Neurobiology of Language

Understanding the cognitive architecture of the human language faculty requires exploring the boundaries of both predisposition and environmental experience. However, previous research on extraordinary multilingualism has often confounded language aptitude with multilingual experience, obscuring their distinct neural correlates. Here, we leveraged a linguistically diverse sample (N=121) and extensive behavioural testing to dissociate language aptitude from multilingual experience, modelling both dimensions continuously in whole-brain speech processing. Language aptitude and multilingual experience were weakly related, and their dissociation was also evident at the neural level. Higher language aptitude showed a neural signature of efficiency, characterised by lower activation in core perisylvian regions. In contrast, higher multilingualism was associated with greater engagement of regions implicated in narrative, multimodal, and memory processing, and with recruitment of traditional language hubs only during degraded speech processing, likely reflecting active attempts to decode unintelligible input. Finally, aptitude and experience interacted within sensorimotor regions. Continuous quantification of multilingual experience proved more sensitive than artificial grouping. By disentangling language aptitude from multilingual experience, this work provides a more precise account of the multilingual brain, and shows that its neurobiology can be better understood by modelling predisposition and experience as distinct but interacting dimensions.

Language experience is a powerful driver of neurocognitive plasticity, shown to modulate attentional and executive processing in speakers of multiple languages. This paper investigated how immersion in a second-language environment shapes attentional processing of speech. Fifty-eight bilingual Chinese-English speakers matched on their English language proficiency listened to competing continuous speech streams in a naturalistic listening task. They were immersed either in their native language environment (Beijing, China), or in the second language environment (Cambridge, UK). In an identical EEG experiment across the two immersion contexts, we assessed the listeners cortical tracking of attended and unattended speech and task-related attentional allocation using Temporal Response Function (mTRF) and Power Spectral Density (PSD) analyses. Behavioral comprehension of the attended narratives was uniformly high. PSD analyses showed no group difference in task-related attentional allocation, but mTRF results revealed robust differences in cortical tracking, with increased tracking of attended - but not unattended - streams in the immersed group, driven by the delta band (1-4 Hz). This boost in tracking of the attended signal declined with prolonged immersion, indicative of changes to attentional speech processing as the language environment stabilizes and consistent with the expansion-renormalization framework of neurocognitive adaptation. Jointly, these data imply that immersion in second-language environments shapes the way listeners encode speech, sharpening the brains ability to extract target auditory information from background noise. They furthermore suggest that, rather than being static or monotonous, this modulation reflects a flexible and dynamic process that is continuously shaped by changes in environmental demands and their duration. Key pointsO_LISecond language immersion boosts cortical tracking of attended speech, driven by the delta band (1-4 Hz). C_LIO_LIThis boost decreases with prolonged immersion, reflecting the dynamic, expansion-renormalization adaptation trajectory. C_LIO_LIImmersion does not influence task-related attentional allocation. C_LI

Aphasia is characterized by impaired word retrieval, yet most cognitive models of word production assume that underlying conceptual-semantic representations are largely preserved. This study investigated whether concept-level semantic structure remains decodable from BOLD signals in chronic post-stroke aphasia and which semantic models best explain neural representational geometry during covert semantic feature generation. Eight healthy adults and six individuals with chronic aphasia completed a dense-sampling fMRI protocol in which they viewed 57 pictured nouns while silently generating semantic features. Representational similarity analysis showed that an experiential model (Exp48) best matched neural geometry in both people with aphasia and controls, outperforming taxonomic (WordNet) and distributional (Word2Vec, GloVe) models. Using representational similarity decoding, concept identity was recovered well above chance in both groups. No relationship was found between decoding accuracy and language measures from individuals with aphasia. These findings suggest that experiential semantic structure remains robustly represented and decodable in chronic aphasia despite lesion-related language impairments, highlighting preserved conceptual representations alongside altered anatomy.

Irony and sarcasm are complex forms of non-literal language that hinge on a misalignment between surface meaning and speaker intent, requiring listeners to integrate contextual, semantic, and prosodic cues. While prior neuroimaging studies have implicated a broad network--including the temporal cortex, the inferior frontal gyrus, and the medial prefrontal cortex--in the comprehension of ironic and sarcastic speech, the precise neural mechanisms underlying the integration of semantic and prosodic information remain unclear. In the present study, we addressed this gap by employing voxel-wise encoding models to systematically identify brain regions specifically involved in combining prosodic and semantic cues during non-literal language comprehension. Participants listened to naturalistic auditory dialogues in which both discourse context and target utterance semantics and prosody were systematically manipulated. We derived custom text embeddings using transformer-based models to capture context-sensitive semantic representations of ironic statements, alongside acoustic features characterizing affective prosody. Ridge regression models were fitted to predict BOLD responses at the voxel level using semantic, prosodic, and combined features, and we identified integration as voxels in which each modality contributed predictive information beyond the other, using a permutation-based conjunction test. The regions integrating prosody and semantics depended on whether discourse context was modeled: integration was confined to the bilateral temporal speech cortex when statements were encoded in isolation, but additionally engaged the left inferior frontal gyrus pars orbitalis (IFGorb) when each statement was weighted by its relevance to the preceding context. These findings indicate that the left IFGorb integrates prosody with context-dependent meaning, engaging beyond the temporal speech cortex specifically when comprehension requires combining semantic, prosodic, and contextual cues--as in irony and sarcasm.

Word classes such as nouns, verbs, and adjectives are fundamental units of language, but their neural encoding remains unclear. Here, we investigate whether word classes are processed as invariant, context-independent lexical categories or depend on sentence context during language processing. We analyse source-localized magnetoencephalography (MEG) data from 200 native Dutch participants who read or listened to sentences and their scrambled counterparts (word lists) from the MOUS dataset. Time-resolved encoding models are used to predict neural responses from major word classes (noun, verb, adjective) and other linguistic variables including word frequency, surprisal, entropy, word length, and ordinal position. Across modalities, we observe a significant interaction between word class and context (sentences vs. word lists), manifest as dynamic modulation of neural responses in a widespread cortical network including bilateral perisylvian, frontal, and midline regions previously implicated in lexicosemantic, structural, and pragmatic processing. This interaction emerges early and reappears later in processing, with distinct temporal profiles for reading and listening. Within-condition effects reveal that word class contributes to neural responses at the level of individual words, but this contribution is context-dependent: it is robust in sentences across modalities and in word-list reading, but absent in auditory word lists. These results indicate that word-class encoding is shaped by the interaction between word-level properties and sentence context during real-time language processing.

Successful speech perception requires listeners to bin continuous acoustic information into discrete phonetic categories. However, some people maintain within-category acoustic information (gradient) while others discard category-irrelevant information (discrete) during perception. Listeners also vary in how consistently they label speech sounds and more gradient/consistent labeling has been linked with better speech-in-noise (SIN) perception. Here, we test how neuroanatomical properties of the brain's major speech-language and auditory pathways relate to individual differences in speech categorization and SIN processing. We measured phonetic categorization and SIN comprehension via phoneme labeling and QuickSIN tasks. Diffusion-weighted imaging (DWI) with probabilistic tractography estimated axonal density within the bilateral arcuate fasciculi and brainstem-cortical auditory projections. Anatomical morphology (surface area, gray matter volume, thickness) was also quantified in the adjacent frontotemporal cortical areas and midbrain. Behaviorally, we found more consistent categorizers had better performance on the QuickSIN. DWI showed that more gradient listeners had greater white matter density in the left arcuate fasciculus and brainstem-cortical auditory pathways, while better SIN performance was predicted by denser white matter in the brainstem-cortical auditory pathways. Morphometric results revealed more consistent listening was associated with greater cortical thickness in right superior temporal gyrus and more gradient listening was associated with greater surface area in right pars opercularis. We infer that individual differences in phonetic categorization relate to SIN comprehension and are at least partially explained by neuroanatomical properties of the auditory-linguistic brain.

Sentence comprehension requires the incremental construction of syntactic structure and semantic interpretation. Prior neural work (Nelson et al., 2017) identified key neural events at major phrase boundaries during sentence comprehension. To investigate a behavioral correlation of these processes, we used self-paced reading to examine the impact of syntactic phase boundaries, semantic congruence, and sentence structure on sentence processing. Participants read object-relative, subject-relative, and canonical control sentences one word at a time and a subsequent comprehension task. Reading times were analyzed relative to phrase boundaries, node-closing operations, and semantic congruence. Object-relative sentences produced the greatest processing difficulty, demonstrated by increased reading times and decreased comprehension accuracy. Reading times peaked at the phrase boundaries, indicating that processing costs are tied to constituent completion rather than individual lexical categories. Reading times also increased with the number of syntactic constituents completed at a phrase boundary. Agent-patient semantic congruence produced its largest effects in object-relative sentences, suggesting that semantic information interacts with syntactic computations when processing demands are greatest. These findings demonstrate that self-paced reading is sensitive to the incremental processing associated with syntactic constituent completion. Processing costs are tied more closely to phrase completion than to individual lexical categories, scale with the amount of syntactic structure completed at a boundary and interact with agent-patient semantic interpretation during object-relative sentence comprehension. Together, these findings support a view of sentence comprehension in which syntactic structure building and semantic interpretation proceed incrementally and interact continuously throughout online language processing.

Second language (L2) processing engages both language-specific and domain-general control systems, yet how these systems vary with L2 proficiency remains unclear. We used functional magnetic resonance imaging (fMRI) to examine neural activity during L2 English processing in Korean-English (K-E) bilinguals across three proficiency levels (beginner, intermediate, advanced). Participants performed rhyme and spelling judgement tasks manipulating orthographic-phonological conflict. Behaviourally, conflict conditions reduced accuracy, with proficiency effects observed selectively in the rhyme task. fMRI results showed that conflict processing recruited frontoparietal control regions, including inferior frontal and parietal cortices, accompanied by deactivation in default mode network regions. Critically, proficiency-related effects differed by task. During rhyme judgement, advanced bilinguals showed greater activation in the left supramarginal gyrus (SMG) and cerebellum, whereas intermediate bilinguals exhibited greater recruitment of the left middle orbital gyrus and dorsomedial prefrontal cortex. During spelling judgement, advanced bilinguals showed greater thalamic activation alongside greater deactivation of the right dorsolateral prefrontal cortex. Activity in the left SMG and cerebellum was positively associated with L2 reading score, and cerebellar activity was also associated with rhyme-task performance, whereas right DLPFC activity was negatively associated with the scores. These findings suggest that increasing L2 proficiency is associated less with altered recruitment of core reading regions than with task-specific shifts in the balance between phonological-specialized, subcortical attentional, and domain-general control systems supporting L2 processing.

Human language processing can be studied through both behavior and brain activity, yet it remains unclear whether these two data types reflect sensitivity to the same information. One influential view holds that both behavioral and neural responses are largely determined by processing effort, often estimated by word surprisal together with the context-independent properties of word frequency and length. At the same time, neural responses have been shown to encode richer aspects of linguistic content, including meaning. Here, we use neural network language models to operationalize these alternatives and systematically compare, within the same analytic computational framework, the predictive power of low-dimensional effort-based predictors and high-dimensional embedding representations that encode contextualized linguistic content, including meaning. Across 8 behavioral datasets and 5 neural datasets (4 fMRI and 1 ERP), we find that processing effort captures substantial variance in both behavioral and neural measures of language processing, in line with much previous work. However, for brain responses---but not for behavioral measures---embedding representations carry substantial predictive power beyond the estimates of processing effort. These results therefore suggest that neural data provide access to rich, high-dimensional dynamics of language comprehension, whereas behavioral data reflect a bottlenecking of these dynamics into a small set of theoretically motivated properties of contextualized linguistic input.

The human brain rapidly transforms continuous speech into structured, meaningful linguistic representations, yet how prior knowledge constrains this process remains unclear. To characterize this influence, we combined MEG recordings acquired during audiobook listening with corpus-derived transition probabilities over syntactic features defined within both phrase-structure and dependency-based grammars. Across grammatical formalisms, prior knowledge selectively sharpened the neural representation of memory-related features indexing syntactic structures that must remain open for future completion. This enhancement was strictly local: immediately preceding contexts improved neural decoding at both word onset and offset, whereas longer histories produced either a return to baseline at the word level or a deterioration in decoding performance. By contrast, integration-related features indexing the completion of syntactic operations showed no benefit from prior knowledge and were represented most strongly at word offset, consistent with their dependence on word-level structural resolution. These dissociable dynamics reveal two concurrent neural computational regimes for syntactic processing: a forward-looking, locally maintained predictive code for pending structure and an integrative code engaged when structure is resolved. More broadly, our findings impose a mechanistic constraint on neural theories of language processing and on accounts that equate prediction in human language comprehension with the comparatively unconstrained operations of large language models (LLMs).

In non-invasive studies of the human brain, semantic processing during language comprehension has been extensively studied using the N400, a component of the electrophysiological evoked response that is strongly modulated by semantic context. More recently, a complementary approach has emerged that uses multivariate pattern analysis to perform neural decoding of semantic vectors from brain activity, operating on the basis that semantic vectors that more closely align with representations in the brain can be more accurately decoded. To consolidate these two approaches, we investigated the relationship between N400 modulation and semantic decoding performance using magnetoencephalography (MEG) in a controlled priming experiment. Twenty-five native speakers of Finnish read word triplets, for which the semantic relatedness between the two primes and the target word was manipulated based on distance in a word2vec embedding space (highly related, moderately related, or unrelated). We found that words presented after unrelated primes elicited higher N400 responses and provided the best examples for training a decoder to map distributed MEG responses to semantic vectors. Semantic information was decodable from approximately 100 to 500 ms after stimulus onset, at all three levels of contextual support. Soon after the N400 peak, neural responses no longer seemed to encode information that could be mapped to context-invariant semantic vectors. This suggests that the end of the N400 window may correspond to a turning point where the representation shifts from being word-specific to encoding the greater semantic context.

Although the brain areas for language processing are well delimited, whether lexical-semantic and syntactic processes are spatially segregated remains debated. To clarify this issue, we conducted two experiments using 7-Tesla functional MRI in 20 participants performing: a functional localizer involving reading sequences of words of increasing linguistic complexity; and a presentation of short, semantically impoverished three-word mini-sentences, flashed in a single glance (e.g., "he does it"), whose grammaticality and syntactic complexity was manipulated through syntactic movement. Our results reveal two functionally dissociable sets of cortical patches within the language system: one sensitive to syntactic structure even in the absence of meaning, and the other involved in semantic composition. This dual-network architecture was consistently observed in the majority of participants, although its precise anatomical localization varied. The two types of voxels coexisted even within a given brain region of the Glasser atlas. Results were confirmed using subject-specific analyses and region-by-condition interactions, as voxels in those two systems displayed markedly different responses to mini-sentences. Thus, high-resolution functional imaging reveals a division of labor between syntactic and semantic composition within the classical language network.

Classic neurobiological models of human speech and language have emphasized the dominant role of temporal lobe in speech perception, while premotor regions including the ventral premotor cortex (PMv) are situated at the level of articulatory processing. However, accumulating evidence from neuroimaging, clinical, and computational studies suggests that premotor cortex may contribute to speech processing beyond articulation. The precise extent and functional organization of these speech-related representations, however, remain unclear. In this study, we combined naturalistic speech perception with computational encoding models to characterize the organization of speech representations within PMv. We functionally localized a cortical region that encompasses previously described premotor speech areas, which we term the premotor language area (PML). Using acoustic, phonemic, semantic, and deep neural speech representations, we found that PML contains representations spanning the full speech-processing hierarchy, from low-level acoustic features to high-level semantic information. These representations are arranged along a smooth posterior-anterior gradient, with increasingly abstract speech representations emerging toward anterior PML. Moreover, this organizational gradient mirrors the canonical speech processing hierarchy in the temporal auditory regions. These findings challenge the traditional view of premotor cortex as primarily an acoustic-articulatory unit, and instead identify PML as a hierarchically organized speech-processing region that parallels the temporal auditory cortex. This provides a new framework for understanding the role of premotor cortex in speech perception.

Language comprehension requires the continuous transformation of speech into a hierarchy of linguistic units, from phonemes to syllables to words. Because speech unfolds rapidly, listeners are thought to predict upcoming content to keep pace. Previous research has provided empirical evidence for predictive processes operating at multiple linguistic levels during naturalistic listening, including words and phonemes. However, it remains unclear whether prediction also operates concurrently at other levels, such as syllabic and phrasal representation. Here we use Mandarin Chinese to examine the neural signatures of predictive processing across multiple levels of linguistic granularity during natural speech comprehension. Mandarin comprises four representational levels: phoneme, sub-syllabic, character and word, and its lexical identity is largely constrained at the sub-syllabic level, potentially redistributing predictive weight across linguistic representations. We recorded magnetoencephalography (MEG) data while 34 native Mandarin speakers (21 females) listened to a naturalistic audiobook and applied linear regression modeling to examine how linguistic features modulated neural activity. We found that the brain activity of listeners segmented speech into hierarchical units, and that surprisal modulated responses simultaneously across sub-syllabic, character and word levels. In contrast to findings from Indo-European languages, however, we did not observe unique surprisal effects at the lowest, phonemic level. Furthermore, the surprisal of lexical tone in Mandarin modulated brain activity only when integrated with its phonological components. These findings suggest that predictive processing during Mandarin speech comprehension operates concurrently across multiple (though not necessarily all) levels of linguistic granularity, with its implementation shaped by language-specific structural properties. Significance statementLanguage comprehension involves segmenting a continuous acoustic stream into multiple linguistic units, from phonemes to words, and generating predictions at these levels. However, direct neural evidence remains limited regarding how segmentation and prediction operate simultaneously across levels of linguistic granularity, particularly outside Indo-European languages. Using temporal response function analysis of magnetoencephalography data recorded during naturalistic Mandarin listening, we show that predictive processing occurs across multiple levels of linguistic granularity. Specifically, we find evidence for prediction-related neural responses at sub-syllabic, character, and word levels, but not a reliable unique effect at the phonemic level. These results indicate that predictive processing also operates during Mandarin speech comprehension, and its neural implementation is shaped by language-specific structural properties.

The Visual Word Form Area (VWFA), located in ventral occipitotemporal cortex, plays a critical role in skilled reading. Researchers have theorized that the VWFA develops in its specific anatomical location due to the convergence of major white matter tracts and proximity to functionally similar regions. This suggests that precise anatomical positioning may be crucial for optimal VWFA function. Previous research has identified several functional differences in this region between typical and struggling readers (i.e. dyslexia): struggling readers show weaker text-selective responses and often exhibit a smaller or even absent VWFA. However, it remains unexplored whether the precise anatomical location of this region also differs between typical and struggling readers. We tested whether VWFA anatomy differs between children with and without dyslexia (N=87). Participants completed a functional localizer, which we used to manually define the VWFA in each individuals native anatomy. We examined whether: (1) VWFA anatomical location relates to reading ability, (2) children with dyslexia show greater variability in VWFA location compared to typical readers, and (3) VWFA location with respect to white matter tracts relates to reading ability. Results reveal that, despite being smaller in children with dyslexia, there is no relationship between VWFA location and reading ability. Specifically, individual VWFA location relative to anatomy, relative to others VWFAs, and relative to white matter tracts, is not related to reading ability. These findings suggest that while the VWFAs general anatomy may be facilitated by development, its precise location remains stable and unrelated to reading proficiency.

Abstract Background: Transcranial direct current stimulation (tDCS) paired with behavioral language therapy can improve naming in persons with aphasia (PWA), yet naming errors persist. Little is known about how naming error patterns change after non-invasive brain stimulation is combined with language treatment. Aims: To examine whether right cerebellar tDCS plus computerized aphasia therapy changes the types of naming errors in people with chronic aphasia across timepoints, and to determine whether effects differ by cerebellar tDCS polarity (anode vs. cathode). Methods and Procedures: In a randomized, double-blind, sham-controlled, within-subject crossover study, we retrospectively analyzed behavioral data from 24 individuals with post-stroke aphasia. Each participant completed two 15-session intervention periods (3-5 sessions/week) with active cerebellar tDCS + computerized aphasia therapy and sham + computerized aphasia therapy, separated by a two-month washout. General linear models (GLMs) assessed longitudinal changes in six error types (semantic, phonological real word, phonological nonword, no response, mixed, unrelated) on an untrained picture naming task (Philadelphia Naming Test; PNT) and a trained task (Naming 80; N80). Additional GLMs evaluated polarity effects with 2 (Group: anode vs. cathode) x 2 (Treatment) interactions, and treatment-order effects with 2 (Group: tDCS-first vs. sham-first) x 2 (Treatment) interactions. Outcomes and Results: Active cerebellar tDCS did not significantly change error types for trained items (N80). For untrained items (PNT), active tDCS reduced several error types relative to sham, with the clearest and most durable reduction in phonological nonword errors; more moderate reductions occurred for phonological real word and unrelated errors. Mixed errors showed a marginally opposite pattern, tending to increase after tDCS and decrease after sham. Polarity analyses indicated broadly similar effects across anodal and cathodal stimulation overall, but only the anode group showed a reliable treatment effect for phonological nonword errors on the PNT. Treatment-order analyses revealed no significant order effects. Conclusions: Our results indicate a shift in naming error types, particularly after tDCS treatment for the untrained naming task (PNT). These findings may help guide the course of treatment approaches of those with aphasia and what error naming pattern types may show changes post stroke when combining non-invasive brain stimulation and computerized aphasia therapy. Clinical Trial Registration: Cerebellar Transcranial Direct Current Stimulation and Aphasia Treatment [NCT02901574] Keywords: aphasia, naming errors, non-invasive brain stimulation, cerebellar tDCS, computerized aphasia treatment

Learning to read a second language requires the brain to incorporate a new writing system into an already established native-language reading network, yet how this process reshapes the Visual Word Form Area (VWFA) remains poorly understood. Using fMRI with a passive viewing paradigm, we systematically investigated how VWFA responses to Chinese (L1) and English (L2) words evolved across three groups of native Mandarin-speaking children at distinct stages of L2 literacy acquisition: L2 pre-readers, L2 beginning readers, and L2 advanced readers. We combined univariate activation analyses, representational similarity analysis, and supervised machine learning classification to investigate two theoretical accounts of VWFA reorganization: the Overlay Model, which predicts stable L1 responses as L2 responses emerge, and the Competing Model, which predicts competitive reallocation of neural resources from L1 to L2. We found that although the VWFA already exhibited robust selective responses to L1 words, L2-word selectivity was absent in L2 pre-readers but emerged robustly in beginning readers, with L1-word selectivity remaining stable throughout. Representational similarity analysis further revealed that robust L2 word representations within the VWFA emerged only after children began learning to read L2, while L1 word representations remained stable across all three groups. Finally, supervised machine learning analyses successfully discriminated among the three L2 literacy groups on the basis of L2 but not L1 activation patterns, indicating that VWFA responses to L2 alone were sufficient to capture childrens stages of L2 literacy acquisition, whereas responses to L1 were insensitive to L2 learning experience, providing no support for the Competing Models prediction of competitive neural reallocation away from L1. Together, these findings support the Overlay Model, demonstrating that the VWFA incorporates a new writing system within its existing cortical resources without compromising L1 print processing.

How internal speech relates to overt speech remains a fundamental question in speech production: do different forms of speech preserve a common neural representation of the intended word, or does that representation change as speech becomes articulated? We used time-resolved electroencephalography to characterize representations of 10 Chinese words during imagined, silent, and overt speech. Word identity was reliably decodable in all three modes, but its temporal dynamics differed: imagined-speech representations peaked earlier and were less temporally stable, whereas silent and overt speech showed stronger and more sustained representations. Cross-mode decoding revealed word-discriminative information shared across all three mode pairs, with substantially stronger generalization between silent and overt speech. However, direct comparison of word-level representational geometry revealed robust correspondence only between silent and overt speech, indicating that transferable information across modes does not necessarily imply preservation of the broader relational structure among words. Representational similarity analyses further showed distinct visual-form, semantic, and phonetic dynamics across speech modes, with late visual-form and phonetic information contributing uniquely to the geometry shared by silent and overt speech. Controlling for time-matched surface electromyography preserved the overall silent-overt neural correspondence and within-mode phonetic representations, while eliminating the unique phonetic contribution to their shared geometry, suggesting that peripheral articulation accounts for part, but not all, of this common structure. Together, these findings show that imagined, silent, and overt speech share word representations at different levels and suggest that representational geometry and temporal stability are progressively reorganized as internal speech is translated into articulation.

Listeners discretize the speech signal by assigning sounds to phonetic categories, though there is variability in how individuals accomplish categorization. Having more consistent categorization of sounds may be advantageous for understanding speech-in-noise (SIN). Though, it is unclear how different levels of neural processing in the auditory system reflect these perceptual differences. We recorded brainstem frequency-following responses (FFRs) and cortical event-related potentials (ERPs) while listeners actively labeled vowels along an acoustic-phonetic continuum using a visual analog scale. We computed intertrial consistency of neural responses to index the stability of listeners' neural speech representations across stimulus presentations. We also assessed how faithfully midbrain and cortical responses represented stimulus acoustics using representational dissimilarity matrices (RDMs) computed across all token pairs. Neural RDMs were then compared with acoustic and phonetic category RDMs to assess whether FFRs and ERPs carried gradient vs. categorical information of the speech signal. We found greater behavioral consistency during phoneme labeling was correlated with improved SIN scores. Neurally, we found greater cortical or subcortical consistency predicted greater behavioral consistency. RDMs revealed subcortical responses retained more acoustic details, while cortical responses more closely reflected abstract phoneme categories. Our findings reveal important benefits of perceptual consistency to other domains of speech perception. We find perceptual consistency is driven by more consistent encoding of speech at either a cortical or subcortical level. More consistent sensory processing could provide a more stable readout of the speech signal to higher cortical brain areas which could confer advantages to later perceptual processes downstream.

Prosody is a fundamental aspect of speech characterized by suprasegmental features such as pitch. Prosodic pitch contours are used to convey speakers intentions, for example, to make a statement or ask a question. Understanding these intentions requires abstracting continuous, variable pitch information into discrete categories. Category perception has been proposed to recruit the motor system in an effector-specific manner, whereby cortical areas controlling motor effectors support speech sound recognition by identifying articulatory gestures. However, it remains unclear whether effectors involved in pitch production similarly contribute to prosodic category perception. To address this question, we collected magnetoencephalography data from 29 participants (15 females) while they first sang pitches arranged in five-tone melodies and then identified the prosody (Statement vs. Question) of single words varying in pitch contour along a five-level continuum. Using a region-restricted searchlight approach to decode singing from rest, we localized two premotor regions for pitch production, corresponding to the ventral and dorsal laryngeal motor cortex (LMC). A separate neural decoding analysis revealed that perceived prosodic categories were decodable in these regions, especially from the dorsal LMC that is more closely associated with pitch regulation. Importantly, decoding performance mirrored behavioral discriminability of prosodic categories across the continuum, suggesting that these regions are involved in perceptual decision-making. Finally, pitch motor areas exchanged category-related information with auditory regions, indicating these areas do not merely echo the processing in auditory regions. Together, these findings highlight effector-specific motor support for prosodic category perception, thereby broadening our understanding of motor involvement in speech perception. Significance StatementSpeech perception has been proposed to recruit the premotor cortex, with different subregions linking speech sounds to the articulatory gestures used to produce them. We investigated this idea through prosody--pitch changes in speech conveying meanings such as statements and questions. Using magnetoencephalography, we identified pitch motor areas during a singing task and tested whether they represent perceived prosodic categories. We found that prosodic categories were distinguishable in these regions and that this neural discriminability mirrored behavioral discriminability across clear and ambiguous prosody, suggesting involvement in perceptual decision-making. These findings are unlikely to reflect passive echoes from auditory regions, as pitch motor areas actively influenced them during categorical processing. Our results highlight effector-specific motor support for forming abstract prosodic representations.