different components of language, for example speech perception and speech comprehension (contrast Norman-Haignere et al. 2015; Overath et al. 2015; Guenther 2016; with Fedorenko et al. 2010, 2011).”

“This system is typically left-lateralized ( Murphy et al., 2019), consistent with the frequent left-lateralization of language processing ( Enge et al., 2020; Fedorenko et al., 2011; Vigneau et al., 2006).”

“For example, language-responsive regions show little or no response to arithmetic (e.g., Amalric & Dehaene, 2019; Fedorenko et al., 2011; Monti et al., 2012), music (e.g., Chen et al., 2023; Fedorenko et al., 2011; Rogalsky et al., 2011), logic (e.g., Monti et al., 2009), executive control (e.g.,…”

“…each layer of a pretrained language model into four experts, each designed to mirror a major cognitive network in the human brain: language (Fedorenko et al., 2011), logic (multiple demand; Duncan, 2010), social reasoning (theory of mind; Saxe & Kanwisher, 2003), and world knowledge…”

“Taken together with previous evidence, our findings suggest that Broca ’ s area is implicated in multifunctional roles within high-level language (Fedorenko et al., 2011b) and domain-general related cognitions (Duncan, 2010), but does not participate in speech output.”

“…avenue for understanding the functional architecture of the cortical language network, where traditional group analyses have led to conflicting conclusions about the functional specialization or selectivity of classic language areas (Fedorenko et al., 2011; Nieto-Casta ˜ n ´ on & Fedorenko, 2012).”

“29 and 70) and their role in lexico - semantic and combinatorial processing during language comprehension and production (e.g., refs.”

“We used this conglomeration of language and social processing regions (Geschwind, 1972; Binder et al., 1997; Frith and Frith, 2007; Olson et al., 2007; Adolphs, 2009; Mitchell, 2009; Fedorenko et al., 2011) as the primary network-of-interest (NOI) for further analyses.”

“…et al., 2013; Scott et al., 2017) across tasks (e.g., Fedorenko et al., 2010; Cheung et al., 2020; Diachek et al., 2020), but show little or no response to diverse non-linguistic functions (e.g., Fedorenko et al., 2011; Monti et al., 2012; Fedorenko and Blank, 2020; Ivanova et al., 2020, 2021).”

“Furthermore, Fedorenko et al. (2011) were able to identify distinct language- versus calculation-related activation within the traditional language zones such as Broca’s area (BA 44/45).”

“The model fine-tuned on AC has higher performance in language-selective areas in the temporal and frontal lobes [Fedorenko et al., 2011, Lipkin et al., 2022].”

“It has recently been demonstrated that this language network exhibits selective activation for high-level linguistic processes in adults (Fedorenko et al., 2011) across 45 languages and 12 language families (Malik--Moraleda et al., 2022), thus representing a universal neural template for language…”

“The design of the task followed the one used by Fedorenko et al. (2011), in which participants completed blocks of the easy and hard versions of the task.”

“…we identify a network that includes the Precuneus and Parahippocampal Gyrus, which are regarded as part of the “extended” language network, compared to the “core” language network including brain regions such as the temporal lobe (Fedorenko et al., 2011; Ferstl et al., 2008; Xiong & Newman, 2021).”

“Together, these findings indicate that the language system is best considered a core-periphery model regulated by a domain-general and domain-specific subnetwork (8), an approach that has been proven to be constructive in understanding the functional architecture of language (9).”

“Second, the fine-resolution version also separates the language network (Fedorenko et al., 2011) and the DMN (Buckner et al., 2008).”

“The separability between language and cognitive control is suggested from the observation that most regions within the language system are not activated by tasks involving the multiple-demand system (Fedorenko et al., 2011).”

“Others have found little-to-no overlap (e.g. Acheson & Hagoort, 2014; Engelhardt et al., 2017; Fedorenko et al., 2011; Van Dyke et al., 2014) and argue for a strong division between the two (Fedorenko, 2014).”

“…studies have examined the relationship between language processing and these types of reasoning and found no overlap: e.g., mathematical reasoning (Fedorenko et al., 2011; Amalric & Dehaene, 2016; Amalric et al., 2019; for evidence from aphasia, see Varley et al., 2005), computer code…”

“The “sentences > non-words” contrast has been shown to reliably identify areas of the brain that are engaged in linguistic processing but not other functions (Fedorenko et al., 2011; Benn et al., 2023; Chen et al., 2023; see Fedorenko et al., 2024 for a review).”