Abstract
Comprehending and producing sentences is a complex endeavor requiring the coordinated activity of multiple brain regions. We examined three issues related to the brain networks underlying sentence comprehension and production in healthy individuals: First, which regions are recruited for sentence comprehension and sentence production? Second, are there differences for auditory sentence comprehension vs. visual sentence comprehension? Third, which regions are specifically recruited for the comprehension of syntactically complex sentences? Results from activation likelihood estimation (ALE) analyses (from 45 studies) implicated a sentence comprehension network occupying bilateral frontal and temporal lobe regions. Regions implicated in production (from 15 studies) overlapped with the set of regions associated with sentence comprehension in the left hemisphere, but did not include inferior frontal cortex, and did not extend to the right hemisphere. Modality differences between auditory and visual sentence comprehension were found principally in the temporal lobes. Results from the analysis of complex syntax (from 37 studies) showed engagement of left inferior frontal and posterior temporal regions, as well as the right insula. The involvement of the right hemisphere in the comprehension of these structures has potentially important implications for language treatment and recovery in individuals with agrammatic aphasia following left hemisphere brain damage.
Keywords: ALE, language, meta‐analysis, sentence comprehension, sentence processing networks, sentence production
1. INTRODUCTION
Comprehending or producing a sentence is a complex endeavor that requires several processing steps and the coordinated activity of multiple brain regions. Auditory and visual sentence comprehension involve conversion of acoustic or visual input signals into phonological or orthographic representations, respectively, which trigger access to stored word representations and associated grammatical and semantic information. For sentence comprehension, an incoming auditory or visual sequence is parsed into a syntactic phrase structure representation, which is used as the basis for computing thematic relations between verbs and arguments (thematic relations consist of functional roles such as agent, theme, etc., that define how a verb's arguments relate to its meaning) and ultimately a conceptual representation of the sentence. Sentence production is equally complex, requiring message generation followed by processes leading to access to and proper ordering of spoken or written word forms. However, the precise characterization of the component processes that are involved in sentence comprehension and production, their relationship to each other, and associated neural mechanisms vary across different neurocognitive models, and has been the subject of debate over the past several decades.
In the current article we used meta‐analysis to examine three issues related to the neural networks involved in sentence processing. First, we aimed to identify the networks for comprehension and production and examine points of difference and similarity in their respective networks. Second, we examined potential modality differences and overlap in the brain regions involved in comprehension of auditorily or visually presented sentences. Third, we examined the neural mechanisms underlying comprehension of syntactically complex structures (i.e., which encode “noncanonical” orders of thematic roles) compared with less complex structures (i.e., which encode standard “canonical” thematic role orders).
1.1. Sentence comprehension and production
One model that links sentence comprehension to underlying brain structures suggests that the comprehension process proceeds through three stages (Friederici, 2011). In the first stage, word category information informs the formation of an initial syntactic structure. The second stage uses lexical and structural information to parse thematic role assignments. In the third stage any errors or mis‐parses in the initial structure are repaired to arrive at the meaning of the sentence. Friederici (2011) proposes that these processes are subserved by a bilateral frontal–temporal network, with left inferior frontal and posterior temporal regions engaged for syntactic and semantic processing, and right hemisphere fronto‐temporal regions involved in prosodic processing.
Alternate models of sentence comprehension offer different sequences of processing and distinct functional roles for the brain regions underlying language. Bornkessel‐Schlesewsky and Schlesewsky (2013) propose that sentence comprehension depends on computation through parallel ventral and dorsal streams in the left hemisphere that begin in left posterior temporal cortex and converge in left inferior frontal cortex. The ventral stream passes through anterior temporal lobe regions and subserves time‐independent processing related to conceptual schema, while the dorsal stream runs through posterior superior temporal lobe and inferior parietal cortical regions and subserves time‐sensitive operations, including syntactic structuring. In contrast to other models, here left inferior frontal cortex is engaged for cognitive control rather than syntactic processing. Still other models emphasize contributions from memory systems to language comprehension (Caplan & Waters, 2013; Ullman, 2004).
Sentence production also proceeds through multiple stages, beginning with conceptualization and planning of the message to be produced. Planning at this stage involves multiple decisions, including how to linearize the message (i.e., deciding what part of the message to say first), and which lexical concepts should be used (e.g., referring to an entity in the sentence as an animal, dog, or labrador) (Indefrey & Levelt, 2000). The output of this stage is a preverbal message. During subsequent grammatical encoding the preverbal message is translated into a syntactically structured sequence of lemmas, which specifies, among other things, the thematic role and syntactic function of each element (e.g., a particular noun may be specified as the agent of the verb and mapped onto the grammatical subject position). Next, the structured sequence of lemmas is encoded to produce a phonological score, which is then translated into a phonetic representation and articulatory movements. See Indefrey and Levelt (2000), Menenti, Gierhan, Segaert, and Hagoort (2011), and Thompson and Kielar (2014) for review and alternative perspectives. Structuring the elements of a sentence for production is posited to depend on left perisylvian brain regions, principally left inferior frontal gyrus (IFG) and left posterior superior temporal gyrus (STG; Bornkessel‐Schlesewsky & Schlesewsky, 2013; Thompson & Kielar, 2014). On one view (Braun, Guillemin, Hosey, & Varga, 2001), bilateral posterior perisylvian regions (along with left fusiform gyrus) may be particularly involved in early conceptual and lexical stages of production. As production proceeds, later stages of phonological encoding and articulation depend on left frontal cortical regions, including the operculum, insula, lateral pre‐motor cortex and anterior supplementary motor area (SMA; Braun et al., 2001).
An important question for the neurocognitive basis of sentence processing is whether (and to what extent) the brain structures that subserve syntactic decoding for sentence comprehension are shared in common with the brain structures engaged in the encoding processes underlying sentence production. Multiple lines of evidence are consistent with at least partially overlapping systems across domains (Grodzinsky, 2000; Kempen, 2000; Menenti et al., 2011; Pickering & Garrod, 2013; Segaert, Menenti, Weber, Petersson, & Hagoort, 2012; Silbert, Honey, Simony, Poeppel, & Hasson, 2014). For example, neuroimaging studies suggest common areas of activation, particularly in left IFG, left middle temporal gyrus (MTG), and bilateral SMA (Indefrey, Hellwig, Herzog, Seitz, & Hagoort, 2004; Menenti et al., 2011; Segaert et al., 2012; Silbert et al., 2014). Studies involving individuals with aphasia and agrammatic language production also suggest at least some overlap, in that patients often have impaired production of simple as well as complex sentences, and they also evince impaired comprehension, particularly for noncanonical syntactic structures with thematically reversible arguments (i.e., where either participant in a verb's action could plausibly be linked to either role; e.g., The boy pushed the girl, where both the boy and the girl could plausibly be the pusher or the pushee) (Caramazza & Zurif, 1976; Grodzinsky, 2000; Love, Swinney, Walenski, & Zurif, 2008; Thompson & Choy, 2009).
Some studies, however, have found differences across domains. Indefrey et al. (2004) found left IFG activation in both comprehension and production but to different degrees, and Silbert et al. (2014) reported that subsets of activated areas in the temporal lobes responded selectively either to production or to comprehension. In addition, results of the latter study showed activation in left IFG (pars opercularis) activation for production, whereas right IFG (pars orbitalis) and bilateral inferior parietal lobule (IPL) were activated for comprehension but not production (Silbert et al., 2014).
1.2. Comprehension of auditory versus visual sentences
The second question we address in the current article concerns input modality differences in sentence comprehension. Multiple arguments and lines of evidence suggest important differences between the processes involved in auditory versus visual sentence comprehension (i.e., reading). In particular, listening to sentences puts different demands on the processing system than reading sentences does (Michael, Keller, Carpenter, & Just, 2001). To start, auditory presentation provides the listener with prosodic cues that the reader must either formulate for themselves (potentially with the aid of punctuation) or do without. Readers are also able to control the rate at which they process information, at least when sentences are presented all at once, while listeners are at the mercy of the speech rate of the speaker (Ferreira & Anes, 1994). In general, fluent auditory presentation engages comprehension processes that are fast, automatic, and not consciously accessible (Swinney, Love, Nicol, Bouck, & Hald, 2000). In contrast, reading engenders greater controlled processing, both with respect to planning and execution of eye‐movements across text and with respect to managing attentional and working‐memory demands on processing (Walczyk, 2000). Readers also have time to engage in explicit, conscious problem‐solving strategies to fit words together into a coherent meaning (Swinney & Osterhout, 1990), and can backtrack to previously read material that may be difficult or ambiguous (Rayner, 1998). Even methods that force people to read sentences one word at a time, such as self‐paced reading and rapid serial visual presentation (where words are presented for a fixed duration in rapid succession) appear to encourage an integrative mode of processing that differs from that subserving uninterrupted auditory sentence comprehension (Love et al., 2008; Nicol, Swinney, Love, & Hald, 2006; Swinney et al., 2000; Swinney & Osterhout, 1990).
Notably, despite these processing differences, event‐related potential studies consistently find that ERP components are largely similar across modalities. N400 effects to semantic violations are found for the same stimuli with auditory and visual presentation, though small differences in the time course or scalp distribution of the effects across modalities have been observed (Balconi & Pozzoli, 2005; Kutas & Federmeier, 2011; Kutas & Van Petten, 1994; Lück, Hahne, & Clahsen, 2006). Likewise, P600 and left anterior negativity (LAN) effects, which are typically evoked in response to (morpho‐) syntactic violations, are also found in both modalities, with only minor modality differences in timing and scalp distribution for both kinds of effects (Balconi & Pozzoli, 2005; Hagoort & Brown, 2000; Lück et al., 2006; Regel, Gunter, & Friederici, 2011). The similarity of these effects suggests similar processing across modalities.
Some fMRI studies directly comparing the same stimuli across modalities have found brain regions that respond to both. These studies consistently report that portions of left IFG and posterior superior/middle temporal gyri (STG/MTG) respond both to auditory and to visual sentences (Buchweitz, Mason, Tomich, & Just, 2009). However, differences between modalities have been found. Temporal‐lobe clusters of activation for auditory input are larger and have a more anterior distribution than active clusters for visual input (Buchweitz et al., 2009; Constable et al., 2004; Michael et al., 2001), and clusters of activation in left frontal cortex for auditory input are anterior and inferior to those in response to visual stimuli (Michael et al., 2001). These findings indicate that networks for listening and reading are at least partially overlapping, though engagement of unique regions in each modality are required to accommodate the processing demands that these language domains each engender.
1.3. Comprehension of complex syntax
The third issue that we address concerns the regions involved in the comprehension of complex syntactic structures. To investigate the issue of syntactic complexity, we targeted the specific contrast between noncanonical and canonical sentence structures. Noncanonical structures are more difficult to process than canonical structures, and comprehension of such sentences is frequently impaired in agrammatic aphasia (Caramazza & Zurif, 1976; Grodzinsky, 2000; Love et al., 2008; Thompson & Choy, 2009). Noncanonical structures are therefore particularly important for investigations of agrammatic language comprehension and, more broadly, for the neural basis of language.
In canonical structures the order of arguments follows the agent–verb–theme order that is the dominant order in English. Thus, in active sentences like The girl watched the boy, the agent (the girl) precedes the theme (the boy). Here agent and theme refer to the thematic roles assigned by the verb to its arguments, which occupy the subject and object positions within the sentence. Thematic roles specify the semantic relation between the argument and the verb—that is, the argument assigned the agent role is interpreted as the doer of the verb's action, whereas the argument assigned the theme role is interpreted as the recipient of the verb's action. In English, any structure in which the theme precedes the agent is considered noncanonical. Such sentences subvert the dominant order of arguments in a language and are, thus, syntactically more complex than sentences which do not. For example, passivized sentences (e.g., The boy was watched by the girl) use morphological markers (‐ed; by) to assign the theme role to the argument in subject position (the boy, which is the recipient of the action of watched) and the agent to the noun phrase in the prepositional phrase (by the girl), contra the dominant, simpler agent–verb–theme order in active English sentences (e.g., The girl watched the boy). Note that other languages have different canonical thematic ordering, and hence different structures that are classified as noncanonical. Note as well that canonical and noncanonical orders are also frequently specified in terms of subjects and objects instead of agents and themes, particularly for languages with flexible word order. See “Method” section below for justification of the structures that were classified as noncanonical.
The left IFG has been claimed to be required for processes that are critical for comprehension of complex syntax (Grodzinsky, 2000), though studies contrasting more‐complex versus less‐complex structures show recruitment not only of left IFG for complex syntactic structures, but also left posterior temporal lobe regions (Bornkessel, Zysset, Friederici, Yves von Cramon, & Schlesewsky, 2005; Mack, Ji, & Thompson, 2013; Meltzer‐Asscher, Schuchard, Den Ouden, & Thompson, 2012; Thompson et al., 2007; Thompson, Bonakdarpour, & Fix, 2010). Studies also have implicated homologous regions within the right hemisphere in complex sentence comprehension, particularly right IFG (Mack, Meltzer‐Asscher, Barbieri, & Thompson, 2013; Meltzer, McArdle, Schafer, & Braun, 2010). In the present investigation we evaluate the results of neuroimaging studies that explicitly compared noncanonical versus canonical structures to better understand whether or not the left IFG supports syntactically complex sentence computation or whether this process engages additional left (and possibly) right hemisphere regions.
1.4. Previous meta‐analyses of sentence‐level processing
A few previous meta‐analyses have addressed these issues. Vigneau et al. (2006) examined left hemisphere fMRI activation associated with sentence processing. Their analysis revealed clusters of activation peaks in left middle frontal gyrus (MFG), IFG (pars opercularis and pars triangularis), posterior superior temporal sulcus (STS), posterior MTG, anterior inferior temporal gyrus (ITG), and temporal pole. However, one issue with this analysis is that it did not include all left hemisphere activation peaks, but only those from a select set of regions of interest (Vigneau et al., 2006). The analysis also included studies using a variety of sentence level tasks, including auditory and visual sentence comprehension tasks and production tasks such as sentence completion; activation differences associated with different processing domains (comprehension, production) or modalities (auditory, visual) were not examined. A follow‐up article examined right‐hemisphere contributions to sentence processing from the same studies included in the left‐hemisphere meta‐analysis and found concentrations of right hemisphere activation peaks in inferior frontal and superior temporal gyri (Vigneau et al., 2011).
Two subsequent meta‐analyses examined comprehension and production in their own right using activation likelihood estimation in two separate studies. Adank (2012a) examined the brain regions involved in the auditory comprehension of language. The results indicated a network of regions in the left hemisphere in the temporal lobe (anterior STS and STG, MTG, fusiform gyrus), inferior parietal lobe (angular gyrus), and frontal lobe (pars opercularis and orbitalis of IFG, ventral premotor cortex, SMA, and precentral gyrus), as well as in the right hemisphere in the insula, anterior STS and STG, and MTG. Adank (2012b) examined the brain regions involved in language production, and found cortical activation clusters in left supplementary motor area, left precentral gyrus, left anterior STG and Heschl's gyrus, right posterior STG, and left IFG and insula. However, neither meta‐analysis was restricted to sentence‐level material. The meta‐analysis of comprehension included words, sentences, and narratives. The meta‐analysis for production included syllables, words, and sentences, but included only two studies with sentence‐level material, and both involved reading aloud, which reduces the need to formulate a message and select lexical items, and may also involve a great deal of processing related to sentence comprehension. Thus, it is not clear to what extent the regions implicated in the production analysis reflected independent contributions from processes specifically required for sentence production.
Two prior activation likelihood estimation meta‐analyses specifically examined the issue of syntactic complexity in language comprehension. Meyer and Friederici (2015) meta‐analyzed neuroimaging activation patterns for complex versus simple syntactic structures, but restricted their analysis to the left hemisphere. Their complex structures included monoclausal sentences with noncanonical thematic orders and (noncanonical) sentences with embedded relative clauses. The authors report two clusters activated for comprehension of these structures: one in frontal (left IFG, the insula, and MFG), and one in posterior (left MTG and STG) regions. Rodd, Vitello, Woollams, and Adank (2015) examined syntactic complexity through the proxy of processing load, and identified contrasts of higher‐load versus lower‐load conditions with either auditory or visual stimulus presentation. Their analysis revealed several clusters, all in the left hemisphere, including both anterior (e.g., pars opercularis and pars triangularis of left IFG) and posterior perisylvian regions (e.g., ITG, MTG, STG, and supramarginal gyrus [SMG]). Notably, however, this analysis combined several dimensions of sentence processing complexity, and included results derived from anomalous versus nonanomalous sentences, ambiguous versus unambiguous sentences, noncanonical versus canonical sentences, and ungrammatical versus grammatical sentence fragments, among others. Thus, it is unclear to what extent canonicity versus other syntactic and/or semantic properties of sentences led to the results obtained. Collectively, these findings point to the need for additional, but more constrained, analyses of the extant literature.
1.5. The current study
Here we use ALE meta‐analysis to investigate three issues. First, we sought to identify the brain structures participating in sentence comprehension and sentence production. We also examined the extent to which the identified regions are overlapping across these domains. Second, we examined whether or to what extent the brain regions for comprehending auditory sentences are the same as those for sentence comprehension in reading. Finally, we examined the set of regions involved in the comprehension of complex structures, where complexity is defined narrowly in terms of deviations from the canonical ordering of thematic roles. For all three investigations, we specifically restrict the domain of inquiry to studies reporting data from whole‐brain analyses of tasks requiring comprehension or production of one or more sentences; thus unlike previous meta‐analyses we do not include studies of smaller units such as sentence fragments, words, or syllables in nonsentential contexts. While sentence processing necessarily includes processing of these smaller units, the processing of these units in isolation omits the sentence‐level processes related to structure and meaning that lie at the heart of our investigation (for a related point, see Adank, 2012a).
2. METHOD
2.1. Literature searches
We conducted a systematic, comprehensive, and objective literature search to find studies to include in the data set for the meta‐analysis. To identify studies of sentence comprehension and production we performed initial http://pubmed.gov searches using the search terms “sentence comprehension” and “syntax comprehension” and “sentence production,” “syntax production,” and “narrative production,” respectively, each in conjunction with “FMRI” or “PET.” In order to find additional original studies of sentence comprehension and production that our initial searches did not identify, we also examined the references cited in the studies that we ultimately included, as well as those from recent meta‐analyses (Adank, 2012a, 2012b; Alain, Du, Bernstein, Barten, & Banai, 2018; DeWitt & Rauschecker, 2012; Martin, Schurz, Kronbichler, & Richlan, 2015; Meyer & Friederici, 2015; Rodd et al., 2015; Vigneau et al., 2006). The searches (from 1992 to present) were completed in May, 2018. Excluding duplicates and studies not written in English, these searches yielded a total of 531 studies for sentence comprehension and 158 for sentence production.
For sentence comprehension, we excluded studies if they did not test sentence comprehension (n = 14), if they were a review or meta‐analysis (n = 22), if they used a method other than FMRI or PET (n = 45), if they did not report coordinates of activation peaks from whole brain analyses in either Talairach or Montreal Neurological Institute (MNI) space (n = 65), or if they did not include a task that required the participant to make an overt response at any point (n = 17) (for discussion of the issue of overt vs. covert responses, see Adank, 2012a; Caplan, 2010). In addition, we excluded studies that did not report results from healthy individuals (n = 54), that reported only results from children (n = 33), or that reported results only from speakers with non‐native language proficiency (n = 5). We selected contrasts from the remaining 277 studies as follows.
First, to address the question of identifying the language network for sentence comprehension, we selected contrasts of normal language against a low‐level baseline intended to subtract out task‐related activation (e.g., activation due to perceiving the stimuli and making a response). Our goal was to include activation from processing at any stage of the comprehension process beyond perceiving the stimulus itself. We therefore did not include contrasts that subtracted activation from potentially important contributing processes (e.g., sentences against nonsense sentences or lists of single words, ambiguous against unambiguous sentences, sentences with metaphoric vs. literal meanings, etc.). We also did not include contrasts with sentences that included pseudowords, as these may engender different or additional processing beyond what is typically required for comprehension as well as interactions between any such processes and the types of materials in the contrasts (see Adank, 2012a).
For studies with auditory stimulus presentation, we identified all contrasts of sentences against an unintelligible auditory baseline (as in Adank, 2012a): white noise, reversed speech, speech‐envelope noise, tones, signal correlated noise, hummed sentences, low‐pass filtered speech (preserving only an intonational contour, thus unintelligible), and noise vocoded speech. Note that noise‐vocoded speech can be intelligible if coded with more than four channels (Davis, Johnsrude, Hervais‐Adelman, Taylor, & McGettigan, 2005), or when preceded by intelligible versions of the same sentence (Hakonen et al., 2017). We therefore, only included contrasts with noise‐vocoded speech as the baseline where it was encoded with four or fewer channels, where it was not preceded by a more intelligible version of the same sentence, and where it was reported to be unintelligible. Thus, for example, Erb, Henry, Eisner, and Obleser (2013) was not included, as the baseline was reported to be 51% intelligible. We also excluded contrasts that reported parametric results including partially intelligible stimuli (e.g., Scott, Rosen, Lang, & Wise, 2006). For studies with visual sentence presentation, we identified all contrasts of normal sentence reading against a baseline condition utilizing stimuli composed of unstructured strings of symbols (i.e., meaningless and not phonologically structured): pseudofonts, lines, symbols, rows of hashmarks or capital Xs, random sequences of consonants or letters, or degraded illegible letters. Coordinates from interaction effects (if reported) were not included. This left us with a total of 45 contrasts (31 auditory, 14 visual) from healthy adult participants that were included in the meta‐analysis.
To address the question of brain regions involved in the comprehension of noncanonical syntactic structures, we examined the set of 277 comprehension studies from our searches to identify any that reported contrasts for sentences with noncanonically ordered argument structures against sentences with canonically ordered argument structures. This included comparisons of sentences with object relative clauses either against subject relative clauses, against active sentences with conjoined objects, or against embedded yes/no questions. We also included contrasts of object‐cleft sentences against subject‐cleft sentences, object wh‐questions against subject‐wh questions, and passive against active sentences. We also included studies conducted in languages with flexible word order. For example, in Kaqchikel the canonical word order is Verb–Object–Subject, though noncanonical Subject–Verb–Object order is allowed (Koizumi & Kim, 2016). German and Japanese have canonical subject‐before‐object word order and noncanonical object‐before‐subject word order (Kim et al., 2009; Kinno, Kawamura, Shioda, & Sakai, 2008; Meyer, Obleser, Anwander, & Friederici, 2012). Contrasts from parametric analyses or activation peaks from interactions again were not included. In total, we identified 37 such contrasts to include in the meta‐analysis.
Finally, for sentence production, we excluded studies that did not test production of sentences (e.g., syllable production; n = 10), if they were a review or meta‐analysis (n = 15), if they used a method other than FMRI or PET (n = 46), if they did not report coordinates of activation peaks from whole brain analyses in either Talairach or Montreal Neurological Institute (MNI) space (n = 25), or if they did not include a task requiring an overt response (n = 8). In addition, we excluded studies that did not report results from healthy individuals (n = 11) or humans (n = 1), that reported only results from children (n = 4), or that reported results only from non‐native speakers (n = 1). We also excluded studies where the production task involved sentence comprehension. We thus excluded production tasks that required rote recitation of trained or memorized speech (whether a story or sentence; n = 5), repetition or shadowing of another speaker (n = 2), reading sentences aloud (n = 5), or producing only a single word to complete a sentence (n = 2). We excluded a further eight studies that reported only main effects across production and comprehension or that reported only results of group differences between healthy and atypical speakers.
The remaining 15 studies used tasks in which participants were prompted to overtly generate one or more sentences. From these 15 studies, we selected all contrasts aimed at removing activation related to peripheral perceptual (i.e., visual decoding) and/or motor (i.e., articulating a response) processes, but that did not remove activation related to any core stages of production of a novel sentence (e.g., message generation, lexical selection, syntactic encoding, etc.). Due to the overall small number of the remaining production studies, we were not able to include a uniform set of contrasts. Note that contrasts against rest were included only if no contrasts against a low‐level baseline condition were reported—that is, if a study reported both a contrast against rest and a contrast against another task or condition, we included only the latter). We did not include contrasts of normal speech against error‐filled or dysfluent speech (Grande et al., 2012).
2.2. ALE method
Activation likelihood estimation (ALE) is a method of meta‐analysis that models the spatial coordinates of reported activation foci as three‐dimensional (3D) Gaussian probability distributions centered at each coordinate (Laird et al., 2005; Turkeltaub, Eden, Jones, & Zeffiro, 2002). We used GingerALE 2.3.6 (http://www.brainmap.org) to conduct the ALE meta‐analyses reported in this article. To perform each analysis, 3D (X, Y, Z) coordinates in MNI space from the included contrasts were combined to create a data set. Coordinates originally reported in Talairach space were converted to MNI space using the Lancaster (SPM) algorithm provided in the GingerAle software package (Lancaster et al., 2007). We also include the originally reported coordinate space in the table entry for each study (in some cases we confirmed the space with the study authors; these are identified in the tables). Coordinates from multiple contrasts from the same participants in a study were not treated as independent observations, but were grouped together in the data set (Turkeltaub et al., 2012). Note that we did treat different subject groups on the same stimuli as independent observations.
In the analyses, full‐width half maximum (FWHM) was subject‐based (Eickhoff et al., 2009), and the modeled activation maps were computed using nonadditive random effects (Turkeltaub et al., 2012). We used cluster level inference thresholding, with false discovery rate correction for multiple comparisons (Eickhoff, Bzdok, Laird, Kurth, & Fox, 2012). The clusters were formed based on the uncorrected p‐values at p < 0.001, with cluster thresholding value of p = 0.05 and 1,000 permutations. For each cluster from each analysis, we report all extrema (i.e., peak ALE probability values within each cluster) with region labels, 3D MNI coordinates, and extrema value. We used the Harvard–Oxford atlas in MRIcron (Rorden, Karnath, & Bonilha, 2007) to assign anatomical region labels to the extrema coordinates. In addition, we report the total cluster volume and a list of studies with one or more foci inside the cluster. Clusters smaller than 250 mm3 or that had fewer than three contributing studies were not reported. The number of contributing studies was based on the GingerALE output, which identifies those studies with at least one focus inside the volume of the cluster. Note that it is possible that an activation peak outside the cluster will nonetheless contribute some probability to it, and so the GingerALE output is a conservative measure of the number of contributing studies.
To examine overlap across ALE‐activation maps between comprehension and production, and between auditory versus visual comprehension, we computed conjunction maps for each of these comparisons using GingerALE. GingerALE computes a conjunction map by comparing two cluster‐thresholded maps and taking the smaller of the ALE values at each voxel. Thus, clusters in the conjunction map only contain voxels that were above threshold in both of the cluster‐thresholded maps, and hence indicate overlap between significant clusters in the two images being compared.
3. RESULTS
3.1. Sentence processing (comprehension and production)
We first analyzed the studies of sentence comprehension and production together, including all of the studies listed in Tables 1, 2, 3. This included 784 foci from 996 participants from 60 studies. All studies used fMRI except five, which used PET. While we placed no particular restrictions on the language in which a study was conducted, most of the included studies were conducted with speakers of a Germanic language (English, Dutch, German, Swedish), with one study representing the Romance languages (Italian), one conducted in Korean, one in Finnish, and one in American Sign Language (ASL) in a group of hearing bilinguals. Of the 60 studies, 5 explicitly stated that their participants were monolingual native speakers of the language the study was conducted in, 38 specified only that participants were native speakers, two specified bilingual native speakers (of both languages), and 15 did not specify the language background of their participants. The majority of individuals were stated to be right handed (n = 967), though the handedness of 24 individuals was not specified (Noppeney & Price, 2004; Wilson, Molnar‐Szakacs, & Iacoboni, 2008), and a total of 5 participants were left handed (all in comprehension studies). The stimuli across the comprehension studies were variable, including simple and complex sentences, either in isolation or sequentially in narratives. The modality (auditory, visual) employed across the comprehension studies also varied, as did the tasks, though the majority made use of comprehension questions or a binary judgment about some aspect of the sentence. All production studies employed an overt sentence generation task (of one or more sentences), though prompts varied, and included pictures, words, and picture/word combinations, in either visual or auditory modalities (for word‐based prompts). The number of participants in each study ranged from a low of 5 to a high of 44. Tables 1, 2, 3 also specify the coordinate space the data were originally reported in, and the statistical threshold for each study's data. For the statistical thresholds, we give the p‐value and the method of correction for multiple comparisons (or whether it was uncorrected). We include random field theory and related methods (e.g., white‐noise based approaches) under the umbrella term of familywise error rate (FWE) based corrections, and distinguish these from Monte Carlo and False Discovery Rate (FDR) based approaches (Nichols, 2012; Nichols & Hayasaka, 2003). Note that authors did not always unambiguously specify the method by which a correction was applied; see individual articles for specific details.
Table 1.
Sentence comprehension studies (auditory)
| ID | Study | Method | Language | Task | Contrast description | N a | # of foci | Reported coordinate space | Statistical thresholdc |
|---|---|---|---|---|---|---|---|---|---|
| 1 | Wong, Miyamoto, Pisoni, Sehgal, and Hutchins (1999) | PET | English | Button press to alternate stimuli | Short simple sentences > reversed speech | 5 | 1 | Talairach | p ≤ 0.05f, 1‐tailed |
| 2 | Kansaku, Yamaura, and Kitazawa (2000) | FMRI | English | Offline comprehension test | Narrative stories > reversed speech (female participants) | 25 | 2 | Talairachb | p = 0.001d |
| 3 | Kansaku et al. (2000) | FMRI | English | Offline comprehension test | Narrative stories > reversed speech (male participants) | 22 | 2 | Talairachb | p = 0.001d |
| 4 | Meyer, Alter, Friederici, Lohmann, and von Cramon (2002) | FMRI | German | Active vs passive sentence decision | Active and passive sentences (normal) > low‐pass filtered speech (prosodic) | 14 | 6 | Talairach | p < 0.001d, 1‐tailed |
| 5 | Crinion, Lambon‐Ralph, Warburton, Howard, and Wise (2003) | PET | English | Offline comprehension test | Stories with simple sentences > reversed speech | 17 | 11 | MNI | p < 0.05f |
| 6 | Tracy et al. (2003) | FMRI | English | Offline comprehension test | Stories > scanner noise (Table 2a) | 15 | 2 | MNI | p < 0.05f , j |
| 7 | Von Kriegstein, Eger, Kleinschmidt, and Giraud (2003) | FMRI | German | Voice, word, or prosody recognition | Normal sentences > speech envelope noise | 14 | 9 | MNIb | p < 0.001d |
| 8 | Constable et al. (2004) | FMRI | English | Acceptability judgment | Subject and object relative sentences > tones | 20 | 8 | Talairach | p < 0.01f , j |
| 9 | Meyer, Steinhauer, Alter, Friederici, and von Cramon (2004) | FMRI | German | Prosody comparison | Normal speech > low‐pass filtered speech (prosodic) | 14 | 6 | Talairach | p = 0.001d |
| 10 | Wildgruber et al. (2004) | FMRI | German | Prosody judgment |
Sentences > scanner noise: Affective judgment Sentences > scanner noise: Linguistic judgment |
10 |
13 12 |
MNI MNI |
|
| 11 | Crinion and Price (2005) | FMRI | English | Offline recognition test | Stories > reversed speech | 18 | 19 | Talairach | p < 0.05f , j |
| 12 | Rodd et al. (2005) | FMRI | English | Probe relatedness judgment | Sentences with lexical ambiguity > signal correlated noise | 30 | 22 | MNI | p < 0.05f , j |
| 13 | Kang et al. (2006) | PET | Korean | Plausibility judgment | Short simple sentences > white noise | 17 | 4 | Talairach | p < 0.05f , j |
| 14 | Love et al. (2006) | FMRI | English | Probe reading, verification, and judgment | Sentences > scanner noise | 10 | 19 | Talairach | p < 0.001d |
| 15 | Davis et al. (2007) | FMRI | English | Offline sentence recognition test | Sentences with lexical ambiguity > signal correlated noise | 12 | 14 | MNI | p < 0.05g , i |
| 16 | Ischebeck, Friederici, and Alter (2008) | FMRI | German | Probe recognition task | Sentences with prosodic breaks > hummed sentences | 14 | 8 | MNIb | p < 0.05e , j |
| 17 | Kovelman, Baker, and Petitto (2008) | FMRI | English | Plausibility judgment | Monolingual English: Sentences > scanner noise | 10 | 17 | MNI | p < 0.001d |
| 18 | Kovelman et al. (2008) | FMRI | English | Plausibility judgment |
Bilingual English: Sentences > scanner noise Bilingual Spanish: Sentences > scanner noise |
11 |
16 16 |
MNI MNI |
p < 0.001d p < 0.001d |
| 19 | Wilson et al. (2008) | FMRI | English | Offline comprehension test | Stories > scanner noise | 12 | 7 | MNI | p < 0.05f , j |
| 20 | Lillywhite et al. (2010) | FMRI | English | Post‐session recall test | First presentation of newspaper story > white noise | 20 | 2 | MNI | p < 0.001d |
| 21 | Okada et al. (2010) | FMRI | English | Intelligibility decision (yes/no) | Intelligible sentences > unintelligible sentences | 20 | 14 | MNI | p < 0.05g , j |
| 22 | Peelle, Troiani, Wingfield, and Grossman (2010) | FMRI | English | Probe relatedness judgment | Simple sentences > signal correlated noise | 6 | 19 | MNI | p < 0.05f , k |
| 23 | Rodd, Longe, Randall, and Tyler (2010) | FMRI | English | Probe relatedness judgment | Sentences with lexical and syntactic ambiguity > signal correlated noise | 14 | 13 | MNI | p < 0.05e , j |
| 24 | Tyler et al. (2010) | FMRI | English | Word‐monitoring task | Normal prose > speech‐envelope noise (young group) | 14 | 4 | MNI | p < 0.05g , j |
| 25 | Tyler et al. (2010) | FMRI | English | Word‐monitoring task | Normal prose > speech‐envelope noise (older group) | 44 | 3 | MNI | p < 0.05g , j |
| 26 | Rodd et al. (2012) | FMRI | English | Probe relatedness judgment | Sentences with lexical ambiguity > signal correlated noise | 15 | 16 | MNI | p < 0.05g , j |
| 27 | Van Leeuwen et al. (2014) | FMRI | Dutch | Sentence‐picture matching | Normal prose > reversed speech | 16 | 13 | MNI | p < 0.05f , j |
| 28 | Vitello, Warren, Devlin, and Rodd (2014) | FMRI | English | Probe relatedness judgment | Sentences with disambiguating information > signal correlated noise | 20 | 7 | MNI | p < 0.05f , i |
| 29 | Tuennerhoff and Noppeney (2016) | FMRI | German | Sentence monitoring (target sentence detection) | Unprimed normal sentences > unprimed degraded sentences | 20 | 6 | MNI | p < 0.05f , j |
| 30 | Willems, Frank, Nijhof, Hagoort, and van den Bosch (2016) | FMRI | Dutch | Post‐task comprehension test | Stories > reversed speech: Entropy measure | 20 | 9 | MNI | p < 0.05h , j |
| Stories > reversed speech: Surprisal measure | 12 | MNI | p < 0.05h , j | ||||||
| 31 | Hakonen et al. (2017) | FMRI | Finnish | Intelligibility judgment | Intact sentences > first presentation of distorted sentences | 20 | 7 | MNI | p < 0.05f , j |
N = Number of participants.
Confirmed with study authors (personal communication).
The method of correction for multiple comparisons (FWE, FDR, Monte‐Carlo, or whether there was no correction) is specified in table notes. FWE, familywise error rate; FDR, false discovery rate. Note that whether a correction was applied voxel‐wise or cluster‐wise was sometimes specified even if the method was not, but was itself not always specified.
Uncorrected for multiple comparisons.
Corrected for multiple comparisons, method not specified.
FWE.
FDR.
Monte‐Carlo based correction.
Voxel‐wise (whole brain).
Voxel‐wise (small volume correction).
Cluster‐wise (whole brain).
Set‐level (whole brain).
Table 2.
Sentence comprehension studies (visual)
| ID | Study | Method | Language | Task | Contrast description | Na | # of foci | Reported coordinate space | Statistical thresholdc |
|---|---|---|---|---|---|---|---|---|---|
| 32 | Robertson et al. (2000) | FMRI | English | Offline old/new probe recognition | Indefinite article sentences > nonletter character strings | 8 | 8 | MNIb | p < 0.05f , i |
| Definite article sentences > nonletter character strings | 8 | MNIb | p < 0.05f , i | ||||||
| 33 | Cooke et al. (2001) | FMRI | English | Gender judgment of agent | Short subject relative > Pseudofont | 7 | 2 | MNI | p < 0.05f , j |
| Long subject relative > Pseudofont | 3 | MNI | p < 0.05f , j | ||||||
| Short object relative > Pseudofont | 1 | MNI | p < 0.05f , j | ||||||
| Long object relative > Pseudofont | 6 | MNI | p < 0.05f , j | ||||||
| 34 | Ferstl and von Cramon (2001) | FMRI | German | Sentence pair coherence judgment | Language > nonword (consonant) sentences | 12 | 10 | Talairach | p < 0.001d |
| 35 | Grossman et al. (2002) | FMRI | English | Gender judgment of agent | Short subject relative > Pseudofont (younger adults) | 13 | 4 | MNI | p < 0.05f, j; variousd |
| Long subject relative > Pseudofont (younger adults) | 5 | MNI | p < 0.05f, j; variousd | ||||||
| Short object relative > Pseudofont (younger adults) | 4 | MNI | p < 0.05f, j; variousd | ||||||
| Long object relative > Pseudofont (younger adults) | 6 | MNI | p < 0.05f, j; variousd | ||||||
| 36 | Grossman et al. (2002) | FMRI | English | Gender judgment of agent | Short subject relative > Pseudofont (older adults) | 11 | 5 | MNI | p < 0.05f, j; variousd |
| Long subject relative > Pseudofont (older adults) | 7 | MNI | p < 0.05f, j; variousd | ||||||
| Short object relative > Pseudofont (older adults) | 6 | MNI | p < 0.05f, j; variousd | ||||||
| Long object relative > Pseudofont (older adults) | 7 | MNI | p < 0.05f, j; variousd | ||||||
| 37 | Noppeney and Price (2004) | FMRI | English | Offline recognition test | Sentences with clause ambiguities > false font | 12 | 7 | Talairach | p < 0.05e |
| 38 | Constable et al. (2004) | FMRI | English | Acceptability judgment | Subject and object relatives > lines | 20 | 10 | Talairach | p < 0.01f, j |
| 39 | Xu, Kemeny, Park, Frattali, and Braun (2005) | FMRI | English | Offline comprehension test | Sentences > random consonant letter strings | 22 | 17 | MNI | p < 0.001d |
| Narratives > random consonant letter strings | 34 | MNI | p < 0.001d | ||||||
| 40 | Bahlmann, Rodriguez‐Fornells, Rotte, and Münte (2007) | FMRI | German | Offline sentence recognition test | Canonical and noncanonical sentences with lexical ambiguity > row of X's | 12 | 5 | MNI | p < 0.001d |
| 41 | Boulenger, Hauk, and Pulvermüller (2009) | FMRI | English | Yes/no comprehension questions | All sentences > Hashmarks (early analysis window) | 18 | 11 | MNI | p < 0.05g , j |
| All sentences > Hashmarks (late analysis window) | 10 | MNI | p < 0.05g , j | ||||||
| 42 | Marques, Canessa, and Cappa (2009) | FMRI | Italian | True/false judgment | Active sentences > row of X's | 21 | 33 | MNI | p < 0.05f , j |
| 43 | Diaz, Barrett, and Hogstrom (2011) | FMRI | English | Emotional valence judgment | Novel metaphors > random letter strings | 16 | 6 | MNI | p < 0.01d |
| Familiar metaphors > random letter strings | 4 | MNI | p < 0.01d | ||||||
| Novel literal sentences > random letter strings | 3 | MNI | p < 0.01d | ||||||
| Familiar literal sentences > random letter strings | 4 | MNI | p < 0.01d | ||||||
| 44 | Schuil, Smits, and Zwaan (2013) | FMRI | Dutch | Probe relatedness judgment | Sentences > illegible sentences | 20 | 7 | MNI | p < 0.05f , j |
| 45 | Van Ettinger‐Veenstra, McAllister, Lundberg, Karlsson, and Engström (2016) | FMRI | Swedish | Indoor/outdoor action judgment | Semantically congruent sentences > symbol font | 27 | 7 | MNI | p < 0.05f , h |
N = Number of participants.
Confirmed with study authors (personal communication).
The method of correction for multiple comparisons (FWE, FDR, Monte‐Carlo, or whether there was no correction) is specified in table notes. FWE, familywise error rate; FDR, false discovery rate. Note that whether a correction was applied voxel‐wise or cluster‐wise was sometimes specified even if the method was not, but was itself not always specified.
Uncorrected for multiple comparisons.
Corrected for multiple comparisons, method not specified.
FWE.
FDR.
Monte‐Carlo based correction.
Voxel‐wise (whole brain).
Voxel‐wise (small volume correction).
Cluster‐wise (whole brain).
Set‐level (whole brain).
Table 3.
Sentence production studies
| ID | Study | Method | Language | Eliciting stimulus | Type of sentence produced | Contrast description | Na | # of foci | Reported coordinate space | Statistical thresholdb |
|---|---|---|---|---|---|---|---|---|---|---|
| 46 | Braun et al. (2001) | PET |
English ASL |
Auditory prompt Auditory prompt |
Personal story (unconstrained) Personal story (unconstrained) |
English > oral‐motor baseline ASL > limb‐motor baseline |
12 |
21 21 |
MNI MNI |
p < 0.01d p < 0.01d |
| 47 | Blank, Scott, Murphy, Warburton, and Wise (2002) | PET | English | Auditory prompt | Personal story (unconstrained) | Speech > counting + nursery rhyme recitation | 8 | 10 | MNI | p < 0.05d , g |
| 48 | Haller, Radue, Erb, Grodd, and Kircher (2005) | FMRI | German | Visual words (verb, noun, noun) | Active transitive | Sentence generation > word reading | 15 | 9 | Talairach | p < 0.05e , g |
| Sentence generation > sentence reading | 7 | Talairach | p < 0.05e , g | |||||||
| 49 | Kemeny, Ye, Birn, and Braun (2005) | FMRI | English | Auditory verb | Constrained (must use verb) | Sentence generation > syllable repetition (pa‐ta‐ka) | 6 | 10 | MNI | p < 0.01d , g |
| 50 | Troiani et al. (2008) | FMRI | English | Sequence of story pictures | Story (unconstrained) | Sentence generation > story viewing + syllable repetition (pa‐da‐ka) | 15 | 5 | Talairach | p < 0.01c |
| 51 | Tremblay and Small (2011) | FMRI | English | Line drawing of an object | Object description | Sentence generation > picture viewing | 21 | 18 | Talairach | p < 0.05d , g |
| 52 | Menenti et al. (2011) | FMRI | Dutch | Complex photograph | Active or passive | Repeated words in production > repeated words in comprehension | 20 | 6 | MNI | p < 0.05d , g |
| 53 | Geranmayeh et al. (2012) | FMRI | English | Picture of an object | Unconstrained | Speech > tongue movement | 19 | 10 | MNI | p < 0.05d , g |
| 54 | Grande et al. (2012) | FMRI | German | Picture of a scene / frequent words | Constrained (avoid frequent words) | Sentence generation > rest | 18 | 7 | MNI | p < 0.05d , g |
| 55 | Menenti, Segaert, and Hagoort (2012) | FMRI | Dutch | Complex photograph | Active/passive | Novel semantics > repeated semantics | 20 | 8 | MNI | p < 0.05d , g |
| Novel words > repeated words | 11 | MNI | p < 0.05d , g | |||||||
| Novel syntax > repeated syntax | 11 | MNI | p < 0.05d , g | |||||||
| 56 | Menenti, Petersson, and Hagoort (2012) | FMRI | Dutch | Complex photograph | Active (transitive) | Novel words > repeated words | 24 | 3 | MNI | p < 0.05d , g |
| Novel meaning > repeated meaning | 9 | MNI | p < 0.05d , g | |||||||
| 57 | Geranmayeh, Wise, Mehta, and Leech (2014) | FMRI | English | Picture of an object | Unconstrained | Speech > counting | 24 | 12 | MNI | p < 0.05d , g |
| 58 | Schönberger et al. (2014) | FMRI | German | Line drawing of a complex scene | Constrained (use only 3 words) | Generation of simple complete sentences > sentences missing verb | 15 | 9 | MNI | p < 0.05f , g |
| 59 | Simmonds, Leech, Collins, Redjep, and Wise (2014) | FMRI | English | Visual word (noun) | Unconstrained | Overt noun definition > rest (look at a row of Xs) | 17 | 14 | MNI | p = 0.05d , g |
| 60 | Matchin and Hickok (2016) | FMRI | English | Sequence of three pictures with arrows and verb | Active or passive | Sentence generation > word lists | 20 | 4 | Talairach | p < 0.05d , g |
N = Number of participants.
Confirmed with study authors (personal communication; none in this Table).
The method of correction for multiple comparisons (FWE, FDR, Monte‐Carlo, or whether there was no correction) is specified in table notes. FWE, familywise error rate; FDR, false discovery rate. Note that whether a correction was applied voxel‐wise or cluster‐wise was sometimes specified even if the method was not, but was itself not always specified.
Uncorrected for multiple comparisons.
Corrected for multiple comparisons, method not specified.
FWE.
FDR.
Monte‐Carlo based correction.
Voxel‐wise (whole brain).
Voxel‐wise (small volume correction).
Cluster‐wise (whole brain).
Set‐level (whole brain).
The analysis identified six significant clusters (Table 4). In the frontal lobe, a large cluster was found in left IFG (pars triangularis and pars opercularis), and insula, crossing as well into left temporal pole. Superior to this cluster was a smaller distinct cluster in left MFG and precentral gyrus, and more superior was a cluster in left SMA and SFG. The largest cluster was in the left MTG (anterior and posterior divisions, extending into the temporooccipital part of this region), posterior STG, and angular gyrus (AG). Another smaller cluster was found in the posterior division of left temporal fusiform cortex and left posterior parahippocampal gyrus. One cluster was found in the right temporal lobe, in right temporal pole, MTG (anterior, posterior, temporooccipital part), and posterior STG.
Table 4.
Significant clusters for sentence processing (comprehension and production combined)
| Cluster | Regiona | X | Y | Z | ALE extrema value | Cluster volume | Studies with foci within clusterb |
|---|---|---|---|---|---|---|---|
| 1 | L IFG, pars triangularis | −48 | 30 | 0 | 0.049 | 15,712 mm3 | 4, 5, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 21, 22, 23, 26, 29, 31, 32, 33, 34, 35, 36, 37, 38, 39, 40, 41, 42, 43, 44, 46, 48, 49, 51, 53, 57 |
| −50 | 22 | 16 | 0.044 | ||||
| L IFG, pars opercularis | −48 | 10 | 28 | 0.027 | |||
| L insula | −34 | 22 | 2 | 0.022 | |||
| L temporal pole | −54 | 8 | −22 | 0.037 | |||
| 2 | L precentral gyrus | −48 | 0 | 50 | 0.046 | 3,936 mm3 | 10, 11, 15, 18, 19, 21, 22, 30, 35, 39, 41, 44, 45, 46, 50, 55, 57 |
| L MFG | −42 | 6 | 50 | 0.042 | |||
| 3 | L SMAc | −4 | 8 | 62 | 0.040 | 4,088 mm3 | 8, 10, 14, 15, 16, 18, 27, 34, 39, 40, 41, 42, 44, 47, 48, 49, 51, 53, 54, 57, 59 |
| L SFG | −4 | 10 | 56 | 0.039 | |||
| 4 | L aMTG | −58 | −6 | −12 | 0.046 | 18,520 mm3 | 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 17, 18, 20, 23, 24, 25, 26, 28, 30, 31, 32, 33, 34, 35, 36, 37, 38, 39, 41, 42, 43, 44, 45, 46, 48, 49, 51, 53, 55, 58, 59 |
| L pMTG | −56 | −38 | 2 | 0.063 | |||
| L MTG, temporooccipital part | −60 | −50 | 8 | 0.049 | |||
| L pSTG | −62 | −18 | −2 | 0.041 | |||
| L angular gyrus | −46 | −60 | 20 | 0.029 | |||
| 5 | L temporal fusiform cortex, posterior division | −38 | −40 | −20 | 0.032 | 2,664 mm3 | 11, 12, 15, 19, 21, 26, 30, 41, 42, 45, 46, 47, 55, 57 |
| L parahippocampal gyrus, posterior division | −26 | −36 | −16 | 0.021 | |||
| 6 | R temporal pole | 50 | 12 | −22 | 0.033 | 9,184 mm3 | 4, 5, 6, 7, 8, 9, 10, 11, 12, 14, 15, 16, 19, 20, 21, 22, 23, 24, 26, 28, 29, 31, 32, 33, 35, 36, 38, 39, 43, 49, 51, 54, 59 |
| R aMTG | 58 | 2 | −22 | 0.025 | |||
| 62 | −6 | −10 | 0.039 | ||||
| R pMTG | 56 | −36 | −12 | 0.018 | |||
| R MTG, temporooccipital part | 52 | −40 | −4 | 0.021 | |||
| 56 | −40 | −4 | 0.022 | ||||
| R pSTG | 60 | −20 | 0 | 0.046 |
L, left; R, right; a, anterior; p, posterior; IFG, inferior frontal gyrus; MFG, middle frontal gyrus; SMA, supplementary motor area; SFG, superior frontal gyrus; MTG, middle temporal gyrus; STG, superior temporal gyrus.
The numbers in this column are the IDs of the studies listed in Tables 1–3 that have at least one focus within the cluster.
The Harvard–Oxford label for this region is given as Juxtapositional lobule cortex (formerly Supplementary motor area).
3.2. Sentence comprehension
A separate meta‐analysis focused only on studies examining sentence comprehension. This included 579 foci from 742 participants in 45 studies (all of the studies listed in Tables 1 and 2). Analysis of these studies revealed six significant clusters with extrema in left perisylvian regions, left temporal occipital fusiform cortex, and the right temporal lobe (Table 5, Figure 1). Three clusters were found in left frontal lobe. The first was in left IFG (with extrema in pars triangularis), extending into insular cortex and left temporal pole; the second in left precentral gyrus, and the third in left SMA. The largest cluster was in left posterior temporal and inferior parietal cortex, with extrema in the MTG (anterior, posterior, and temporooccipital part), posterior STG, and AG. Another cluster in the left temporal lobe peaked in temporal occipital fusiform cortex. One cluster was found in the right temporal lobe, spanning right temporal pole, MTG (anterior, posterior, and temporooccipital part), and posterior STG.
Table 5.
Significant clusters for sentence comprehension (combined auditory and visual studies)
| Cluster | Regiona | X | Y | Z | ALE extrema value | Cluster volume | Studies with foci within clusterb |
|---|---|---|---|---|---|---|---|
| 1 | L IFG, pars triangularis | −48 | 30 | 0 | 0.046 | 14,688 mm3 | 4, 5, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 21, 22, 23, 26, 29, 31, 32, 33, 34, 35, 36, 37, 38, 39, 40, 41, 42, 43, 44 |
| −50 | 22 | 16 | 0.034 | ||||
| L insula | −34 | 22 | 2 | 0.018 | |||
| L temporal pole | −54 | 8 | −22 | 0.037 | |||
| −44 | 18 | −20 | 0.020 | ||||
| 2 | L precentral gyrus | −48 | −2 | 50 | 0.038 | 3,144 mm3 | 10, 11, 15, 18, 19, 21, 22, 30, 32, 35, 39, 41, 44, 45 |
| 3 | L SMAc | −4 | 8 | 58 | 0.031 | 1,944 mm3 | 8, 10, 14, 15, 18, 27, 39, 40, 41, 42, 44 |
| −6 | 8 | 52 | 0.031 | ||||
| 4 | L aMTG | −58 | −6 | −12 | 0.046 | 18,184 mm3 | 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 17, 18, 19, 20, 21, 23, 24, 25, 26, 28, 30, 31, 32, 33, 34, 35, 36, 37, 38, 39, 41, 42, 43, 44, 45 |
| L pMTG | −58 | −36 | 0 | 0.057 | |||
| L MTG, temporooccipital part | −56 | −58 | 10 | 0.032 | |||
| L pSTG | −62 | −18 | −2 | 0.041 | |||
| L angular gyrus | −56 | −56 | 14 | 0.032 | |||
| 5 | L temporal occipital fusiform cortex | −42 | −44 | −18 | 0.026 | 1,776 mm3 | 11, 12, 15, 19, 21, 26, 30, 32, 41, 42, 45 |
| 6 | R temporal pole | 50 | 12 | −22 | 0.033 | 9,264 mm3 | 4, 5, 6, 7, 8, 9, 10, 11, 12, 14, 15, 16, 19, 20, 21, 22, 23, 24, 26, 28, 29, 31, 32, 33, 35, 36, 38, 39, 43 |
| R aMTG | 62 | −6 | −10 | 0.037 | |||
| 58 | 2 | −22 | 0.033 | ||||
| R pMTG | 56 | −36 | −12 | 0.017 | |||
| R MTG, temporooccipital part | 58 | −40 | −4 | 0.017 | |||
| 52 | −40 | −4 | 0.016 | ||||
| R pSTG | 60 | −20 | 0 | 0.045 |
L, left; R, right; a, anterior; p, posterior; IFG, inferior frontal gyrus; SMA, supplementary motor area; MTG, middle temporal gyrus; STG, superior temporal gyrus.
The numbers in this column are the IDs of the studies listed in Tables 1 and 2 that have at least one focus within the cluster.
The Harvard‐Oxford label for this region is given as Juxtapositional lobule cortex (formerly Supplementary motor area).
Figure 1.
Significant clusters associated with sentence comprehension (both auditory and visual) derived from ALE analysis: Left lateral (a), from above (left hemisphere is on the left) (b), right lateral (c), axials by z‐coordinates in MNI space (d). The scale bar reflects the ALE values. Cluster labels correspond to the numbering in Table 5
3.3. Sentence production
For sentence production, 15 studies (Table 3) with 205 foci from 254 participants were included. The ALE analysis of these production studies identified five significant clusters (Table 6a; Figure 2). One cluster peaked in left MFG. The second cluster peaked in bilateral SFG, extending posteriorly and inferiorly into left SMA. The third was found in the temporooccipital part of left MTG. The fourth cluster was in the superior division of left lateral occipital cortex; the fifth in the superior division of right lateral occipital cortex.
Table 6.
Significant clusters for the sentence production analysis (a), and for the conjunction of comprehension and production (b)
| Cluster | Regiona | X | Y | Z | Extrema value | Volume | Studies with foci within clusterb |
|---|---|---|---|---|---|---|---|
| a. Sentence production | |||||||
| 1 | L MFG | −42 | 6 | 50 | 0.024 | 920 mm3 | 46, 50, 55, 57 |
| 2 | L superior frontal gyrusc | −4 | 10 | 66 | 0.023 | 2,304 mm3 | 47, 48, 49, 51, 53, 54, 57, 59 |
| −8 | 18 | 52 | 0.015 | ||||
| R superior frontal gyrus | 4 | 14 | 58 | 0.018 | |||
| 3 | L MTG, temporooccipital part | −58 | −48 | 8 | 0.020 | 1,056 mm3 | 48, 51, 55, 58 |
| 4 | L lateral occipital cortex, superior division | −44 | −74 | 24 | 0.017 | 792 mm3 | 46, 56, 58 |
| −42 | −68 | 18 | 0.013 | ||||
| 5 | R lateral occipital cortex, superior division | 46 | −66 | 22 | 0.017 | 960 mm3 | 46, 55, 56 |
| 42 | −72 | 26 | 0.014 | ||||
| b. Clusters in common between comprehension and production | |||||||
| 1 | L SMAd | −4 | 8 | 64 | 0.019 | 512 mm3 | 15, 39, 54 |
| Medial SFG | 0 | 12 | 58 | 0.014 | |||
| 2 | L MFG | −44 | 6 | 50 | 0.020 | 536 mm3 | 32, 46, 50, 55, 57 |
| 3 | L MTG, temporooccipital part | −58 | −48 | 8 | 0.020 | 1,056 mm3 | 5, 30, 32, 38, 48, 51, 55, 58 |
L, left; MFG, middle frontal gyrus; MTG, middle temporal gyrus; SFG, superior frontal gyrus; SMA, supplementary motor area.
The numbers in this column are the IDs of the studies listed in Tables 1, 2, 3 that have at least one focus within the cluster.
The cluster extends posteriorly into SMA.
The Harvard‐Oxford label for this region is given as Juxtapositional lobule cortex (formerly Supplementary motor area).
Figure 2.
Clusters of significant activation derived from the production meta‐analysis: Left lateral (a), from above (left hemisphere is on the left) (b), right lateral (c), axials by z‐coordinates in MNI space (d). The scale bar reflects the ALE values. Cluster labels correspond to the numbering in Table 6a
With respect to the question of overlap between comprehension and production, the conjunction map of these comprehension and production results yielded three clusters (Table 6b), all in the left hemisphere. The first was in left SMA and medial SFG, the second peaked in left MFG, and the third was found in the temporooccipital portion of left MTG.
3.4. Auditory versus visual comprehension
To examine modality differences between auditory and visual comprehension, we split the 45 comprehension studies into two data sets by modality of stimulus presentation. Thirty‐one studies contributed to the auditory modality (all studies from Table 1), with 14 in the visual modality (all studies from Table 2).
For the auditory modality, results showed six significant clusters (Table 7a, Figure 3a), derived from 339 foci from 523 participants. Three clusters were found in the frontal lobe. The largest was in left IFG (pars orbitalis, pars triangularis, and pars opercularis) and the insula, extending into left temporal pole. The second occupied left precentral gyrus. The third peaked in left SMA and left paracingulate gyrus. In the temporal lobe, a large cluster spread across left anterior and posterior MTG, posterior STG, and into the AG. Another cluster had a peak within the posterior division of left temporal fusiform cortex. Another large cluster was found in right temporal lobe (temporal pole, anterior and posterior MTG, and posterior STG).
Table 7.
Significant clusters for auditory (a) and visual (b) sentence comprehension
| Cluster | Regiona | X | Y | Z | ALE extrema value | Cluster volume | Studies with foci within clusterb |
|---|---|---|---|---|---|---|---|
| a. Sentence comprehension (auditory) | |||||||
| 1 | L IFG pars orbitalis | −44 | 26 | −18 | 0.024 | 8,880 mm3 | 4, 5, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 21, 22, 23, 26, 29, 31 |
| −48 | 26 | −8 | 0.020 | ||||
| −38 | 30 | −2 | 0.019 | ||||
| L IFG, pars triangularis | −56 | 24 | 12 | 0.025 | |||
| L IFG, pars opercularis | −50 | 14 | 16 | 0.022 | |||
| −44 | 28 | −2 | 0.019 | ||||
| L insula | −34 | 22 | 0 | 0.016 | |||
| L temporal pole | −56 | 8 | −22 | 0.033 | |||
| 2 | L precentral gyrus | −50 | −2 | 50 | 0.023 | 1,568 mm3 | 10, 11, 15, 18, 19, 21, 22, 30 |
| 3 | L SMAc | −2 | 8 | 58 | 0.018 | 1,000 mm3 | 8, 10, 14, 15, 18 |
| L paracingulate gyrus | −6 | 10 | 52 | 0.017 | |||
| 4 | L aMTG | −58 | −6 | −12 | 0.042 | 13,064 mm3 | 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 17, 18, 19, 20, 21, 23, 24, 25, 26, 28, 30, 31 |
| L pMTG | −58 | −38 | 2 | 0.035 | |||
| L pSTG | −62 | −18 | −2 | 0.041 | |||
| L angular gyrus | −52 | −52 | 14 | 0.022 | |||
| 5 | L temporal fusiform cortex, posterior division | −34 | −40 | −22 | 0.020 | 872 mm3 | 11, 12, 21, 26 |
| 6 | R temporal pole | 50 | 14 | −22 | 0.031 | 8,728 mm3 | 4, 5, 6, 7, 8, 9, 10, 12, 13, 14, 15, 19, 20, 21, 22, 23, 24, 26, 28, 29, 30, 31 |
| R aMTG | 58 | 2 | −22 | 0.023 | |||
| 62 | −6 | −10 | 0.037 | ||||
| R pMTG | 58 | −34 | 0 | 0.013 | |||
| R pSTG | 60 | −18 | 0 | 0.035 | |||
| b. Sentence comprehension (visual) | |||||||
| 1 | L IFG, pars triangularis | −50 | 30 | 0 | 0.031 | 3,976 mm3 | 32, 33, 34, 35, 36, 37, 38, 39, 41, 42, 43, 44 |
| 2 | L IFG, pars triangularis | −48 | 24 | 16 | 0.022 | 2,464 mm3 | 32, 34, 36, 39, 40, 41, 42 |
| L MFG | −48 | 8 | 28 | 0.016 | |||
| 3 | L precentral gyrus | −46 | 0 | 48 | 0.018 | 1,384 mm3 | 32, 34, 36, 39, 40, 41, 42 |
| 4 | L supplementary motor areac | −6 | 8 | 60 | 0.016 | 848 mm3 | 39, 40, 41, 42, 44 |
| 5 | L pMTG | −58 | −36 | 0 | 0.025 | 7,120 mm3 | 32, 33, 34, 35, 36, 37, 38, 39, 41, 42, 43, 44, 45 |
| −58 | −42 | −4 | 0.025 | ||||
| L MTG, temporooccipital part | −60 | −56 | 10 | 0.021 | |||
| −58 | −58 | 10 | 0.021 | ||||
| L angular gyrus | −58 | −52 | 18 | 0.013 | |||
| −46 | −56 | 14 | 0.013 | ||||
| −48 | −58 | 18 | 0.013 | ||||
| 6 | R MTG, temporooccipital part | 56 | −42 | −6 | 0.015 | 720 mm3 | 35, 36, 39 |
| 52 | −40 | −4 | 0.014 | ||||
L, left; R, right; a, anterior; p, posterior; IFG, inferior frontal gyrus; MFG, middle frontal gyrus; SMA, supplementary motor area; MTG, middle temporal gyrus; STG, superior temporal gyrus.
The numbers in this column are the IDs of the studies listed in Table 1 or Table 2 that have at least one focus within the cluster.
The Harvard–Oxford label for this region is given as Juxtapositional lobule cortex (formerly Supplementary motor area).
Figure 3.
Left hemisphere, midline, and right hemisphere views of significant clusters: Auditory comprehension vs. baseline (a), visual comprehension vs. baseline (b), conjunction map between auditory and visual studies (c). Cluster labels correspond to the numbering in Table 7a (auditory), Table 7b (visual), and Table 8 (conjunction). The scale bar reflects the ALE values
For the visual modality, the ALE analysis included 240 foci from 219 participants. The results indicated six significant clusters (Table 7b, Figure 3b). Four clusters were found in the frontal lobe: one in left IFG (pars triangularis), one spanning L IFG (pars triangularis) and L MFG, one in left precentral gyrus, and one in left SMA. Another large cluster was found in left posterior temporal / inferior parietal cortex (posterior MTG and the temporooccipital part of MTG, extending up into AG), and a small cluster was found in the right hemisphere (in the temporooccipital part of MTG).
To examine the overlap between auditory and visual modalities (Table 8, Figure 3c) we computed the conjunction map between them. Results revealed five significant clusters, all in the left hemisphere: four in frontal cortex and one in temporal and parietal cortices. In the frontal lobe, clusters of significant overlap were in IFG (pars orbitalis, pars triangularis, and pars opercularis), a distinct cluster in pars opercularis of IFG, another in precentral gyrus, and one more in SMA. In the temporal lobe, one cluster was found in MTG (posterior and temporooccipital parts), extending superiorly to AG.
Table 8.
Significant clusters derived from the conjunction map for auditory versus visual comprehension
| Cluster | Regiona | X | Y | Z | ALE Extrema (z‐score) | Cluster volume | Studies with foci within clusterb |
|---|---|---|---|---|---|---|---|
| 1 | L IFG, pars orbitalis | −48 | 26 | −6 | 0.019 | 1,376 mm3 | 8, 9, 17, 29, 33, 36, 41, 43, 44 |
| L IFG, pars triangularis | −50 | 30 | 4 | 0.016 | |||
| L IFG, pars opercularis | −46 | 28 | −2 | 0.018 | |||
| 2 | L IFG, pars opercularis | −50 | 20 | 16 | 0.016 | 608 mm3 | 18, 21, 22, 41 |
| 3 | L precentral gyrus | −48 | −2 | 50 | 0.018 | 664 mm3 | 15, 18, 22, 39, 44 |
| 4 | L SMAc | −4 | 8 | 60 | 0.015 | 432 mm3 | 10, 14, 18, 41, 42 |
| −6 | 8 | 54 | 0.015 | ||||
| 5 | L MTG, temporooccipital part | −56 | −56 | 12 | 0.015 | 3,280 mm3 | 5, 6, 7, 8, 11, 12, 17, 24, 25, 30, 32, 38, 41, 42, 43, 44, 45 |
| −62 | −46 | 6 | 0.016 | ||||
| −60 | −50 | 8 | 0.014 | ||||
| L pMTG | −58 | −36 | 0 | 0.025 | |||
| L angular gyrus | −58 | −52 | 18 | 0.013 |
L, left; p, posterior; IFG, inferior frontal gyrus; SMA, supplementary motor area; MTG, middle temporal gyrus.
The numbers in this column are the IDs of the studies listed in Tables 1 and 2 that have at least one focus within the cluster.
The Harvard–Oxford label for this region is given as juxtapositional lobule cortex (formerly Supplementary motor area).
3.5. Comprehension of noncanonical sentences
Finally, we examined studies that reported a contrast of sentences with noncanonically ordered arguments compared against those with canonically ordered arguments. A total of 226 foci from 665 unique participants in 37 studies were included in the meta‐analysis (Table 9). All but eight studies (contributing PET contrasts) used fMRI. The studies were conducted with speakers of English, German, Hebrew, Kaqchikel, Japanese, and Mandarin. In 15 of the studies, participants were explicitly claimed to be monolingual native speakers of the language used in the study, in 16 of the studies participants were claimed only to be native speakers, one study employed a bilingual group (native in both languages), and 5 studies did not specify the language background of their participants. Most of the participants were right handed (n = 645); handedness was not specified for 18 people (both groups from Waters, Caplan, Alpert, & Stanczak, 2003), and 2 participants from one study were left handed. The tasks employed across the studies were variable, including comprehension questions, sentence picture‐matching, or a binary judgment about some aspect of the sentence. The number of participants in each study ranged from a low of 7 to a high of 50. Table 9 also specifies the coordinate space the data were originally reported in, and the statistical threshold for each study's data. For the statistical thresholds, we give the p‐value and the method of correction for multiple comparisons (or whether it was uncorrected). We include random field theory and related methods (e.g., white‐noise based approaches) under the umbrella term of familywise error rate (FWE) based corrections, and distinguish these from Monte Carlo and False Discovery Rate (FDR) based approaches (Nichols, 2012; Nichols & Hayasaka, 2003). Note that authors did not always unambiguously specify the method by which a correction was applied; see individual articles for specific details.
Table 9.
Studies contrasting noncanonical and canonical sentences
| ID | Study | Method | Language | Task | Contrast description | Na | # of foci | Reported coordinate space | Statistical thresholdb |
|---|---|---|---|---|---|---|---|---|---|
| 61 | Stromswold, Caplan, Alpert, and Rauch (1996) | PET | English | Plausibility judgment | Visual object relative > subject relative | 8 | 1 | Talairach | p < 0.05e |
| 62 | Caplan, Alpert, and Waters (1999) | PET | English | Plausibility judgment | Auditory object cleft > subject clefts | 16 | 3 | Talairach | Not specifiede, h, i |
| 63 | Caplan, Alpert, Waters, and Olivieri (2000) | PET | English | Plausibility judgment | Visual object relative > subject relative | 11 | 1 | Talairach | p < 0.01e , i |
| 64 | Cooke et al. (2001) | FMRI | English | Gender judgment of agent | Visual short object relatives > short subject relatives | 7 | 3 | Talairach | p < 0.002c |
| Visual long object relatives > long subject relatives | 4 | Talairach | p < 0.002c | ||||||
| 65 | Caplan, Waters, and Alpert (2003) | PET | English | Plausibility judgment | Visual object relatives > subject relatives (elderly slow responders) | 13 | 2 | Talairach | Not specifiede , i |
| 66 | Caplan et al. (2003) | PET | English | Plausibility judgment | Visual object relatives > subject relatives (young slow responders) | 8 | 2 | Talairach | Not specifiede , i |
| 67 | Caplan et al. (2003) | PET | English | Plausibility judgment | Visual object relatives > subject relatives (elderly fast responders) | 9 | 1 | Talairach | Not specifiede , i |
| 68 | Waters et al. (2003) | PET | English | Plausibility judgment | Visual object relatives > subject relatives (high working memory subjects) | 9 | 3 | MNI | Not specifiede , i |
| 69 | Waters et al. (2003) | PET | English | Plausibility judgment | Visual object relatives > subject relatives (low working memory subjects) | 9 | 3 | MNI | Not specifiede , i |
| 70 | Constable et al. (2004) | FMRI | English | Acceptability judgment | Auditory and visual object relative > subject relative | 20 | 8 | Talairach | p < 0.01c |
| 71 | Grewe et al. (2005) | FMRI | German | Acceptability judgment | Visual nominal object‐before‐subject > nominal subject‐before object | 15 | 7 | Talairach | p < 0.001c |
| 72 | Yokoyama et al. (2006) | FMRI | Japanese | Plausibility judgment | Visual passive vs. active (Japanese) | 36 | 4 | Talairach | p < 0.001c |
| 73 | Bahlmann et al. (2007) | FMRI | German | Offline sentence recognition test | Visual noncanonical sentences (object initial) > canonical sentences (subject initial) | 12 | 2 | MNI | p < 0.001c |
| 74 | Yokoyama et al. (2007) | FMRI | Japanese | Agent/patient identification | Visual passives > actives | 20 | 2 | Talairach | p < 0.001c |
| 75 | Caplan, Stanczak, and Waters (2008) | FMRI | English | Plausibility judgment | Visual object relatives > subject relatives (unconstrained context) | 16 | 8 | MNI | p < 0.001e , h |
| Visual object relatives > subject relatives (constrained context) | 6 | MNI | p < 0.001e , h | ||||||
| 76 | Caplan, Chen, and Waters (2008) | FMRI | English |
Sentence verification (early TR) Sentence verification (late TR) |
Visual object relatives > subject relatives | 15 |
1 10 |
Talairach Talairach |
|
| 77 | Kinno et al. (2008) | Japanese | Sentence‐picture matching | Visual passive > active | 14 | 1 | MNI | p < 0.05f , i | |
| Visual scrambled > active | 3 | MNI | p < 0.05f , i | ||||||
| 78 | Kovelman et al. (2008) | FMRI | English | Plausibility judgment | Monolingual English: Visual object relatives > subject relatives | 10 | 8 | MNI | p < 0.001c |
| 79 | Kovelman et al. (2008) | FMRI | English | Plausibility judgment | Bilingual English: Visual object relatives > subject relatives | 11 | 9 | MNI | p < 0.001c |
| 80 | Rogalsky, Matchin, and Hickok (2008) | FMRI | English | Plausibility judgment | Auditory object relatives > subject relatives | 15 | 7 | Talairach | p < 0.005c |
| Plausibility judgment with concurrent articulation task | Auditory object relatives > subject relatives | 4 | Talairach | p < 0.005c | |||||
| Plausibility judgment with concurrent motor task | Auditory object relatives > subject relatives | 6 | Talairach | p < 0.005c | |||||
| 81 | Bornkessel‐Schlesewsky, Schlesewsky, and von Cramon (2009) | FMRI | German | Grammaticality judgment | Visual noncanonical sentences (object first) > canonical sentences (subject first) | 30 | 7 | Talairach | p < 0.001c |
| 82 | Kim et al. (2009) | FMRI | Japanese | Acceptability judgment | Visual scrambled noncanonical sentences (OSV) > canonical (SOV) sentences | 36 | 2 | Talairach | p < 0.05e , j |
| 83 | Ye and Zhou (2009) | FMRI | Mandarin | Probe verification | Visual passive > active | 19 | 4 | MNI | p < 0.001c |
| 84 | Meltzer et al. (2010) | FMRI | English | Sentence‐picture matching | Auditory object relative > subject relative (sentence) | 24 | 1 | Talairach | p < 0.05g , j |
| Auditory reversible object relatives > reversible subject relatives (sentence) | 3 | Talairach | p < 0.05g , j | ||||||
| Auditory irreversible object relatives > irreversible subject relatives (sentence) | 1 | Talairach | p < 0.05g , j | ||||||
| Auditory reversible object relatives > reversible subject relatives (post‐sentence picture) | 4 | Talairach | p < 0.05g , j | ||||||
| 85 | Santi and Grodzinsky (2010) | FMRI | English | Sentence comparison | Auditory object relatives > subject relatives | 17 | 1 | Talairach | p < 0.05g , j |
| 86 | Thompson, Bonakdarpour, and Fix (2010) | FMRI | English | Sentence‐picture matching | Auditory object clefts > subject clefts | 12 | 8 | MNI | p < 0.05f , j |
| 87 | Wilson et al. (2010) | FMRI | English | Sentence‐picture matching | Auditory noncanonical (passives, object relatives) > canonical (actives, subject relatives) | 24 | 12 | MNI | p < 0.05e , j |
| 88 | Hirotani, Makuuchi, Ruschemeyer, and Friederici (2011) | FMRI | Japanese | Sentence comprehension question | Auditory passives > actives | 16 | 10 | MNI | p < 0.05d , j |
| 89 | Prat and Just (2010) | FMRI | English | True/false memory probes | Visual object relatives > conjoined actives | 27 | 19 | MNI | p < 0.05f , j |
| 90 | Bornkessel‐Schlesewsky, Grewe, and Schlesewsky (2012) | FMRI | German | Acceptability judgment | Visual noncanonical sentences (object first) > canonical sentences (subject first) | 18 | 6 | Talairach | p < 0.05g , j |
| 91 | Meyer et al. (2012) | FMRI | German | Comprehension questions | Auditory noncanonical (OSV) > canonical (SOV) | 24 | 1 | MNI | p < 0.05g , j |
| 92 | Santi and Grodzinsky (2012) | FMRI | English | Sentence comprehension question | Auditory object Wh‐question > subject Wh‐question | 14 | 11 | Talairach | p < 0.05g , j |
| 93 | Mack, Ji, and Thompson (2013) | FMRI | English | Sentence‐picture matching | Auditory passives > actives | 27 | 6 | MNI | p < 0.001c |
| 94 | Newman et al. (2013) | FMRI | English | Sentence reading and comprehension probes | Visual object relatives > conjoined actives (sentence reading) | 50 | 3 | MNI | p < 0.001c |
| Visual object relatives > conjoined actives (probe phase) | 6 | MNI | p < 0.001c | ||||||
| 95 | Shetreet and Friedmann (2014) | FMRI | Hebrew | Semantic decision | Auditory noncanonical sentences with wh‐movement > canonical sentences | 22 | 6 | Talairach | p < 0.05d , j |
| 96 | Rogalsky, Almeida, Sprouse, and Hickok (2015) | FMRI | English | True/false questions or memorizing words | Auditory long object relatives > long yes/no embedded questions | 15 | 3 | Talairach | p = 0.005c |
| Auditory short object relatives > short yes/no embedded questions | 2 | Talairach | p = 0.005c | ||||||
| 97 | Koizumi and Kim (2016) | FMRI | Kaqchikel | Plausibility judgment | Auditory noncanonical (SVO) > canonical (VOS) | 16 | 1 | MNI | p < 0.05e |
N = Number of participants.
Confirmed with study authors (personal communication; none in this Table).
The method of correction for multiple comparisons (FWE, FDR, Monte‐Carlo, or whether there was no correction) is specified in table notes. FWE, familywise error rate; FDR, false discovery rate. Note that whether a correction was applied voxel‐wise or cluster‐wise was sometimes specified even if the method was not, but was itself not always specified.
Uncorrected for multiple comparisons.
Corrected for multiple comparisons, method not specified.
FWE.
FDR.
Monte‐Carlo based correction.
Voxel‐wise (whole brain).
Voxel‐wise (small volume correction).
Cluster‐wise (whole brain).
Set‐level (whole brain).
The results revealed seven significant clusters (Table 10, Figure 4). There was a small cluster in left frontal pole and a large cluster in left IFG (pars orbitalis and opercularis), with additional peaks in left MFG. There was another distinct cluster in left MFG. The fourth cluster was in SFG, and extended to the right and left paracingulate gyri (but did not extend into SMA). The fifth was in right insular cortex. The sixth cluster was in left posterior MTG, while the seventh was in left AG.
Table 10.
Significant clusters for noncanonical versus canonical sentence comprehension
| Cluster | Regiona | X | Y | Z | ALE Extrema value | Cluster volume | Studies with foci within clusterb |
|---|---|---|---|---|---|---|---|
| 1 | L frontal pole | −46 | 40 | 0 | 0.015 | 864 mm3 | 63, 69, 80, 97 |
| −42 | 44 | 4 | 0.014 | ||||
| 2 | L IFG, pars orbitalis | −34 | 26 | −2 | 0.026 | 10,296 mm3 | 61, 62, 68, 70, 71, 72, 73, 75, 76, 77, 78, 79, 80, 81, 83, 84, 86, 87, 88, 89, 90, 91, 92, 94 |
| L IFG, pars opercularis | −52 | 16 | 10 | 0.036 | |||
| L MFG | −44 | 22 | 30 | 0.022 | |||
| −40 | 10 | 32 | 0.012 | ||||
| 3 | L MFG | −42 | 4 | 48 | 0.026 | 1,712 mm3 | 70, 77, 82, 84, 86, 89, 93 |
| 4 | L SFG | −10 | 8 | 60 | 0.012 | 1,208 mm3 | 66, 69, 78, 83, 84, 87, 88, 93 |
| Medial SFG | 0 | 16 | 54 | 0.015 | |||
| L paracingulate gyrus | −6 | 10 | 52 | 0.013 | |||
| R paracingulate gyrus | 8 | 20 | 46 | 0.016 | |||
| 5 | R insula | 36 | 22 | −4 | 0.019 | 1,040 mm3 | 71, 81, 84, 87, 88 |
| 6 | L pMTG | −52 | −36 | −2 | 0.023 | 848 mm3 | 75, 76, 87, 88 |
| 7 | L angular gyrus | −44 | −56 | 18 | 0.020 | 1,024 mm3 | 75, 86, 87, 88, 96 |
L, left; R, right; p, posterior; IFG, inferior frontal gyrus; MFG, middle frontal gyrus; SFG, superior frontal gyrus; MTG, middle temporal gyrus.
The numbers in this column correspond to the IDs of the studies listed in Table 9 that have at least one focus within the cluster.
Figure 4.
Significant clusters for noncanonical vs. canonical contrasts: Left lateral (a), from above (left hemisphere is on the left) (b), right lateral (c). The scale bar reflects the ALE values. Cluster labels correspond to the numbering in Table 10
4. DISCUSSION
We used ALE meta‐analysis to identify the brain structures participating in sentence comprehension and sentence production, and the extent to which the identified regions are overlapping across these domains. We also examined whether or to what extent the brain regions for comprehending auditory sentences are the same as those for sentence comprehension in reading. Finally, we examined the set of regions involved in the comprehension of complex noncanonical structures. We discuss our findings for each of these questions of interest in turn below.
Before we turn to that discussion however, we briefly note that the choice of contrasts in fMRI does not necessarily allow for precise delineation of the functional roles of any activated regions (Caplan, 2009). That is, the contrasts between two conditions in fMRI studies (e.g., sentence comprehension vs. baseline, or even noncanonical vs. canonical sentences) do not allow one to identify a single process that differs between conditions—there may be any number of sub‐processes that contribute to the observed differences in activation. This is particularly the case for studies of sentence comprehension against a low‐level baseline, where activated regions could participate in any part of the process (e.g., lexical retrieval, structure building, etc.), and there is insufficient information to delimit functional roles for particular regions. Despite these limitations on interpretation, in what follows we briefly cover what's known about the likely functions of each region implicated in our analyses, with additional discussion of our questions of interest related to comparisons of production and comprehension, comparisons across modalities, and the comprehension of noncanonical sentences.
Note as well that regions may be active in fMRI because they contribute to task performance, not because they are related to language processing per se (Caplan, 2010). However, one advantage of meta‐analysis is that differential activation of regions due to such ancillary processes across studies that used different tasks or parameters may be factored out when results from these studies are pooled. Nevertheless, we also briefly consider nonlanguage functions for activated regions, particularly those areas that are not typically associated with language functions.
4.1. Language networks for comprehension and production
The results of the meta‐analysis of sentence comprehension against low‐level baselines revealed large clusters of activation in left temporo‐parietal‐occipital, left frontal, and right temporal lobe cortical regions. These results are largely consistent with regions proposed by previous models of sentence comprehension (Bornkessel‐Schlesewsky & Schlesewsky, 2013; Friederici, 2011). In what follows we describe functions that have been ascribed to each region where significant clusters were found. These functions are also broadly summarized in Figure 5.
Figure 5.
Important functions for each region contributing to sentence comprehension (labels in white boxes) and production (labels in gray boxes) in the left and right hemispheres. Colors highlight approximate locations for each region. See text for additional details
The left inferior frontal gyrus (IFG) has been implicated as an important region for sentence processing (Grodzinsky & Amunts, 2006). The functions of this region have been the subject of intense debate, with various claims for language‐specific roles as well as domain‐general roles. First, it has been claimed that the left IFG is critical for syntactic processing, in particular for phrase structure building and syntactic working memory (Friederici, 2002; Ullman, 2006), as well as the computation of complex, noncanonical sentences (Grodzinsky, 2000). In our analysis, the IFG was activated for comprehension of all sentences (both visually and auditorily), including complex structures. The IFG has also been implicated in lexical selection (Hickok & Poeppel, 2007; Thompson‐Schill, D'Esposito, Aguirre, & Farah, 1997) as well as in procedural and working memory (Ullman, 2006) and cognitive control (Bornkessel‐Schlesewsky & Schlesewsky, 2013), and may also respond to other general factors, such as processing load, attentional demands, task and stimulus complexity, and executive function (Colman, Koerts, Stowe, Leenders, & Bastiaanse, 2011; Friederici, 2002; Hickok, 2000; Love, Haist, Nicol, & Swinney, 2006; Michael et al., 2001; Rogalsky et al., 2017). Nearby, the insula is involved with functions supporting language comprehension as well as production, with bilateral involvement of this region in both domains, and has also been implicated in attention and working memory (Adank, 2012b; Oh, Duerden, & Pang, 2014).
We also found significant clusters spanning left supplementary motor area (SMA) and SFG (left lateralized in the comprehension analysis and combined comprehension/production analysis, but bilateral for production). Although more often associated with production processes, including those engaged for sentences (see below), the SMA has also been implicated in sentence comprehension (Jednoróg et al., 2015; Méndez Orellana et al., 2014), and may be particularly engaged when comprehension is effortful, such as when listening to noisy or acoustically degraded speech (Adank, 2012a, 2012b). The SMA and nearby cortical regions are also involved in nonlanguage functions, including response preparation (Corbetta & Shulman, 2002; Kristensen, Wang, Petersson, & Hagoort, 2013), cognitive control (Dosenbach et al., 2006; Henry, Berman, Nagarajan, Mukherjee, & Berger, 2004), attention, and motor sequencing (Cho et al., 2012).
With respect to clusters in left precentral and middle frontal gyri, the involvement of these areas in motor functions is well established (Petrides, 2016). The MFG has also been found to be involved in working memory and cognitive control (Abutalebi, 2008; Abutalebi & Green, 2007; Aron, 2007; Osaka, Komori, Morishita, & Osaka, 2007; Trumbo et al., 2016) and is part of a broader set of domain‐general regions engaged for cognitively demanding tasks (Duncan & Owen, 2000; Fedorenko, Duncan, & Kanwisher, 2013; Novais‐Santos et al., 2007).
In contrast to the frontal regions in the left hemisphere, clusters in the temporal lobes were bilateral in posterior STG and MTG (anterior, posterior, and temporooccipital part) and the temporal pole for comprehension, with clusters extending into the angular gyrus (AG) in the left hemisphere.
The posterior STG is strongly associated with lexical processing, including word‐level phoneme identification and word recognition (Bornkessel‐Schlesewsky & Schlesewsky, 2013; Hullett, Hamilton, Mesgarani, Schreiner, & Chang, 2016; Michael et al., 2001). This region also has long been associated with sentence comprehension, largely based on neuropsychological studies of patients with comprehension impairments (Baldo & Dronkers, 2007; Dronkers, Wilkins, Van Valin Jr., Redfern, & Jaeger, 2004; Magnusdottir et al., 2013). In a recent voxel‐based lesion symptom mapping study with 66 participants with left hemisphere stroke and aphasia (and agrammatic aphasia), poor performance on both canonical and noncanonical sentences was associated with damage to the left superior temporal gyrus, extending into the parietal lobe regions (Rogalsky et al., 2017). Thompson and colleagues have also found this pattern in several studies with agrammatic aphasic patients (Barbieri et al., 2018; Thompson, den Ouden, Bonakdarpour, Garibaldi, & Parrish, 2010; see also Thompson & Kielar, 2014 for review). However, recent studies of patients with primary progressive aphasia (PPA), a neurodegenerative disease affecting language networks, suggest that cortical tissue within this region may not comprise a “hub” for sentence comprehension. Mesulam, Thompson, Weintraub, and Rogalski (2015) examined the relation between sentence comprehension ability (based on the Northwestern Assessment of Verbs and Sentences [NAVS]; Thompson, 2011) and cortial atrophy in 63 PPA patients and found no relation between atrophy in the posterior STG and task performance. Even among the 10 patients who showed the lowest sentence comprehension scores, only half showed atrophy in this region. Rather, atrophy in the left inferior parietal lobule, as well as in anterior brain regions, was positively correlated with sentence comprehension impairments.
Peak regions of activation also were found in the MTG, extending to temporooccipital cortex. Like the STG, this region has been implicated in lexical processing (Hickok & Poeppel, 2000, 2004, 2007; Hagoort, 2005; see Lau, Phillips, & Poeppel, 2008 for review) and has been shown in previous neuroimaging studies to be responsive to both visual and auditory sentence structural manipulations (Wilson, Bautista, & McCarron, 2018). A recent study using electrocorticography (ECoG) (Nelson et al., 2017) also suggests that the posterior MTG is associated with predictive (top‐down) processing, which is engaged for sentence comprehension in healthy listeners (Lau, Stroud, Plesch, & Phillips, 2006; Mack, Ji, & Thompson, 2013; and others). These findings, coupled with the results of the present study, indicate a role for the pMTG in syntactic processing.
The temporal pole (anterior temporal lobe [ATL]), located within the posteroanterior ventral pathway and connecting to inferior frontal regions (Friederici, 2012), also has been associated with sentence processing, with studies reporting activation in this region for sentences contrasted with word lists, foreign speech, etc. (Bemis & Pylkkänen, 2011, 2012; Fedorenko, Nieto‐Castanon, & Kanwisher, 2012; Humphries, Binder, Medler, & Liebenthal, 2006; Rogalsky & Hickok, 2009), although damage to this region by surgical resection or PPA does not impair sentence comprehension (Cotelli et al., 2007; Gorno‐Tempini et al., 2004; Grossman, Rhee, & Moore, 2005; Kho et al., 2008; Mesulam, Wieneke, Thompson, Rogalski, & Weintraub, 2012). Rather, current research suggests that the ATL subserves semantic processing (Lambon Ralph, Sage, Jones, & Mayberry, 2010; Mesulam et al., 2012; Patterson, Nestor, & Rogers, 2007; Visser, Jefferies, & Lambon Ralph, 2010) and facilitates the combination of words into complex semantic representations (Pylkkänen, 2015; Zaccarella, Schell, & Friederici, 2017) in line with the idea that this region is crucial for object/entity knowledge and word comprehension ability (Mesulam et al., 2015; Mesulam et al., 2019).
The left inferior parietal lobule (i.e., the AG and SMG) also is associated with sentence level semantic representations and sentence comprehension (Indefrey, 2011; Indefrey & Levelt, 2004; Schönberger et al., 2014; Vigneau et al., 2006); though note that our analyses revealed no activation peaks in the SMG for sentence comprehension. Studies with stroke patients show, for example, that damage to the AG is associated with thematic (i.e., relational) errors (e.g., dog for bone). Similarly, a greater hemodynamic response in the AG has been reported for thematic compared with taxonomic (i.e., featural, categorical; e.g., dog–cat) relations between words (Boylan, Trueswell, & Thompson‐Schill, 2015, 2017; Kalénine et al., 2009; Lewis, Poeppel, & Murphy, 2015). Several fMRI studies also have shown increased activation in the AG as a function of the number of arguments (i.e., thematic roles, participant roles) and subcategorization frames encoded within the representation of action verbs in both young and older participants (Kang, Constable, Gore, & Avrutin, 1999; Kuperberg et al., 2000; Meltzer‐Asscher, Mack, Barbieri, & Thompson, 2015; Newman, Just, Keller, Roth, & Carpenter, 2003; Ni et al., 2000; Shetreet, Palti, Friedmann, & Hadar, 2007; Thompson et al., 2007; Thompson, Bonakdarpour, & Fix, 2010). Summarizing this latter literature, Thompson and Meltzer‐Asscher (2014) proposed that the left AG is associated with retrieval of argument structure information whereas the pMTG is involved in integrating verbs with their arguments. This is in line with models of sentence comprehension implicating the AG in integration of semantic and syntactic information at the sentence level (Friederici, Opitz, & von Cramon, 2000; Just, Carpenter, Keller, Eddy, & Thulborn, 1996; Stowe et al., 1998).
Finally, the posterior division of the temporal fusiform cortex is part of Brodmann area 37 (BA 37) and located in the ventrocaudal aspect of the temporal lobe, medial to the inferior temporal gyrus (ITG). A recent connectivity meta‐analysis on left BA 37 suggests that this region is a node in the semantic language network (Ardila, Bernal, & Rosselli, 2015). Connections from left BA 37 to left prefrontal cortex may be utilized for verbal reasoning while connections from BA 37 to left ITG may subserve lexical–semantic processing (Berthier, 1999). The temporal fusiform cortex has also been associated with semantic disambiguation (Davis et al., 2007; Rodd, Davis, & Johnsrude, 2005; Rodd, Johnsrude, & Davis, 2012; Zempleni, Renken, Hoeks, Hoogduin, & Stowe, 2007).
4.2. Comparing sentence comprehension and sentence production
The networks for sentence comprehension (across both modalities, relative to a low‐level baseline) and sentence production were similar in many respects (Tables 5 and 6 and Figures 1 and 2). Both networks contained clusters in left precentral/middle frontal gyrus and SMA. In the temporal lobe, both networks contained clusters in left posterior regions, particularly the temporooccipital portion of MTG. Two of the regions that we found in common between comprehension and production are consistent with those found in prior studies that compared processing across domains: left MTG and SMA (Indefrey et al., 2004;Menenti et al., 2011 ; Segaert et al., 2012 ; Silbert et al., 2014).
However, there were differences as well. In the comprehension analysis, temporal lobe activation was much more extensive in the left hemisphere than for production, and was largely bilateral as well. A cluster was found in fusiform cortex for comprehension but not production, though note that four production studies did contribute activation to the cluster in this region in the omnibus analysis of sentence comprehension and production. For production, clusters were found in bilateral lateral superior occipital cortex that were not found for comprehension. Activity in this region may be related to the earliest stages of lexical access (Braun et al., 2001; Menenti, Petersson, & Hagoort, 2012), and so may reflect domain‐specific differences in these early stages of processing.
The biggest difference that we found concerned the left inferior frontal cortex. We found a large cluster in this region for sentence comprehension, but no clusters in this region for sentence production, in spite of the purported importance of this region for sentence production (Menenti et al., 2011; Segaert et al., 2012; Silbert et al., 2014). One possibility for this finding is that it reflects low power. Adank (2012b) reports a cluster in this region for language production in a meta‐analysis based on 21 studies (we had 15); recent recommendations are for 20 studies to enable adequate power in cluster‐based meta‐analysis (Eickhoff et al., 2016). In Adank's results, that cluster was the smallest one resulting from her analysis (456 mm3), and it is possible that it reflected contributions from production of syllables or words in isolation, or from sentence comprehension processes required for the production tasks, rather than from sentence production per se. Related to concerns of low power, it may be that due to the variety of structures being produced, different portions of this region were active in different studies, but did not overlap sufficiently to produce a significant cluster. Consistent with this argument, we note that in the results of the omnibus analysis of sentence comprehension and production (Table 4), six production studies had activation peaks within the large cluster in left inferior frontal cortex (cluster 1).
Alternatively, it may be the case that the choice of baseline in many of the included contrasts worked against a finding of activation in left IFG. That is, because left IFG is activated by single word processing as well as by sentence processing, activation due to sentence production may not be revealed against baseline conditions that involve language (whether sentences or single words). Consistent with this suggestion, of the eight studies that had such linguistic baseline conditions (Table 3 IDs: 47, 48, 52, 55, 56, 57, 58, 60), only two had activation within the L IFG cluster from the omnibus analysis (cluster 1 in Table 4).
A third possibility is that the IFG is particularly engaged only when processing demands are high, for example, as in noncanonical sentences (Hickok, 2000; Love et al., 2006; Michael et al., 2001). Thus, the sentence types and/or production tasks included in published studies were not sufficiently structurally demanding to consistently engage this region. While some of the included studies involved unconstrained production, which could have included complex structures, only three studies that met criteria for inclusion examined noncanonical structures specifically, and all were passive sentences; none examined the more complex object‐relative clause or object cleft structures. Of the three studies, only one reported a noncanonical versus canonical contrast, for passives versus actives (Matchin & Hickok, 2016), with only a single activation peak in left postcentral gyrus.
Low power may have also contributed to the differences in the temporal lobes, as the clusters were much smaller and less widespread in the left hemisphere for production. It is also possible that differences in the selection of contrasts contributed to these domain differences. Given the small number of available production studies, we were not able to choose only contrasts against similar baselines, as we were able to do for the sentence comprehension analysis. As well, there was variation in terms of the sentence structures that were studied and whether the structures to be produced were constrained or not.
Nevertheless, our findings are broadly consistent with those from previous comprehension and production meta‐analyses, perhaps despite methodological differences. For comprehension, our choice of baseline was similar to that of Adank (2012a), though we included studies of sentence reading and did not include studies of single word comprehension. We used a different meta‐analysis method than Vigneau et al. (2006, 2011), with different study selection criteria as well. Perhaps the largest difference between our results and both of those prior comprehension analyses is that they both report activation in right inferior frontal regions (insula and inferior frontal gyrus), which we did not find here. This region has been implicated in sentence level prosody (Friederici, 2011). However, the right hemisphere may be recruited for prosodic processing when it is isolated from segmental information and/or tasks are difficult (Friederici, 2011; Plante, Creusere, & Sabin, 2002). Vigneau et al. (2011) suggests this region reflects recruitment of attentional or working memory processes. We return to this point below when we discuss the results of our noncanonical analysis.
4.3. Auditory versus visual comprehension
Several regions involved for auditory versus visual sentence comprehension were overlapping, though modality differences were present. With respect to their similarities across modalities, the conjunction map from the contrast analysis revealed five overlapping clusters in the left hemisphere: four clusters in the frontal lobe and one cluster in posterior temporal lobe and inferior parietal lobe. The overlap across modalities for clusters in IFG and posterior temporal/inferior parietal cortex are consistent with the hypothesis that these regions are modality independent, transmodal processing hubs (Mesulam, 1998). The similar and overlapping clusters in the SMA may reflect similar processing difficulty or attentional demands in the two modalities. However, Adank (2012b) suggested that activity in the precentral gyrus and SMA reflect difficulties associated with auditory comprehension related to perceiving speech in noise. This argument does not hold for reading comprehension. One possibility may be that activity in these regions reflects difficult processing, but for different reasons across modalities: perceiving speech in noisy conditions in the auditory modality, and increased demands for controlled processing in the visual modality (as we argued in the introduction).
With respect to modality differences, the overall volume of significant clusters for auditory studies was twice that of visual studies, consistent with a prior finding (Buchweitz et al., 2009). This difference was most pronounced in the temporal lobes, where the auditory clusters were much larger overall than the visual clusters, bilaterally. The clusters for auditory comprehension were also spatially more anterior than the visual clusters, particularly in the right hemisphere, where the clusters did not overlap between modalities, consistent with prior findings (Buchweitz et al., 2009; Constable et al., 2004; Michael et al., 2001). Widespread processing in the temporal lobes, particularly in the MTG and STG and nearby regions, may be required to decode the continuous auditory signal for word recognition (Hickok & Poeppel, 2007; Mesgarani, Cheung, Johnson, & Chang, 2014). This difference may, therefore, reflect pre‐lexical differences between auditory comprehension and reading. Note that the contrasts we chose to include in this meta‐analysis were intended only to remove low‐level processing related to sensory perception of the stimuli, and thus should not be expected to remove other aspects of pre‐lexical processing.
Another difference is that a cluster was found in left posterior temporal fusiform cortex for auditory comprehension, whereas no similar cluster was found for reading. Given the purported role of this region in lexical–semantic processing, it is surprising that it did not emerge in our analysis of reading comprehension. However, close inspection of the data from the overall comprehension analysis revealed that four reading studies did contribute to the temporal occipital fusiform cluster found in that analysis (cluster 5 in Table 5). Thus, it is possible that the lack of a significant cluster in temporal‐fusiform cortex in the analysis of reading by itself reflects low power to detect a small effect (our analysis of reading comprehension did only include 14 studies), rather than a real lack of involvement of that region in reading comprehension.
Finally, we note that the distribution of the clusters in left inferior frontal regions does not appear consistent with previous claims that auditory activation in left inferior frontal cortex is anterior and inferior to activation in response to visual stimuli, with auditory activity in pars triangularis and visual in pars opercularis (Michael et al., 2001). In our results, auditory and visual stimuli elicited overlapping clusters with peak in pars orbitalis, pars triangularis, and pars opercularis (Table 8, Figure 3). Thus, differences between modalities appear to be driven primarily by differences in the temporal lobes, which may reflect differences in pre‐lexical processing.
4.4. Noncanonical sentence comprehension
The analysis of noncanonical sentence comprehension revealed clusters in left IFG and MFG, as well as a cluster in left SFG extending into left and right paracingulate gyrus. An additional cluster peaked in the right insular cortex. Two more clusters were found in left posterior perisylvian areas, one in left posterior MTG, and one in left AG. Activation in these regions is consistent with their purported roles in sentence comprehension (see above), but does not differentiate between models of the functional role of the regions—for example, whether left inferior frontal cortex is principally responsible for processing complex syntax or whether it is engaged due to the higher processing load of these complex sentences.
The involvement of right insular cortex in our results suggests that this region does play a role in sentence comprehension, despite the results from our analysis of sentence comprehension against a low‐level baseline, where no activation in this region was found. It may be that this region is engaged only in limited circumstances, as we argued above for left inferior frontal cortical regions (including the left insula), though whether these noncanonical sentences elicit activity in this region because of difficult prosody (as has been suggested for right inferior frontal cortex; Friederici, 2011) or for other reasons remains to be determined. As a second point, the engagement of this region in the comprehension of complex sentences in healthy individuals is consistent with the hypothesis that this region (and nearby and/or functionally connected regions) might be capable of contributing to the recovery of comprehension of complex sentences in individuals with agrammatic aphasia, though precisely how and to what extent is still an active area of investigation (Gainotti, 2015; Lukic et al., 2017).
While neither of the prior meta‐analyses of complex syntax found activation in right insular cortex, our results from the left hemisphere were largely similar to those from previous meta‐analyses of complex sentences. Nevertheless, there were differences. In comparison to Rodd et al. (2015), who defined complexity more broadly than we did (see Introduction), we found an activation peak in pars orbitalis of left IFG and two peaks within a cluster in left frontal pole that they did not find. In the parietal lobe, we found activation in left AG, whereas their left parietal activation was in inferior parietal lobule and SMG. They found activation in left ITG, while we did not. These differences are unlikely to reflect a discrepancy in power, particularly for regions where they found activation and we did not, as they had fewer studies (n = 28). More likely, the differences in activated regions reflect differences in the definition of complexity across studies, though precisely why the differences manifest in the regions they do is not yet clear.
Our results were also similar to those of Meyer and Friederici (2015) for the left hemisphere, except that they did not find activation peaks in left angular gyrus or anywhere near left superior frontal or paracingulate gyri, contra our results. Note that they restricted their analysis to the left hemisphere, precluding any finding of right hemisphere activation. The activation differences in the left hemisphere are unlikely to reflect differences in how complexity was defined, as Meyer and Friederici (2015) defined complexity in terms of noncanonical structures, similar to our definition. It is possible that the additional left hemisphere clusters we found could reflect increased power in our study, as we had 37 studies to their 20.
Meyer and Friederici (2015) also included many studies that we did not include, and these different choices related to inclusion criteria could also have contributed to the different results. Nine of the studies they included did not meet our inclusion criteria: one reported only data from individuals with neurodegenerative disease (Amici et al., 2007); three reported results from ROI analyses, not whole‐brain analyses (Ben‐Shachar, Hendler, Kahn, Ben‐Bashat, & Grodzinsky, 2003; Ben‐Shachar, Palti, & Grodzinsky, 2004; Michael et al., 2001); one reported coordinates only from an interaction, not from simple effects (Bornkessel et al., 2005); one included embedded clauses with canonical thematic order (Makuuchi, Bahlmann, Anwander, & Friederici, 2009); one reported combined data from passive (noncanonical) sentences and sentences with shifted prepositional phrases that did not involve verbal thematic role assignments (Dapretto & Bookheimer, 1999); one reported results from a parametric analysis between canonical, noncanonical, and ungrammatical sentences (Friederici, Fiebach, Schlesewsky, Bornkessel, & von Cramon, 2006); another (Obleser & Kotz, 2010) reported one set of results a parametric analysis (experiment 1), and one set of results with no task (experiment 2); a tenth study that we excluded (Newman, Ikuta, & Burns Jr, 2010) reported a subset of the same data from a study that we did include (Newman, Malaia, Seo, & Cheng, 2013).
5. CONCLUSIONS
In sum, our meta‐analyses revealed a bilateral fronto‐temporal network of brain regions involved in sentence comprehension and production. While the sets of implicated regions were similar in many respects across domains, particularly in the left hemisphere, the network for production was not found to extend to the right hemisphere, likely due to low power for the relatively smaller effects in that hemisphere. We also did not see activation in left inferior frontal cortex for production, suggesting that left inferior frontal cortex is engaged for production only when task or stimulus demands lead to particularly costly processing. Modality differences between auditory and visual sentence comprehension appear to be relatively minor, and clear differences were not seen in frontal cortical regions. However, it may be that regions implicated in difficult processing in both modalities (precentral gyrus, SMA) may be engaged for different reasons across modalities. Finally, our meta‐analysis examining the comprehension of complex syntax (narrowly defined), revealed active regions in both left and right frontal cortex. The involvement of the right hemisphere in the comprehension of these structures has potentially important implications for language treatment and recovery in individuals with agrammatic aphasia following left hemisphere brain damage.
Supporting information
Appendix S1 Data_Comprehension_and_production
Appendix S2 Data_Comprehension
Appendix S3 Data_Production
Appendix S4 Data_Auditory_Comprehension
Appendix S5 Data_Visual_Comprehension
Appendix S6 Data_noncanonical
ACKNOWLEDGMENTS
This work was supported by the National Institutes of Health‐NIDCD, Clinical Research Center Grant, P50DC012283 (PI: C. K. Thompson). The authors wish to thank Dr. Jennifer Mack, Dr. Elena Barbieri, Katharine Aveni, Devin St. John, Kathy Xie, Brianne Chiappetta, Sarah Chandler, and Lucia Gurrola for helpful comments and assistance with the literature review and data entry. The authors have no conflicts of interest to declare.
Walenski M, Europa E, Caplan D, Thompson CK. Neural networks for sentence comprehension and production: An ALE‐based meta‐analysis of neuroimaging studies. Hum Brain Mapp. 2019;40:2275–2304. 10.1002/hbm.24523
Funding information National Institutes of Health‐NIDCD, Clinical Research Center, Grant/Award Number: P50DC012283
REFERENCES
- Abutalebi, J. (2008). Neural processing of second language representation and control. Acta Psychologica, 128, 466–478. [DOI] [PubMed] [Google Scholar]
- Abutalebi, J. , & Green, D. W. (2007). Bilingual language production: The neurocognition of language representation and control. Journal of Neurolinguistics, 20, 242–275. [Google Scholar]
- Adank, P. (2012a). Design choices in imaging speech comprehension: An activation likelihood estimation (ALE) meta‐analysis. NeuroImage, 63, 1601–1613. [DOI] [PubMed] [Google Scholar]
- Adank, P. (2012b). The neural bases of difficult speech comprehension and speech production: Two activation likelihood estimation (ALE) meta‐analyses. Brain and Language, 122, 42–54. [DOI] [PubMed] [Google Scholar]
- Alain, C. , Du, Y. , Bernstein, L. J. , Barten, T. , & Banai, K. (2018). Listening under difficult conditions: An activation likelihood estimation meta‐analysis. Human Brain Mapping, 39, 2695–2709. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Amici, S. , Brambati, S. M. , Wilkins, D. P. , Ogar, J. , Dronkers, N. L. , Miller, B. L. , & Gorno‐Tempini, M. L. (2007). Anatomical correlates of sentence comprehension and verbal working memory in neurodegenerative disease. The Journal of Neuroscience, 27, 6282–6290. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Ardila, A. , Bernal, B. , & Rosselli, M. (2015). Language and visual perception associations: Meta‐analytic connectivity modeling of Brodmann area 37. Behavioural Neurology, 2015, 1–14. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Aron, A. R. (2007). The neural basis of inhibition in cognitive control. The Neuroscientist, 13, 214–218. [DOI] [PubMed] [Google Scholar]
- Bahlmann, J. , Rodriguez‐Fornells, A. , Rotte, M. , & Münte, T. F. (2007). An fMRI study of canonical and noncanonical word order in German. Human Brain Mapping, 28, 940–949. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Balconi, M. , & Pozzoli, U. (2005). Comprehending semantic and grammatical violations in Italian. N400 and P600 comparison with visual and auditory stimuli. Journal of Psycholinguistic Research, 34, 71–98. [DOI] [PubMed] [Google Scholar]
- Baldo, J. , & Dronkers, N. (2007). Neural correlates of arithmetic and language comprehension: A common substrate? Neuropsychologia, 45, 229–235. [DOI] [PubMed] [Google Scholar]
- Barbieri, E. , Mack, J.E. , Chiappetta, B. , Europa, E. , Thompson, C.K. (2018) Mechanisms of neural plasticity and recovery from sentence processing deficits in chronic stroke‐induced aphasia: an fMRI study.
- Bemis, D. K. , & Pylkkänen, L. (2011). Simple composition: A magnetoencephalography investigation into the comprehension of minimal linguistic phrases. The Journal of Neuroscience, 31, 2801–2814. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Bemis, D. K. , & Pylkkänen, L. (2012). Basic linguistic composition recruits the left anterior temporal lobe and left angular gyrus during both listening and reading. Cerebral Cortex, 23, 1859–1873. [DOI] [PubMed] [Google Scholar]
- Ben‐Shachar, M. , Hendler, T. , Kahn, I. , Ben‐Bashat, D. , & Grodzinsky, Y. (2003). The neural reality of syntactic transformations: Evidence from functional magnetic resonance imaging. Psychological Science, 14, 433–440. [DOI] [PubMed] [Google Scholar]
- Ben‐Shachar, M. , Palti, D. , & Grodzinsky, Y. (2004). Neural correlates of syntactic movement: Converging evidence from two fMRI experiments. NeuroImage, 21, 1320–1336. [DOI] [PubMed] [Google Scholar]
- Berthier, M. L. (1999). Transcortical aphasias. Hove, England: Psychology Press. [Google Scholar]
- Blank, S. C. , Scott, S. K. , Murphy, K. , Warburton, E. , & Wise, R. J. (2002). Speech production: Wernicke, Broca and beyond. Brain, 125, 1829–1838. [DOI] [PubMed] [Google Scholar]
- Bornkessel, I. , Zysset, S. , Friederici, A. D. , Yves von Cramon, D. , & Schlesewsky, M. (2005). Who did what to whom? The neural basis of argument hierarchies during language comprehension. NeuroImage, 26, 221–233. [DOI] [PubMed] [Google Scholar]
- Bornkessel‐Schlesewsky, I. , Grewe, T. , & Schlesewsky, M. (2012). Prominence vs. aboutness in sequencing: A functional distinction within the left inferior frontal gyrus. Brain and Language, 120, 96–107. [DOI] [PubMed] [Google Scholar]
- Bornkessel‐Schlesewsky, I. , & Schlesewsky, M. (2013). Reconciling time, space and function: A new dorsal–ventral stream model of sentence comprehension. Brain and Language, 125, 60–76. [DOI] [PubMed] [Google Scholar]
- Bornkessel‐Schlesewsky, I. , Schlesewsky, M. , & von Cramon, D. Y. (2009). Word order and Broca's region: Evidence for a supra‐syntactic perspective. Brain and Language, 111, 125–139. [DOI] [PubMed] [Google Scholar]
- Boulenger, V. , Hauk, O. , & Pulvermüller, F. (2009). Grasping ideas with the motor system: Semantic somatotopy in idiom comprehension. Cerebral Cortex, 19, 1905–1914. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Boylan, C. , Trueswell, J. C. , & Thompson‐Schill, S. L. (2015). Compositionality and the angular gyrus: A multi‐voxel similarity analysis of the semantic composition of nouns and verbs. Neuropsychologia, 78, 130–141. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Boylan, C. , Trueswell, J. C. , & Thompson‐Schill, S. L. (2017). Relational vs. attributive interpretation of nominal compounds differentially engages angular gyrus and anterior temporal lobe. Brain and Language, 169, 8–21. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Braun, A. R. , Guillemin, A. , Hosey, L. , & Varga, M. (2001). The neural organization of discourse: An H2 15O‐PET study of narrative production in English and American sign language. Brain, 124, 2028–2044. [DOI] [PubMed] [Google Scholar]
- Buchweitz, A. , Mason, R. A. , Tomich, L. M. B. , & Just, M. A. (2009). Brain activation for reading and listening comprehension: An fMRI study of modality effects and individual differences in language comprehension. Psychology & Neuroscience, 2, 111–123. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Caplan, D. (2009). Experimental design and interpretation of functional neuroimaging studies of cognitive processes. Human Brain Mapping, 30, 59–77. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Caplan, D. (2010). Task effects on BOLD signal correlates of implicit syntactic processing. Language & Cognitive Processes, 25, 866–901. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Caplan, D. , Alpert, N. , & Waters, G. (1999). PET studies of syntactic processing with auditory sentence presentation. NeuroImage, 9, 343–351. [DOI] [PubMed] [Google Scholar]
- Caplan, D. , Alpert, N. , Waters, G. , & Olivieri, A. (2000). Activation of Broca's area by syntactic processing under conditions of concurrent articulation. Human Brain Mapping, 9, 65–71. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Caplan, D. , Chen, E. , & Waters, G. (2008). Task‐dependent and task‐independent neurovascular responses to syntactic processing. Cortex, 44, 257–275. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Caplan, D. , Stanczak, L. , & Waters, G. (2008). Syntactic and thematic constraint effects on blood oxygenation level dependent signal correlates of comprehension of relative clauses. Journal of Cognitive Neuroscience, 20, 643–656. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Caplan, D. , & Waters, G. (2013). Memory mechanisms supporting syntactic comprehension. Psychonomic Bulletin and Review, 20, 243–268. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Caplan, D. , Waters, G. , & Alpert, N. (2003). Effects of age and speed of processing on rCBF correlates of syntactic processing in sentence comprehension. Human Brain Mapping, 19, 112–131. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Caramazza, A. , & Zurif, E. (1976). Dissociation of algorithmic and heuristic processes in language comprehension: Evidence from aphasia. Brain and Language, 3, 572–582. [DOI] [PubMed] [Google Scholar]
- Cho, Y. W. , Song, H.‐J. , Lee, J. J. , Lee, J. H. , Lee, H. J. , Yi, S. D. , … Chang, Y. (2012). Age‐related differences in the brain areas outside the classical language areas among adults using category decision task. Brain and Language, 120, 372–380. [DOI] [PubMed] [Google Scholar]
- Colman, K. S. F. , Koerts, J. , Stowe, L. A. , Leenders, K. L. , & Bastiaanse, R. (2011). Sentence comprehension and its association with executive functions in patients with Parkinson's disease. Parkinson's Disease, 2011, 15. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Constable, R. T. , Pugh, K. R. , Berroya, E. , Mencl, W. E. , Westerveld, M. , Ni, W. , & Shankweiler, D. (2004). Sentence complexity and input modality effects in sentence comprehension: An fMRI study. NeuroImage, 22, 11–21. [DOI] [PubMed] [Google Scholar]
- Cooke, A. , Zurif, E. B. , DeVita, C. , Alsop, D. , Koenig, P. , Detre, J. , … Grossman, M. (2001). Neural basis for sentence comprehension: Grammatical and short‐term memory components. Human Brain Mapping, 15, 80–94. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Corbetta, M. , & Shulman, G. L. (2002). Control of goal‐directed and stimulus‐driven attention in the brain. Nature Reviews Neuroscience, 3, 201–215. [DOI] [PubMed] [Google Scholar]
- Cotelli, M. , Borroni, B. , Manenti, R. , Ginex, V. , Calabria, M. , Moro, A. , … Padovani, A. (2007). Universal grammar in the frontotemporal dementia spectrum: Evidence of a selective disorder in the corticobasal degeneration syndrome. Neuropsychologia, 45, 3015–3023. [DOI] [PubMed] [Google Scholar]
- Crinion, J. , & Price, C. J. (2005). Right anterior superior temporal activation predicts auditory sentence comprehension following aphasic stroke. Brain, 128, 2858–2871. [DOI] [PubMed] [Google Scholar]
- Crinion, J. T. , Lambon‐Ralph, M. A. , Warburton, E. A. , Howard, D. , & Wise, R. J. (2003). Temporal lobe regions engaged during normal speech comprehension. Brain, 126, 1193–1201. [DOI] [PubMed] [Google Scholar]
- Dapretto, M. , & Bookheimer, S. Y. (1999). Form and content: Dissociating syntax and semantics in sentence comprehension. Neuron, 24, 427–432. [DOI] [PubMed] [Google Scholar]
- Davis, M. H. , Coleman, M. R. , Absalom, A. R. , Rodd, J. M. , Johnsrude, I. S. , Matta, B. F. , … Menon, D. K. (2007). Dissociating speech perception and comprehension at reduced levels of awareness. Proceedings of the National Academy of Sciences, 104, 16032–16037. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Davis, M. H. , Johnsrude, I. S. , Hervais‐Adelman, A. , Taylor, K. , & McGettigan, C. (2005). Lexical information drives perceptual learning of distorted speech: Evidence from the comprehension of noise‐Vocoded sentences. Journal of Experimental Psychology. General, 134, 222–241. [DOI] [PubMed] [Google Scholar]
- DeWitt, I. , & Rauschecker, J. P. (2012). Phoneme and word recognition in the auditory ventral stream. Proceedings of the National Academy of Sciences, 109, E505–E514. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Diaz, M. T. , Barrett, K. T. , & Hogstrom, L. J. (2011). The influence of sentence novelty and figurativeness on brain activity. Neuropsychologia, 49, 320–330. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Dosenbach, N. U. , Visscher, K. M. , Palmer, E. D. , Miezin, F. M. , Wenger, K. K. , Kang, H. C. , … Petersen, S. E. (2006). A core system for the implementation of task sets. Neuron, 50, 799–812. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Dronkers, N. F. , Wilkins, D. P. , Van Valin, R. D., Jr. , Redfern, B. B. , & Jaeger, J. J. (2004). Lesion analysis of the brain areas involved in language comprehension. Cognition, 92, 145–177. [DOI] [PubMed] [Google Scholar]
- Duncan, J. , & Owen, A. M. (2000). Common regions of the human frontal lobe recruited by diverse cognitive demands. Trends in Neurosciences, 23, 475–483. [DOI] [PubMed] [Google Scholar]
- Eickhoff, S. B. , Bzdok, D. , Laird, A. R. , Kurth, F. , & Fox, P. T. (2012). Activation likelihood estimation meta‐analysis revisited. NeuroImage, 59, 2349–2361. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Eickhoff, S. B. , Laird, A. R. , Grefkes, C. , Wang, L. E. , Zilles, K. , & Fox, P. T. (2009). Coordinate‐based activation likelihood estimation meta‐analysis of neuroimaging data: A random‐effects approach based on empirical estimates of spatial uncertainty. Human Brain Mapping, 30, 2907–2926. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Eickhoff, S. B. , Nichols, T. E. , Laird, A. R. , Hoffstaedter, F. , Amunts, K. , Fox, P. T. , … Eickhoff, C. R. (2016). Behavior, sensitivity, and power of activation likelihood estimation characterized by massive empirical simulation. NeuroImage, 137, 70–85. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Erb, J. , Henry, M. J. , Eisner, F. , & Obleser, J. (2013). The brain dynamics of rapid perceptual adaptation to adverse listening conditions. The Journal of Neuroscience, 33, 10688–10697. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Fedorenko, E. , Duncan, J. , & Kanwisher, N. (2013). Broad domain generality in focal regions of frontal and parietal cortex. Proceedings of the National Academy of Sciences of the United States of America, 110, 16616–16621. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Fedorenko, E. , Nieto‐Castanon, A. , & Kanwisher, N. (2012). Lexical and syntactic representations in the brain: An fMRI investigation with multi‐voxel pattern analyses. Neuropsychologia, 50, 499–513. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Ferreira, F. , & Anes, M. (1994). Why study spoken language processing? In Gernsbacher M. (Ed.), Handbook of psycholinguistics. San Diego, CA: Academic Press. [Google Scholar]
- Ferstl, E. C. , & von Cramon, D. Y. (2001). The role of coherence and cohesion in text comprehension: An event‐related fMRI study. Cognitive Brain Research, 11, 325–340. [DOI] [PubMed] [Google Scholar]
- Friederici, A. D. (2002). Towards a neural basis of auditory sentence processing. Trends in Cognitive Sciences, 6, 78–84. [DOI] [PubMed] [Google Scholar]
- Friederici, A. D. (2011). The brain basis of language processing: From structure to function. Physiological Reviews, 91, 1357–1392. [DOI] [PubMed] [Google Scholar]
- Friederici, A. D. (2012). The cortical language circuit: From auditory perception to sentence comprehension. Trends in Cognitive Sciences, 16, 262–268. [DOI] [PubMed] [Google Scholar]
- Friederici, A. D. , Fiebach, C. J. , Schlesewsky, M. , Bornkessel, I. D. , & von Cramon, Y. (2006). Processing linguistic complexity and grammaticality in the left frontal cortex. Cerebral Cortex, 16, 1709–1717. [DOI] [PubMed] [Google Scholar]
- Friederici, A. D. , Opitz, B. , & von Cramon, D. Y. (2000). Segregating semantic and syntactic aspects of processing in the human brain: An fMRI investigation of different word types. Cerebral Cortex, 10, 698–705. [DOI] [PubMed] [Google Scholar]
- Gainotti, G. (2015). Contrasting opinions on the role of the right hemisphere in the recovery of language. A critical survey. Aphasiology, 29, 1020–1037. [Google Scholar]
- Geranmayeh, F. , Brownsett, S. L. E. , Leech, R. , Beckmann, C. F. , Woodhead, Z. , & Wise, R. J. S. (2012). The contribution of the inferior parietal cortex to spoken language production. Brain and Language, 121, 47–57. [DOI] [PubMed] [Google Scholar]
- Geranmayeh, F. , Wise, R. J. S. , Mehta, A. , & Leech, R. (2014). Overlapping networks engaged during spoken language production and its cognitive control. The Journal of Neuroscience, 34, 8728–8740. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Gorno‐Tempini, M. L. , Dronkers, N. F. , Rankin, K. P. , Ogar, J. M. , Phengrasamy, L. , Rosen, H. J. , … Miller, B. L. (2004). Cognition and anatomy in three variants of primary progressive aphasia. Annals of Neurology, 55, 335–346. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Grande, M. , Meffert, E. , Schoenberger, E. , Jung, S. , Frauenrath, T. , Huber, W. , … Heim, S. (2012). From a concept to a word in a syntactically complete sentence: An fMRI study on spontaneous language production in an overt picture description task. NeuroImage, 61, 702–714. [DOI] [PubMed] [Google Scholar]
- Grewe, T. , Bornkessel, I. , Zysset, S. , Wiese, R. , von Cramon, D. Y. , & Schlesewsky, M. (2005). The emergence of the unmarked: A new perspective on the language‐specific function of Broca's area. Human Brain Mapping, 26, 178–190. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Grodzinsky, Y. (2000). The neurology of syntax: Language use without Broca's area. The Behavioral and Brain Sciences, 23, 1–21 discussion 21‐71. [DOI] [PubMed] [Google Scholar]
- Grodzinsky Y., & Amunts K. (Eds.). (2006). Broca's region. New York, NY: Oxford University Press. [Google Scholar]
- Grossman, M. , Cooke, A. , De Vita, C. , Alsop, D. , Detre, J. , Chen, W. , & Gee, J. (2002). Age‐related changes in working memory during sentence comprehension: An fMRI study. NeuroImage, 15, 302–317. [DOI] [PubMed] [Google Scholar]
- Grossman, M. , Rhee, J. , & Moore, P. (2005). Sentence processing in frontotemporal dementia. Cortex, 41, 764–777. [DOI] [PubMed] [Google Scholar]
- Hagoort, P. (2005). On Broca, brain, and binding: A new framework. Trends in Cognitive Sciences, 9, 416–423. [DOI] [PubMed] [Google Scholar]
- Hagoort, P. , & Brown, C. (2000). ERP effects of listening to speech compared to reading: The P600/SPS to syntactic violations in spoken sentences and rapid serial visual presentation. Neuropsychologia, 38, 1531–1549. [DOI] [PubMed] [Google Scholar]
- Hakonen, M. , May, P. J. C. , Jääskeläinen, I. P. , Jokinen, E. , Sams, M. , & Tiitinen, H. (2017). Predictive processing increases intelligibility of acoustically distorted speech: Behavioral and neural correlates. Brain and Behavior: A Cognitive Neuroscience Perspective, 7, 1–15. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Haller, S. , Radue, E. , Erb, M. , Grodd, W. , & Kircher, T. (2005). Overt sentence production in event‐related fMRI. Neuropsychologia, 43, 807–814. [DOI] [PubMed] [Google Scholar]
- Henry, R. G. , Berman, J. I. , Nagarajan, S. S. , Mukherjee, P. , & Berger, M. S. (2004). Subcortical pathways serving cortical language sites: Initial experience with diffusion tensor imaging fiber tracking combined with intraoperative language mapping. NeuroImage, 21, 616–622. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Hickok, G. (2000). The left frontal convolution plays no special role in syntactic comprehension. The Behavioral and Brain Sciences, 23, 35–36. [Google Scholar]
- Hickok, G. , & Poeppel, D. (2000). Towards a functional neuroanatomy of speech perception. Trends in Cognitive Sciences, 4, 131–138. [DOI] [PubMed] [Google Scholar]
- Hickok, G. , & Poeppel, D. (2004). Dorsal and ventral streams: A framework for understanding aspects of the functional anatomy of language. Cognition. Special issue: Towards a new functions anatomy of. Language, 92, 67–99. [DOI] [PubMed] [Google Scholar]
- Hickok, G. , & Poeppel, D. (2007). The cortical organization of speech processing. Nature Reviews Neuroscience, 8, 393–402. [DOI] [PubMed] [Google Scholar]
- Hirotani, M. , Makuuchi, M. , Ruschemeyer, S. , & Friederici, A. D. (2011). Who was the agent? The neural correlates of reanalysis processes during sentence comprehension. Human Brain Mapping, 32, 1775–1787. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Hullett, P. W. , Hamilton, L. S. , Mesgarani, N. , Schreiner, C. E. , & Chang, E. F. (2016). Human superior temporal Gyrus Organization of Spectrotemporal Modulation Tuning Derived from speech stimuli. The Journal of Neuroscience, 36, 2014–2026. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Humphries, C. , Binder, J. R. , Medler, D. A. , & Liebenthal, E. (2006). Syntactic and semantic modulation of neural activity during auditory sentence comprehension. Journal of Cognitive Neuroscience, 18, 665–679. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Indefrey, P. (2011). The spatial and temporal signatures of word production components: A critical update. Frontiers in Psychology, 2, 255. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Indefrey, P. , Hellwig, F. , Herzog, H. , Seitz, R. J. , & Hagoort, P. (2004). Neural responses to the production and comprehension of syntax in identical utterances. Brain and Language, 89, 312–319. [DOI] [PubMed] [Google Scholar]
- Indefrey, P. , & Levelt, W. J. (2004). The spatial and temporal signatures of word production components. Cognition, 92, 101–144. [DOI] [PubMed] [Google Scholar]
- Indefrey, P. , & Levelt, W. J. M. (2000). The neural correlates of language production In Gazzaniga M. S. (Ed.), The new cognitive neurosciences (pp. 845–866). Cambridge, MA: MIT Press. [Google Scholar]
- Ischebeck, A. , Friederici, A. , & Alter, K. (2008). Processing prosodic boundaries in natural and hummed speech: An fMRI study. Cerebral Cortex, 18, 541–552. [DOI] [PubMed] [Google Scholar]
- Jednoróg, K. , Bola, Ł. , Mostowski, P. , Szwed, M. , Boguszewski, P. M. , Marchewka, A. , & Rutkowski, P. (2015). Three‐dimensional grammar in the brain: Dissociating the neural correlates of natural sign language and manually coded spoken language. Neuropsychologia, 71, 191–200. [DOI] [PubMed] [Google Scholar]
- Just, M. A. , Carpenter, P. A. , Keller, T. A. , Eddy, W. F. , & Thulborn, K. R. (1996). Brain activation modulated by sentence comprehension. Science, 274, 114–116. [DOI] [PubMed] [Google Scholar]
- Kalénine, S. , Peyrin, C. , Pichat, C. , Segebarth, C. , Bonthoux, F. , & Baciu, M. (2009). The sensory‐motor specificity of taxonomic and thematic conceptual relations: A behavioral and fMRI study. NeuroImage, 44, 1152–1162. [DOI] [PubMed] [Google Scholar]
- Kang, A. , Constable, R. , Gore, J. , & Avrutin, S. (1999). An event‐related fMRI study of implicit phrase‐level syntactic and semantic processing. NeuroImage, 10, 555–561. [DOI] [PubMed] [Google Scholar]
- Kang, E. , Lee, D. S. , Kang, H. , Hwang, C. H. , Oh, S.‐H. , Kim, C.‐S. , … Lee, M. C. (2006). The neural correlates of cross‐modal interaction in speech perception during a semantic decision task on sentences: A PET study. NeuroImage, 32, 423–431. [DOI] [PubMed] [Google Scholar]
- Kansaku, K. , Yamaura, A. , & Kitazawa, S. (2000). Sex differences in lateralization revealed in the posterior language areas. Cerebral Cortex, 10, 866–872. [DOI] [PubMed] [Google Scholar]
- Kemeny, S. , Ye, F. Q. , Birn, R. , & Braun, A. R. (2005). Comparison of continuous overt speech fMRI using BOLD and arterial spin labeling. Human Brain Mapping, 24, 173–183. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Kempen, G. (2000). Could grammatical encoding and grammatical decoding be subserved by the same processing module? The Behavioral and Brain Sciences, 23, 38–39. [Google Scholar]
- Kho, K. H. , Indefrey, P. , Hagoort, P. , van Veelen, C. W. , van Rijen, P. C. , & Ramsey, N. F. (2008). Unimpaired sentence comprehension after anterior temporal cortex resection. Neuropsychologia, 46, 1170–1178. [DOI] [PubMed] [Google Scholar]
- Kim, J. , Koizumi, M. , Ikuta, N. , Fukumitsu, Y. , Kimura, N. , Iwata, K. , … Kawashima, R. (2009). Scrambling effects on the processing of Japanese sentences: An fMRI study. Journal of Neurolinguistics, 22, 151–166. [Google Scholar]
- Kinno, R. , Kawamura, M. , Shioda, S. , & Sakai, K. L. (2008). Neural correlates of noncanonical syntactic processing revealed by picture‐sentence matching task. Human Brain Mapping, 29, 1015–1027. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Koizumi, M. , & Kim, J. (2016). Greater left inferior frontal activation for SVO than VOS during sentence comprehension in Kaqchikel. Frontiers in Psychology, 7, 1–9. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Kovelman, I. , Baker, S. A. , & Petitto, L.‐A. (2008). Bilingual and monolingual brains compared: A functional magnetic resonance imaging investigation of syntactic processing and a possible “neural signature” of bilingualism. Journal of Cognitive Neuroscience, 20, 153–169. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Kristensen, L. B. , Wang, L. , Petersson, K. M. , & Hagoort, P. (2013). The interface between language and attention: Prosodic focus marking recruits a general attention network in spoken language comprehension. Cerebral Cortex, 23, 1836–1848. [DOI] [PubMed] [Google Scholar]
- Kuperberg, G. R. , McGuire, P. K. , Bullmore, E. T. , Brammer, M. J. , Rabe‐Hesketh, S. , Wright, I. C. , … David, A. S. (2000). Common and distinct neural substrates for pragmatic, semantic, and syntactic processing of spoken sentences: An fMRI study. Journal of Cognitive Neuroscience, 12, 321–341. [DOI] [PubMed] [Google Scholar]
- Kutas, M. , & Federmeier, K. D. (2011). Thirty years and counting: Finding meaning in the N400 component of the event‐related brain potential (ERP). Annual Review of Psychology, 62, 621–647. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Kutas, M. , & Van Petten, C. K. (1994). Psycholinguistics electrified: Event‐related potential investigations In Gernsbacher M. A. (Ed.), Handbook of psycholinguistics. San Diego, CA: Academic Press. [Google Scholar]
- Lambon Ralph, M. A. , Sage, K. , Jones, R. W. , & Mayberry, E. J. (2010). Coherent concepts are computed in the anterior temporal lobes. Proceedings of the National Academy of Sciences of the United States of America, 107, 2717–2722. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Lancaster, J. L. , Tordesillas‐Gutiérrez, D. , Martinez, M. , Salinas, F. , Evans, A. , Zilles, K. , … Fox, P. T. (2007). Bias between MNI and Talairach coordinates analyzed using the ICBM‐152 brain template. Human Brain Mapping, 28, 1194–1205. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Lau, E. , Stroud, C. , Plesch, S. , & Phillips, C. (2006). The role of structural prediction in rapid syntactic analysis. Brain and Language, 98, 74–88. [DOI] [PubMed] [Google Scholar]
- Lau, E. F. , Phillips, C. , & Poeppel, D. (2008). A cortical network for semantics: (de)constructing the N400. Nature Reviews Neuroscience, 9, 920–933. [DOI] [PubMed] [Google Scholar]
- Lewis, G. A. , Poeppel, D. , & Murphy, G. L. (2015). The neural bases of taxonomic and thematic conceptual relations: An MEG study. Neuropsychologia, 68, 176–189. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Lillywhite, L. M. , Saling, M. M. , Demutska, A. , Masterton, R. , Farquharson, S. , & Jackson, G. D. (2010). The neural architecture of discourse compression. Neuropsychologia, 48, 873–879. [DOI] [PubMed] [Google Scholar]
- Love, T. , Haist, F. , Nicol, J. , & Swinney, D. (2006). A functional neuroimaging investigation of the roles of structural complexity and task‐demand during auditory sentence processing. Cortex, 42, 577–590. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Love, T. , Swinney, D. , Walenski, M. , & Zurif, E. (2008). How left inferior frontal cortex participates in syntactic processing: Evidence from aphasia. Brain and Language, 107, 203–219. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Lück, M. , Hahne, A. , & Clahsen, H. (2006). Brain potentials to morphologically complex words during listening. Brain Research, 1077, 144–152. [DOI] [PubMed] [Google Scholar]
- Lukic, S. , Barbieri, E. , Wang, X. , Caplan, D. , Kiran, S. , Rapp, B. , … Thompson, C. K. (2017). Right hemisphere Grey matter volume and language functions in stroke aphasia. Neural Plasticity, 2017, 1–14. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Mack, J. E. , Ji, W. , & Thompson, C. K. (2013). Effects of verb meaning on lexical integration in agrammatic aphasia: Evidence from eyetracking. Journal of Neurolinguistics, 26, 619–636. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Mack, J. E. , Meltzer‐Asscher, A. , Barbieri, E. , & Thompson, C. K. (2013). Neural correlates of processing passive sentences. Brain Sciences, 3, 1198–1214. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Magnusdottir, S. , Fillmore, P. , den Ouden, D. B. , Hjaltason, H. , Rorden, C. , Kjartasson, O. , … Fridriksson, J. (2013). Damage to left anterior temporal cortex predicts impairment of complex syntactic processing: A lesion‐symptom mapping study. Human Brain Mapping, 34, 2715–2723. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Makuuchi, M. , Bahlmann, J. , Anwander, A. , & Friederici, A. D. (2009). Segregating the core computational faculty of human language from working memory. PNAS, 106, 8362–8367. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Marques, J. F. , Canessa, N. , & Cappa, S. (2009). Neural differences in the processing of true and false sentences: Insights into the nature of ‘truth’ in language comprehension. Cortex, 45, 759–768. [DOI] [PubMed] [Google Scholar]
- Martin, A. , Schurz, M. , Kronbichler, M. , & Richlan, F. (2015). Reading in the brain of children and adults: A meta‐analysis of 40 functional magnetic resonance imaging studies. Human Brain Mapping, 36, 1963–1981. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Matchin, W. , & Hickok, G. (2016). Syntactic perturbation’ during production activates the right IFG, but not Broca's area or the ATL. Frontiers in Psychology, 7, 241. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Meltzer, J. A. , McArdle, J. J. , Schafer, R. J. , & Braun, A. R. (2010). Neural aspects of sentence comprehension: Syntactic complexity, reversibility, and reanalysis. Cerebral Cortex, 20, 1853–1864. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Meltzer‐Asscher, A. , Mack, J. E. , Barbieri, E. , & Thompson, C. K. (2015). How the brain processes different dimensions of argument structure complexity: Evidence from fMRI. Brain and Language, 142, 65–75. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Meltzer‐Asscher, A. , Schuchard, J. , Den Ouden, D. B. , & Thompson, C. K. (2012). The neural substrates of complex argument structure representations: Processing "alternating transitivity" verbs. Language & Cognitive Processes, 28, 1154–1168. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Méndez Orellana, C. P. , Sandt‐Koenderman, M. E. , Saliasi, E. , Meulen, I. , Klip, S. , Lugt, A. , & Smits, M. (2014). Insight into the neurophysiological processes of melodically intoned language with functional MRI. Brain and Behavior: A Cognitive Neuroscience Perspective, 4, 615–625. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Menenti, L. , Gierhan, S. M. , Segaert, K. , & Hagoort, P. (2011). Shared language: Overlap and segregation of the neuronal infrastructure for speaking and listening revealed by functional MRI. Psychological Science, 22, 1173–1182. [DOI] [PubMed] [Google Scholar]
- Menenti, L. , Petersson, K. M. , & Hagoort, P. (2012). From reference to sense: How the brain encodes meaning for speaking. Frontiers in Psychology, 2. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Menenti, L. , Segaert, K. , & Hagoort, P. (2012). The neuronal infrastructure of speaking. Brain and Language, 122, 71–80. [DOI] [PubMed] [Google Scholar]
- Mesgarani, N. , Cheung, C. , Johnson, K. , & Chang, E. F. (2014). Phonetic feature encoding in human superior temporal Gyrus. Science, 343, 1006–1010. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Mesulam, M. (1998). From sensation to cognition. Brain, 121, 1013–1052. [DOI] [PubMed] [Google Scholar]
- Mesulam, M.‐M. , Rader, B. , Sridhar, J. , Nelson, M. , Hyun, J. , Rademaker, A. , Geula, C. , … Rogalski, E. (2019). Word comprehension in temporal cortex and Wernicke area: A PPA perspective. Neurology, 92(3), e224–e233; DOI: 10.1212/WNL.0000000000006788. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Mesulam, M. M. , Thompson, C. K. , Weintraub, S. , & Rogalski, E. J. (2015). The Wernicke conundrum and the anatomy of language comprehension in primary progressive aphasia. Brain, 138, 2423–2437. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Mesulam, M. M. , Wieneke, C. , Thompson, C. , Rogalski, E. , & Weintraub, S. (2012). Quantitative classification of primary progressive aphasia at early and mild impairment stages. Brain, 135, 1537–1553. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Meyer, L. , & Friederici, A. D. (2015). Neural systems underlying the processing of complex sentences In Hickok G. & Small S. (Eds.), Neurobiology of language (pp. 597–606). Amsterdam (The Netherlands) and Boston (Massachusetts): Academic Press. [Google Scholar]
- Meyer, M. , Alter, K. , Friederici, A. D. , Lohmann, G. , & von Cramon, D. Y. (2002). FMRI reveals brain regions mediating slow prosodic modulations in spoken sentences. Human Brain Mapping, 17, 73–88. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Meyer, M. , Obleser, J. , Anwander, A. , & Friederici, A. D. (2012). Linking ordering in Broca's area to storage in left temporo‐parietal regions: The case of sentence processing. NeuroImage, 62, 1987–1998. [DOI] [PubMed] [Google Scholar]
- Meyer, M. , Steinhauer, K. , Alter, K. , Friederici, A. D. , & von Cramon, D. Y. (2004). Brain activity varies with modulation of dynamic pitch variance in sentence melody. Brain and Language, 89, 277–289. [DOI] [PubMed] [Google Scholar]
- Michael, E. B. , Keller, T. A. , Carpenter, P. A. , & Just, M. A. (2001). fMRI investigation of sentence comprehension by eye and by ear: Modality fingerprints on cognitive processes. Human Brain Mapping, 13, 239–252. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Nelson, M. J. , El Karoui, I. , Giber, K. , Yang, X. , Cohen, L. , Koopman, H. , … Dehaene, S. (2017). Neurophysiological dynamics of phrase‐structure building during sentence processing. Proceedings of the National Academy of Sciences of the United States of America, 114, E3669–E3678. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Newman, S. D. , Ikuta, T. , & Burns, T., Jr. (2010). The effect of semantic relatedness on syntactic analysis: An fMRI study. Brain and Language, 113, 51–58. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Newman, S. D. , Just, M. A. , Keller, T. A. , Roth, J. , & Carpenter, P. A. (2003). Differential effects of syntactic and semantic processing on the subregions of Broca's area. Cognitive Brain Research, 16, 297–307. [DOI] [PubMed] [Google Scholar]
- Newman, S. D. , Malaia, E. , Seo, R. , & Cheng, H. (2013). The effect of individual differences in working memory capacity on sentence comprehension: An fMRI study. Brain Topography, 26, 458–467. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Ni, W. , Constable, R. T. , Menci, W. E. , Pugh, K. R. , Fulbright, R. K. , Shaywitz, S. E. , … Shankweiler, D. (2000). An event‐related neuroimaging study distinguishing form and content in sentence processing. Journal of Cognitive Neuroscience, 12, 120–133. [DOI] [PubMed] [Google Scholar]
- Nichols, T. , & Hayasaka, S. (2003). Controlling the familywise error rate in functional neuroimaging: A comparative review. Statistical Methods in Medical Research, 12, 419–446. [DOI] [PubMed] [Google Scholar]
- Nichols, T. E. (2012). Multiple testing corrections, nonparametric methods, and random field theory. NeuroImage, 62, 811–815. [DOI] [PubMed] [Google Scholar]
- Nicol, J. , Swinney, D. , Love, T. , & Hald, L. (2006). The on‐line study of sentence comprehension: An examination of dual task paradigms. Journal of Psycholinguistic Research, 35, 215–231. [DOI] [PubMed] [Google Scholar]
- Noppeney, U. , & Price, C. J. (2004). An fMRI study of syntactic adaptation. Journal of Cognitive Neuroscience, 16, 702–713. [DOI] [PubMed] [Google Scholar]
- Novais‐Santos, S. , Gee, J. , Shah, M. , Troiani, V. , Work, M. , & Grossman, M. (2007). Resolving sentence ambiguity with planning and working memory resources: Evidence from fMRI. NeuroImage, 37, 361–378. [DOI] [PubMed] [Google Scholar]
- Obleser, J. , & Kotz, S. A. (2010). Expectancy constraints in degraded speech modulate the language comprehension network. Cerebral Cortex, 20, 633–640. [DOI] [PubMed] [Google Scholar]
- Oh, A. , Duerden, E. G. , & Pang, E. W. (2014). The role of the insula in speech and language processing. Brain and Language, 135, 96–103. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Okada, K. , Rong, F. , Venezia, J. , Matchin, W. , Hsieh, I.‐H. , Saberi, K. , … Hickok, G. (2010). Hierarchical Organization of Human Auditory Cortex: Evidence from acoustic invariance in the response to intelligible speech. Cerebral Cortex, 20, 2486–2495. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Osaka, M. , Komori, M. , Morishita, M. , & Osaka, N. (2007). Neural bases of focusing attention in working memory: An f MRI study based on group differences. Cognitive, Affective, & Behavioral Neuroscience, 7, 130–139. [DOI] [PubMed] [Google Scholar]
- Patterson, K. , Nestor, P. J. , & Rogers, T. T. (2007). Where do you know what you know? The representation of semantic knowledge in the human brain. Nature Reviews. Neuroscience, 8, 976–987. [DOI] [PubMed] [Google Scholar]
- Peelle, J. E. , Troiani, V. , Wingfield, A. , & Grossman, M. (2010). Neural processing during older Adults' comprehension of spoken sentences: Age differences in resource allocation and connectivity. Cerebral Cortex, 20, 773–782. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Petrides, M. (2016). The ventrolateral frontal region In Hickok G. & Small S. L. (Eds.), The neurobiology of language. New York, NY: Academic Press. [Google Scholar]
- Pickering, M. J. , & Garrod, S. (2013). An integrated theory of language production and comprehension. The Behavioral and Brain Sciences, 36, 329–347. [DOI] [PubMed] [Google Scholar]
- Plante, E. , Creusere, M. , & Sabin, C. (2002). Dissociating sengential prosody from setence processing: Activation interacts with task demands. NeuroImage, 17, 401–410. [DOI] [PubMed] [Google Scholar]
- Prat, C. S. , & Just, M. A. (2010). Exploring the neural dynamics underpinning individual differences in sentence comprehension. Cerebral Cortex, 21, 1747–1760. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Pylkkänen, L. (2015). Composition of complex meaning: Interdisciplinary perspectives on the left anterior temporal lobe In Hickok G. & Small S. L. (Eds.), Neurobiology of language (pp. 622–629). Amsterdam, The Netherlands: Elsevier. [Google Scholar]
- Rayner, K. (1998). Eye movements in reading and information processing: 20 years of research. Psychological Bulletin, 124, 372–422. [DOI] [PubMed] [Google Scholar]
- Regel, S. , Gunter, T. C. , & Friederici, A. D. (2011). Isn't it ironic? An electrophysiological exploration of figurative language processing. Journal of Cognitive Neuroscience, 23, 277–293. [DOI] [PubMed] [Google Scholar]
- Robertson, D. A. , Gernsbacher, M. A. , Guidotti, S. J. , Robertson, R. R. W. , Irwin, W. , Mock, B. J. , & Campana, M. E. (2000). Functional neuroanatomy of the cognitive process of mapping during discourse comprehension. Psychological Science, 11, 255–260. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Rodd, J. M. , Davis, M. H. , & Johnsrude, I. S. (2005). The neural mechanisms of speech comprehension: fMRI studies of semantic ambiguity. Cerebral Cortex, 15, 1261–1269. [DOI] [PubMed] [Google Scholar]
- Rodd, J. M. , Johnsrude, I. S. , & Davis, M. H. (2012). Dissociating frontotemporal contributions to semantic ambiguity resolution in spoken sentences. Cerebral Cortex, 22, 1761–1773. [DOI] [PubMed] [Google Scholar]
- Rodd, J. M. , Longe, O. A. , Randall, B. , & Tyler, L. K. (2010). The functional organisation of the fronto‐temporal language system: Evidence from syntactic and semantic ambiguity. Neuropsychologia, 48, 1324–1335. [DOI] [PubMed] [Google Scholar]
- Rodd, J. M. , Vitello, S. , Woollams, A. M. , & Adank, P. (2015). Localising semantic and syntactic processing in spoken and written language comprehension: An activation likelihood estimation meta‐analysis. Brain and Language, 141, 89–102. [DOI] [PubMed] [Google Scholar]
- Rogalsky, C. , Almeida, D. , Sprouse, J. , & Hickok, G. (2015). Sentence processing selectivity in Broca's area: Evident for structure but not syntactic movement. Language, Cognition and Neuroscience, 30, 1326–1338. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Rogalsky, C. , & Hickok, G. (2009). Selective attention to semantic and syntactic features modulates sentence processing networks in anterior temporal cortex. Cerebral Cortex, 19, 786–796. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Rogalsky, C. , LaCroix, A. N. , Chen, K.‐H. , Anderson, S. W. , Damasio, H. , Love, T. , & Hickok, G. (2017). The neurobiology of Agrammatic sentence comprehension: A lesion study. Journal Cognitive Neuroscience, 30, 234–255. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Rogalsky, C. , Matchin, W. , & Hickok, G. (2008). Broca's area, sentence comprehension, and working memory: An fMRI study. Frontiers in Human Neuroscience, 2, 14. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Rorden, C. , Karnath, H.‐O. , & Bonilha, L. (2007). Improving lesion‐symptom mapping. Journal Cognitive Neuroscience, 19, 1081–1088. [DOI] [PubMed] [Google Scholar]
- Santi, A. , & Grodzinsky, Y. (2010). fMRI adaptation dissociates syntactic complexity dimensions. NeuroImage, 51, 1285–1293. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Santi, A. , & Grodzinsky, Y. (2012). Broca's area and sentence comprehension: A relationship parasitic on dependency, displacement or predictability? Neuropsychologia, 50, 821–832. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Schönberger, E. , Heim, S. , Meffert, E. , Pieperhoff, P. , da Costa Avelar, P. , Huber, W. , … Grande, M. (2014). The neural correlates of agrammatism: Evidence from aphasic and healthy speakers performing an overt picture description task. Frontiers in Psychology, 5, 246. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Schuil, K. D. , Smits, M. , & Zwaan, R. A. (2013). Sentential context modulates the involvement of the motor cortex in action language processing: An fMRI study. Frontiers in Human Neuroscience, 7, 100. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Scott, S. K. , Rosen, S. , Lang, H. , & Wise, R. J. S. (2006). Neural correlates of intelligibility in speech investigated with noise vocoded speech—A positron emission tomography study. The Journal of the Acoustical Society of America, 120, 1075–1083. [DOI] [PubMed] [Google Scholar]
- Segaert, K. , Menenti, L. , Weber, K. , Petersson, K. M. , & Hagoort, P. (2012). Shared syntax in language production and language comprehension‐‐an FMRI study. Cerebral Cortex, 22, 1662–1670. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Shetreet, E. , & Friedmann, N. (2014). The processing of different syntactic structures: fMRI investigation of the linguistic distinction between wh‐movement and verb movement. Journal of Neurolinguistics, 27, 1–14. [Google Scholar]
- Shetreet, E. , Palti, D. , Friedmann, N. , & Hadar, U. (2007). Cortical representation of verb processing in sentence comprehension: Number of complements, subcategorization, and thematic frames. Cerebral Cortex, 17, 1958–1969. [DOI] [PubMed] [Google Scholar]
- Silbert, L. J. , Honey, C. J. , Simony, E. , Poeppel, D. , & Hasson, U. (2014). Coupled neural systems underlie the production and comprehension of naturalistic narrative speech. Proceedings of the National Academy of Sciences, 111, E4687–E4696. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Simmonds, A. J. , Leech, R. , Collins, C. , Redjep, O. , & Wise, R. J. S. (2014). Sensory‐motor integration during speech production localizes to both left and right Plana Temporale. The Journal of Neuroscience, 34, 12963–12972. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Stowe, L. A. , Broere, C. A. , Paans, A. M. , Wijers, A. A. , Mulder, G. , Vaalburg, W. , & Zwarts, F. (1998). Localizing components of a complex task: Sentence processing and working memory. Neuroreport, 9, 2995–2999. [DOI] [PubMed] [Google Scholar]
- Stromswold, K. , Caplan, D. , Alpert, N. , & Rauch, S. (1996). Localization of syntactic comprehension by positron emission tomography. Brain and Language, 52, 452–473. [DOI] [PubMed] [Google Scholar]
- Swinney, D. , Love, T. , Nicol, J. , Bouck, V. , & Hald, L. A. (2000). Neuroanatomical organization of sentential processing operations: Evidence from aphasia on the (modular) processing' of discontinuous dependencies In Bastiaane R. & Grodzinsky Y. (Eds.), Grammatical disorders in aphasia: A neurolinguistic perspective (pp. 51–66). London: Whurr Publishers. [Google Scholar]
- Swinney, D. , & Osterhout, L. (1990). Inference generation during auditory language comprehension In Graesser A. C. & Bower G. H. (Eds.), Inference and text comprehension: The psychology of learning and motivation. San Diego, CA: Academic Press. [Google Scholar]
- Thompson, C. , & Kielar, A. (2014). Neural bases of sentence processing: Evidence from neurolinguistic and neuroimaging studies In Ferreira V., Goldrick M., & Miozzo M. (Eds.), The Oxford handbook of language production (pp. 47–69). New York, NY: Oxford University Press. [Google Scholar]
- Thompson, C. K. (2011). Northwestern assessment of verbs and sentences. Evanston, IL: Northwestern University. [Google Scholar]
- Thompson, C. K. , Bonakdarpour, B. , Fix, S. C. , Blumenfeld, H. K. , Parrish, T. B. , Gitelman, D. R. , & Mesulam, M. M. (2007). Neural correlates of verb argument structure processing. Journal of Cognitive Neuroscience, 19, 1753–1767. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Thompson, C. K. , Bonakdarpour, B. , & Fix, S. F. (2010). Neural mechanisms of verb argument structure processing in agrammatic aphasic and healthy age‐matched listeners. Journal of Cognitive Neuroscience, 22, 1993–2011. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Thompson, C. K. , & Choy, J. J. (2009). Pronominal resolution and gap filling in agrammatic aphasia: Evidence from eye movements. Journal of Psycholinguistic Research, 38, 255–283. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Thompson, C. K. , den Ouden, D.‐B. , Bonakdarpour, B. , Garibaldi, K. , & Parrish, T. B. (2010). Neural plasticity and treatment‐induced recovery of sentence processing in agrammatism. Neuropsychologia, 48, 3211–3227. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Thompson, C. K. , & Meltzer‐Asscher, A. (2014). Neurocognitive mechanisms of verb argument structure processing. Amsterdam, The Netherlands: John Benjamins Publishing Company. [Google Scholar]
- Thompson‐Schill, S. L. , D'Esposito, M. , Aguirre, G. K. , & Farah, M. J. (1997). Role of left inferior prefrontal cortex in retrieval of semantic knowledge: A reevaluation. Proceedings of the National Academy of Sciences of the United States of America, 94, 14792–14797. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Tracy, J. , Flanders, A. , Madi, S. , Natale, P. , Delvecchio, N. , Pyrros, A. , & Laskas, J. (2003). The Brain's response to incidental intruded words during focal text processing. NeuroImage, 18, 117–126. [DOI] [PubMed] [Google Scholar]
- Tremblay, P. , & Small, S. L. (2011). Motor response selection in overt sentence production: A functional MRI study. Frontiers in Psychology, 2, 253. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Troiani, V. , Fernández‐Seara, M. A. , Wang, Z. , Detre, J. A. , Ash, S. , & Grossman, M. (2008). Narrative speech production: An fMRI study using continuous arterial spin labeling. NeuroImage, 40, 932–939. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Trumbo, M. C. , Matzen, L. E. , Coffman, B. A. , Hunter, M. A. , Jones, A. P. , Robinson, C. S. H. , & Clark, V. P. (2016). Enhanced working memory performance via transcranial direct current stimulation: The possibility of near and far transfer. Neurpsychologia, 93, 85–96. [DOI] [PubMed] [Google Scholar]
- Tuennerhoff, J. , & Noppeney, U. (2016). When sentences live up to your expectations. NeuroImage, 124, 641–653. [DOI] [PubMed] [Google Scholar]
- Turkeltaub, P. , Eden, G. , Jones, K. , & Zeffiro, T. (2002). Meta‐analysis of the functional neuroanatomy of single‐word reading: Method and validation. NeuroImage, 16, 765–780. [DOI] [PubMed] [Google Scholar]
- Turkeltaub, P. E. , Eickhoff, S. B. , Laird, A. R. , Fox, M. , Wiener, M. , & Fox, P. (2012). Minimizing within‐experiment and within‐group effects in activation likelihood estimation meta‐analyses. Human Brain Mapping, 33, 1–13. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Tyler, L. K. , Shafto, M. A. , Randall, B. , Wright, P. , Marslen‐Wilson, W. D. , & Stamatakis, E. A. (2010). Preserving syntactic processing across the adult life span: The modulation of the frontotemporal language system in the context of age‐related atrophy. Cerebral Cortex, 20, 352–364. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Ullman, M. (2006). Is Broca's area part of a basal ganglia thalamocortical circuit? Cortex, 42, 480–485. [DOI] [PubMed] [Google Scholar]
- Ullman, M. T. (2004). Contributions of memory circuits to language: The declarative/procedural model. Cognition, 92, 231–270. [DOI] [PubMed] [Google Scholar]
- Van Ettinger‐Veenstra, H. , McAllister, A. , Lundberg, P. , Karlsson, T. , & Engström, M. (2016). Higher language ability is related to angular gyrus activation increase during semantic processing, independent of sentence incongruency. Frontiers in Human Neuroscience, 10, 110. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Van Leeuwen, T. M. , Lamers, M. J. , Petersson, K. M. , Gussenhoven, C. , Rietveld, T. , Poser, B. , & Hagoort, P. (2014). Phonological markers of information structure: An fMRI study. Neuropsychologia, 58, 64–74. [DOI] [PubMed] [Google Scholar]
- Vigneau, M. , Beaucousin, V. , Herve, P.‐Y. , Duffau, H. , Crivello, F. , Houde, O. , … Tzourio‐Mazoyer, N. (2006). Meta‐analyzing left hemisphere language areas: Phonology, semantics, and sentence processing. NeuroImage, 30, 1414–1432. [DOI] [PubMed] [Google Scholar]
- Vigneau, M. , Beaucousin, V. , Hervé, P.‐Y. , Jobard, G. , Petit, L. , Crivello, F. , … Tzourio‐Mazoyer, N. (2011). What is right‐hemisphere contribution to phonological, lexico‐semantic, and sentence processing? Insights from a meta‐analysis. NeuroImage, 54, 577–593. [DOI] [PubMed] [Google Scholar]
- Visser, M. , Jefferies, E. , & Lambon Ralph, M. A. (2010). Semantic processing in the anterior temporal lobes: A meta‐analysis of the functional neuroimaging literature. Journal of Cognitive Neuroscience, 22, 1083–1094. [DOI] [PubMed] [Google Scholar]
- Vitello, S. , Warren, J. E. , Devlin, J. T. , & Rodd, J. M. (2014). Roles of frontal and temporal regions in reinterpreting semantically ambiguous sentences. Frontiers in Human Neuroscience, 8, 530. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Von Kriegstein, K. , Eger, E. , Kleinschmidt, A. , & Giraud, A. L. (2003). Modulation of neural responses to speech by directing attention to voices or verbal content. Cognitive Brain Research, 17, 48–55. [DOI] [PubMed] [Google Scholar]
- Walczyk, J. J. (2000). The interplay between automatic and control processes in reading. Reading Research Quarterly, 35, 554–566. [Google Scholar]
- Waters, G. , Caplan, D. , Alpert, N. , & Stanczak, L. (2003). Individual differences in rCBF correlates of syntatic processing in sentence comprehension: Effects of working memory and speed of processing. NeuroImage, 19, 101–112. [DOI] [PubMed] [Google Scholar]
- Wildgruber, D. , Hertrich, I. , Riecker, A. , Erb, M. , Anders, S. , Grodd, W. , & Ackermann, H. (2004). Distinct frontal regions subserve evaluation of linguistic and emotional aspects of speech intonation. Cerebral Cortex, 14, 1384–1389. [DOI] [PubMed] [Google Scholar]
- Willems, R. M. , Frank, S. L. , Nijhof, A. D. , Hagoort, P. , & van den Bosch, A. (2016). Prediction during natural language comprehension. Cerebral Cortex, 26, 2506–2516. [DOI] [PubMed] [Google Scholar]
- Wilson, S. M. , Bautista, A. , & McCarron, A. (2018). Convergence of spoken and written language processing in the superior temporal sulcus. NeuroImage, 171, 62–74. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Wilson, S. M. , Dronkers, N. F. , Ogar, J. M. , Jang, J. , Growdon, M. E. , Agosta, F. , … Gorno‐Tempini, M. L. (2010). Neural correlates of syntactic processing in the nonfluent variant of primary progressive aphasia. The Journal of Neuroscience, 30, 16845–16854. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Wilson, S. M. , Molnar‐Szakacs, I. , & Iacoboni, M. (2008). Beyond superior temporal cortex: Intersubject correlations in narrative speech comprehension. Cerebral Cortex, 18, 230–242. [DOI] [PubMed] [Google Scholar]
- Wong, D. , Miyamoto, R. T. , Pisoni, D. B. , Sehgal, M. , & Hutchins, G. D. (1999). PET imaging of cochlear‐implant and normal‐hearing subjects listening to speech and nonspeech. Hearing Research, 132, 34–42. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Xu, J. , Kemeny, S. , Park, G. , Frattali, C. , & Braun, A. (2005). Language in context: Emergent features of word, sentence, and narrative comprehension. NeuroImage, 25, 1002–1015. [DOI] [PubMed] [Google Scholar]
- Ye, Z. , & Zhou, X. (2009). Conflict control during sentence comprehension: fMRI evidence. NeuroImage, 48, 280–290. [DOI] [PubMed] [Google Scholar]
- Yokoyama, S. , Okamoto, H. , Miyamoto, T. , Yoshimoto, K. , Kim, J. , Iwata, K. , … Kawashima, R. (2006). Cortical activation in the processing of passive sentences in L1 and L2: An fMRI study. NeuroImage, 30, 570–579. [DOI] [PubMed] [Google Scholar]
- Yokoyama, S. , Watanabe, J. , Iwata, K. , Ikuta, N. , Haji, T. , Usui, N. , … Sato, S. (2007). Is Broca's area involved in the processing of passive sentences? An event‐related fMRI study. Neuropsychologia, 45, 989–996. [DOI] [PubMed] [Google Scholar]
- Zaccarella, E. , Schell, M. , & Friederici, A. D. (2017). Reviewing the functional basis of the syntactic merge mechanism for language: A coordinate‐based activation likelihood estimation meta‐analysis. Neuroscience and Biobehavioral Reviews, 80, 646–656. [DOI] [PubMed] [Google Scholar]
- Zempleni, M.‐Z. , Renken, R. , Hoeks, J. C. J. , Hoogduin, J. M. , & Stowe, L. A. (2007). Semantic ambiguity processing in sentence context: Evidence from event‐related fMRI. NeuroImage, 34, 1270–1279. [DOI] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
Appendix S1 Data_Comprehension_and_production
Appendix S2 Data_Comprehension
Appendix S3 Data_Production
Appendix S4 Data_Auditory_Comprehension
Appendix S5 Data_Visual_Comprehension
Appendix S6 Data_noncanonical