CoChat
Sign Up
Cited by 595Neurobiology of Language and BilingualismReading and Literacy DevelopmentLanguage Development and Disorders

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

Open access
Fedorenko, Evelina, Behr, Michael K., Kanwisher, Nancy·2011·Proceedings of the National Academy of Sciences
DOI 10.1073/pnas.1112937108PMC PMC3182706
Open this paper in CoChat — free account

Abstract

Neuroscientists have debated for centuries whether some regions of the human brain are selectively engaged in specific high-level mental functions or whether, instead, cognition is implemented in multifunctional brain regions. For the critical case of language, conflicting answers arise from the neuropsychological literature, which features striking dissociations between deficits in linguistic and nonlinguistic abilities, vs. the neuroimaging literature, which has argued for overlap between activations for linguistic and nonlinguistic processes, including arithmetic, domain general abilities like cognitive control, and music. Here, we use functional MRI to define classic language regions functionally in each subject individually and then examine the response of these regions to the nonlinguistic functions most commonly argued to engage these regions: arithmetic, working memory, cognitive control, and music. We find little or no response in language regions to these nonlinguistic functions. These data support a clear distinction between language and other cognitive processes, resolving the prior conflict between the neuropsychological and neuroimaging literatures.

AI snapshot

Written from the paper’s full text.

TL;DR

Using fMRI, the authors functionally defined language regions in each subject and tested whether those regions respond to arithmetic, working memory, cognitive control, or music. They found little or no response to these nonlinguistic functions, supporting a clear distinction between language and other cognitive processes.

Key findings

  • Across seven nonlinguistic tasks, language-sensitive ROIs showed little or no response; no other ROIs showed reliable responses to any nonlinguistic task after correction, except one left middle frontal gyrus (LMFG) ROI that responded significantly to verbal working memory.
  • The music contrast produced some response in several language ROIs, but effects were weak overall and none survived correction for multiple comparisons.
  • Whole-brain overlap analyses at the ROI-analysis threshold found no spatially systematic regions where language activations overlapped with any of the seven nonlinguistic tasks; with liberal thresholding, three overlap regions emerged, including posterior left inferior frontal gyrus (LIFG) overlap with verbal working memory in 85% of subjects and a similar region with Stroop in 71%.
  • The authors conclude that most key cortical regions for high-level linguistic processing are not engaged by mental arithmetic, general working memory, cognitive control, or musical processing, with the LMFG as the main exception.

Methods

  • Forty-eight right-handed native English speakers (31 females, ages 18–40) participated; each was run on a language localizer and one or more nonlinguistic tasks. Task-specific samples were: math n = 11, spatial WM n = 12, verbal WM n = 13, MSIT n = 15, vMSIT n = 12, Stroop n = 14, and music n = 12, with overlapping subjects across tasks.
  • The language localizer contrasted reading sentences vs. pronounceable nonwords with a memory probe; nonlinguistic experiments contrasted harder vs. easier conditions for arithmetic, spatial WM, verbal WM, MSIT, vMSIT, and Stroop, and intact vs. scrambled music.
  • fMRI data were collected on a 3-T Siemens Trio, analyzed with SPM5 and custom MATLAB scripts; language ROIs were defined per subject by intersecting group-level parcels with individual localizer maps thresholded at P < 0.001 uncorrected, and responses were estimated using cross-validation across runs.

Limitations

  • The paper states that the investigated regions are not the only ones engaged in or necessary for language: language also requires auditory/visual and motor cortex, and domain-general frontal/parietal regions may be necessary for many linguistic functions.
  • The paper states that the computations in the posterior LIFG overlap region remain an open question and that higher spatial resolution might reveal the apparent overlap to be artifactual due to proximity of language-sensitive and demanding-task regions.
  • The paper notes that the localizer targets high-level linguistic processing and does not address other aspects of language such as sound-level processes; evident from the text, the whole-brain overlap findings required liberal thresholding, and per-task samples were small (11–15 participants) with overlapping subjects across experiments.

Model-written summary. Check it against the paper before relying on it.

Integrity signals

Registry

Crossref · 10.1073/pnas.1112937108

Cites

0 citation errors · 0 not checkable

Statistics

Nothing to recompute.

Claims the paper makes

0 of 39 cited sources checked · 0 supported · 0 flagged

  • Findingown data
    § Discussion

    The response to intact music in the left inferior frontal gyrus is lower than the response to the language localizer control task.

    “the response to intact music is barely above the fixation baseline (and lower than the response to the language localizer control task)”
    Quote in the paperSubstantial claim
  • Findingown data
    § Results

    At the threshold used for the region-of-interest analyses, no spatially systematic regions were detected in which language activations overlapped with any of the seven nonlinguistic tasks.

    “At the threshold used for the ROI analyses, no spatially systematic regions were detected in which language activations overlapped with any of the seven nonlinguistic tasks, confirming the overall picture of language specificity found in the ROI analyses.”
    Quote in the paperSubstantial claim
  • Findingown data
    § Discussion

    The response to intact music in the left inferior frontal gyrus is barely above the fixation baseline.

    “the response to intact music is barely above the fixation baseline (and lower than the response to the language localizer control task)”
    Quote in the paperSubstantial claim
  • Findingown data
    § Discussion

    Most of the key cortical regions engaged in high-level linguistic processing are not engaged by mental arithmetic, general working memory, cognitive control, or musical processing.

    “The current results show that most of the key cortical regions engaged in high-level linguistic processing are not engaged by mental arithmetic, general working memory, cognitive control, or musical processing, indicating a high degree of functional specificity in the brain regions that support language.”
    Quote in the paperSubstantial claim
  • Findingown data
    § Results

    At the same threshold, robust overlap regions in the frontal and parietal lobes between Stroop and the multisource interference task were discovered in a subset of subjects run on both tasks.

    “[This threshold is not too stringent to discover true overlap when it exists; at the same threshold, we discovered robust overlap regions in the frontal and parietal lobes between Stroop and the multisource interference task (MSIT) cognitive control task in a subset of the subjects run on both tasks.]”
    Quote in the paperSubstantial claim
  • May combine two claims. 10
  • Classified as probably not a substantial claim. 8

  • May combine two claims.

    Findingown data
    § Results

    The music contrast effects in language regions of interest were weak overall and none survived correction for multiple comparisons.

    “The music contrast did produce some response in several language ROIs (SI Text), suggesting possibly overlapping circuits; however, these effects were weak overall, and none survived correction for multiple comparisons.”
    Quote in the paperSubstantial claim

  • May combine two claims.

    Findingown data
    § Results

    A portion of the left inferior frontal gyrus language region of interest is selective for high-level linguistic processing and not engaged in either verbal working memory or Stroop.

    “Nevertheless, a complementary analysis shows that a portion of the LIFG language ROI is selective for high-level linguistic processing and not engaged in either verbal WM or Stroop (Fig. 4).”
    Quote in the paperSubstantial claim

  • Probably not a substantial claim: may not pass S3 (load-bearing). May combine two claims.

    Findingown data
    § Materials and Methods

    In experiment 7, activation patterns were similar across the two task groups (passive listening vs. liking question), so data from the two groups were collapsed for the current analyses.

    “Because the activation patterns were similar across the two tasks, we collapsed the data from these two groups of participants for the current analyses.”
    Quote in the paperNot a substantial claim

  • Probably not a substantial claim: may not pass S3 (load-bearing).

    Interpretationreasoning
    § Discussion

    The brain regions investigated here are not the only ones engaged in or necessary for language processing.

    “Note that the brain regions investigated here are surely not the only ones engaged in or necessary for language processing.”
    Quote in the paperNot a substantial claim

  • Probably not a substantial claim: may not pass S3 (load-bearing). May combine two claims.

    Study design and conductnone given
    § Materials and Methods

    The order of conditions was counterbalanced; each condition was equally likely to appear in the earlier vs. later parts of each run and was as likely to follow every other condition as it was to precede it.

    “The order of conditions was counterbalanced; each condition was equally likely to appear in the earlier vs. later parts of each run and was as likely to follow every other condition as it was to precede it.”
    Quote in the paperNot a substantial claim

  • Classified as probably not a substantial claim. May combine two claims.

    Study design and conductreasoning
    § Materials and Methods

    The scrambled musical pieces had component sounds similar to the intact music but lacked musical structure.

    “The resulting piece had component sounds similar to the intact music but lacked musical structure.”
    Quote in the paperSubstantial claim

  • May combine two claims.

    Interpretationown data
    § Abstract

    These data support a clear distinction between language and other cognitive processes, resolving the prior conflict between the neuropsychological and neuroimaging literatures.

    “These data support a clear distinction between language and other cognitive processes, resolving the prior conflict between the neuropsychological and neuroimaging literatures.”
    Quote in the paperSubstantial claim

  • May combine two claims.

    Premisereasoning
    § Results

    Region-of-interest analyses can obscure functional heterogeneity within a region of interest or miss important overlapping regions outside the borders of the region of interest.

    “However, ROI analyses can obscure functional heterogeneity within an ROI or miss important overlapping regions outside the borders of the ROI.”
    Quote in the paperSubstantial claim

  • Probably not a substantial claim: may not pass S3 (load-bearing).

    Interpretationnone given
    § Acknowledgments

    The current language localizer task does not target sound-level processes.

    “Of course, the question of the relationship between linguistic and nonlinguistic processes extends to other aspects of language (e.g., sound-level processes), which we are not targeting with the current language localizer task.”
    Quote in the paperNot a substantial claim

  • Classified as probably not a substantial claim.

    Study design and conductcitation
    § Materials and Methods

    Responses to the sentences and nonwords conditions were estimated using cross-validation across functional runs, ensuring that the data used to estimate the effects were always independent of the data used for ROI definition.

    “The responses to the sentences and nonwords conditions were estimated using cross-validation across the functional runs, and therefore, the data used to estimate the effects were always independent of the data used for ROI definition (46).”
    Quote in the paperSubstantial claim0 of 1 sources checked
    • cites [46] Vul E, Kanwisher N. Begging the question: The non-independence error in fMRI data analysis. In: Hanson S, Bunzl M, editors. Foundational issues for human brain mapping. Cambridge, MA: MIT Press; 2010. pp. 71–91. [Google Scholar] source

      Not checked in the free run.

  • May combine two claims.

    Premisecitation
    § Abstract

    The language localizer is robust to changes in stimulus modality (visual/auditory), specific stimuli, and task (passive reading vs. reading with a probe task).

    “and to be robust to changes in stimulus modality (visual/auditory) (see also ref. 32), specific stimuli, and task (passive reading vs. reading with a probe task, which is used here).”
    Quote in the paperSubstantial claim0 of 1 sources checked
    • cites [32] Braze D, et al. Unification of sentence processing via ear and eye: An fMRI study. Cortex. 2011;47:416–431. doi: 10.1016/j.cortex.2009.11.005. [DOI] [PMC free article] [PubMed] [Google Scholar] source

      Not checked in the free run.

  • May combine two claims.

    Premisereasoning
    § Abstract

    The neuroimaging claims are based largely on observations of activation for a nonlanguage task in or near a general brain region previously implicated in language, without a direct demonstration of overlap of linguistic and nonlinguistic activations in the same subjects.

    “However, the neuroimaging claims are based largely on observations of activation for a nonlanguage task in or near a general brain region previously implicated in language (e.g., Broca's area), without a direct demonstration of overlap of linguistic and nonlinguistic activations in the same subjects.”
    Quote in the paperSubstantial claim0 of 10 sources checked
    • cites [7] Dehaene S, Spelke E, Pinel P, Stanescu R, Tsivkin S. Sources of mathematical thinking: Behavioral and brain-imaging evidence. Science. 1999;284:970–974. doi: 10.1126/science.284.5416.970. [DOI] [PubMed] [Google Scholar] source

      Not checked in the free run.

    • cites [8] Novick JM, Trueswell JC, Thompson-Schill SL. Cognitive control and parsing: Reexamining the role of Broca's area in sentence comprehension. Cogn Affect Behav Neurosci. 2005;5:263–281. doi: 10.3758/cabn.5.3.263. [DOI] [PubMed] [Google Scholar] source

      Not checked in the free run.

    • cites [9] Levitin DJ, Menon V. Musical structure is processed in “language” areas of the brain: A possible role for Brodmann Area 47 in temporal coherence. Neuroimage. 2003;20:2142–2152. doi: 10.1016/j.neuroimage.2003.08.016. [DOI] [PubMed] [Google Scholar] source

      Not checked in the free run.

    • cites [10] Awh E, et al. Dissociation of storage and rehearsal in verbal working memory: Evidence from positron emission tomography. Psych Science. 1996;7:25–31. [Google Scholar] source

      Not checked in the free run.

    • cites [11] Badre D, Poldrack RA, Paré-Blagoev EJ, Insler RZ, Wagner AD. Dissociable controlled retrieval and generalized selection mechanisms in ventrolateral prefrontal cortex. Neuron. 2005;47:907–918. doi: 10.1016/j.neuron.2005.07.023. [DOI] [PubMed] [Google Scholar] source

      Not checked in the free run.

    • cites [12] Blumstein S. Reflections on the cognitive neuroscience of language. In: Gazzaniga M, editor. The Cognitive Neurosciences. Cambridge, MA: MIT Press; 2009. pp. 1235–1240. [Google Scholar]

      Not checked in the free run.

    • cites [13] Braver TS, et al. A parametric study of prefrontal cortex involvement in human working memory. Neuroimage. 1997;5:49–62. doi: 10.1006/nimg.1996.0247. [DOI] [PubMed] [Google Scholar] source

      Not checked in the free run.

    • cites [14] Bunge SA, Klingberg T, Jacobsen RB, Gabrieli JDE. A resource model of the neural basis of executive working memory. Proc Natl Acad Sci USA. 2000;97:3573–3578. doi: 10.1073/pnas.050583797. [DOI] [PMC free article] [PubMed] [Google Scholar] source

      Not checked in the free run.

    • cites [15] Cohen JD, et al. Activation of the prefrontal cortex in a nonspatial working memory task with functional MRI. Hum Brain Mapp. 1994;1:293–304. doi: 10.1002/hbm.460010407. [DOI] [PubMed] [Google Scholar] source

      Not checked in the free run.

    • cites [16] Fadiga L, Craighero L, D'Ausilio A. Broca's area in language, action, and music. Ann N Y Acad Sci. 2009;1169:448–458. doi: 10.1111/j.1749-6632.2009.04582.x. [DOI] [PubMed] [Google Scholar] source

      Not checked in the free run.

  • Probably not a substantial claim: may not pass S4 (based). May combine two claims.

    Premisenone given
    § Results

    Region-of-interest-based analyses afford great statistical power and reduce the blurring of functional activations inevitable in group analyses.

    “ROI-based analyses afford great statistical power and reduce the blurring of functional activations inevitable in group analyses.”
    Quote in the paperNot a substantial claim

  • Classified as probably not a substantial claim.

    Study design and conductnone given
    § Results

    Responses were estimated using data from left-out runs not used for defining the regions of interest.

    “(here and throughout the study, responses were estimated using data from left-out runs not used for defining the ROIs).”
    Quote in the paperSubstantial claim
14 of 14

  • The paper relies on this without support.

    The individual-subject overlap analysis strategy has almost never been applied in neuroimaging investigations of high-level linguistic processing.

    “a data analysis strategy that has almost never been applied in neuroimaging investigations of high-level linguistic processing”

  • The paper relies on this without support.

    Most claims about the overlap between linguistic and nonlinguistic processes have concerned syntactic processing.

    “However, most claims about the overlap between linguistic and nonlinguistic processes have concerned (i) syntactic processing, which is included in our functional contrast, and (ii) brain regions, which are robustly activated by our localizer (e.g., Broca's area).”

  • The paper relies on this without support.

    Most claims about the overlap between linguistic and nonlinguistic processes have concerned brain regions robustly activated by the language localizer.

    “However, most claims about the overlap between linguistic and nonlinguistic processes have concerned (i) syntactic processing, which is included in our functional contrast, and (ii) brain regions, which are robustly activated by our localizer (e.g., Broca's area).”
3 of 3

Claims other papers make about this one

Each is checked against this paper’s own text: does it say what the citing paper says it does?

  • The hard > easy contrast in the verbal working memory task has been found to robustly activate regions of the MD network.

    “The hard > easy contrast has been found to robustly activate regions of the MD network (Fedorenko et al., 2011(Fedorenko et al., , 2013)).”

    From Functional Specificity and Sex Differences in the Neural Circuits Supporting the Inhibition of Automatic Imitation (2018) · cites this paper as Fedorenko et al., 2011

    The Materials and Methods describe a hard-versus-easy verbal working-memory task, and the Results report robust activation from each nonlinguistic task in brain regions near the language ROIs (page not provided). The paper does not identify those verbal-working-memory activations specifically as regions of the MD network, so that part of the claim is not established by our copy. (LLM, conf=0.81)

    Experimental — accuracy not yet measured

  • Neuroimaging studies have consistently shown that the language network remains largely inactive during various reasoning tasks.

    “Neuroimaging studies have consistently shown that the language network -a set of brain regions responsible for language comprehension and production -remains largely inactive during various reasoning tasks”

    From Training Large Language Models to Reason in a Continuous Latent Space (2024) · cites this paper as Fedorenko et al., 2011

    The Results report little response in language-sensitive regions during the seven nonlinguistic tasks tested, including arithmetic, working memory, and cognitive control. But the Abstract says prior neuroimaging work had suggested overlap between linguistic and nonlinguistic activations, so this paper does not support the claim that neuroimaging studies have consistently shown inactivity across reasoning tasks. (LLM, conf=0.88)

    Experimental — accuracy not yet measured

  • Perisylvian regions are more involved in processing language inputs than some other brain systems.

    “There is little doubt that perisylvian regions are more involved in processing language inputs than some other brain systems (see, e.g., Blank & Fedorenko, 2017), and may be particularly involved in language tasks (e.g., Fedorenko, Behr, & Kanwisher, 2011).”

    From Grounding the neurobiology of language in first principles: The necessity of non-language-centric explanations for language comprehension (2018) · cites this paper as Fedorenko, Behr, & Kanwisher, 2011

  • fMRI reveals that the language network responds reliably when people listen to, read, or generate sentences, but not when they perform arithmetic tasks, engage in logical reasoning, understand computer programs, listen to music, categorize objects or events, reason about people's mental states, or process non-verbal communicative information like facial expressions or gestures.

    “it responds reliably when people listen to, read, or generate sentences, but not when they perform arithmetic tasks, engage in logical reasoning, understand computer programs, listen to music, categorize objects or events, reason about people’s mental states, or process non-verbal communicative information like facial expressions or gestures [37, 41, 42, 43, 44, 45, 46, 47, 48].”

    From Dissociating language and thought in large language models (2024) · cites this paper as [37]

  • The language network is sensitive to linguistic regularities at multiple levels, from phonological/sub-lexical to phrase/sentence level.

    “is sensitive to linguistic regularities at multiple levels: from phonological/sub-lexical [36] to phrase/sentence level [37, 38]”

    From Dissociating language and thought in large language models (2024) · cites this paper as [37]

Cites

Flagged first, then by citations

Research works printed but not linked: 1

  1. P Broca, [Notes on the seat of the faculty of articulate language, followed by an observation of aphemia. Bull Soc Anat Paris 6, 330–357, French. (1861).

Cited by

Sorted by citations

Authors
Fedorenko, E., Behr, M. K., & Kanwisher, N.
Affiliations
  • Massachusetts Institute of Technology (US)
  • Brain and Cognitive Sciences Department and
  • McGovern Institute for Brain Research (US)
ORCID
Venue
Proceedings of the National Academy of Sciences
Publisher
National Academy of Sciences
Published
2011-09
Volume
108
Issue
39
Pages
16428-16433
ISSN
  • 0027-8424 (ISSN-L)
  • 1091-6490
DOI
10.1073/pnas.1112937108
PMC
PMC3182706
OpenAlex
W2096462008
Semantic Scholar
b62b60be4fe51188d90a27dd9fc695c924a55b0b
Cited by
595
References
0
Citations by year
2026: 43 · 2025: 52 · 2024: 75 · 2023: 42 · 2022: 47 · 2021: 61 · 2020: 38 · 2019: 32 · 2018: 38 · 2017: 33 · 2016: 34 · 2015: 35 · 2014: 22 · 2013: 24 · 2012: 17
FWCI
10.27
Type
journal article
Language
en
Open access
Locations
  • Proceedings of the National Academy of Sciences (publishedVersion)
  • Europe PMC (PubMed Central) (publishedVersion)
  • PubMed Central (publishedVersion)
  • PubMed (publishedVersion)
  • DSpace@MIT (Massachusetts Institute of Technology) (submittedVersion, cc-by-nc)
Funders
  • Massachusetts Institute of Technology
  • National Institutes of Health (5k99hd057522-02)
  • McGovern Institute for Brain Research, Massachusetts Institute of Technology
  • National Institute of Child Health and Human Development
  • Eunice Kennedy Shriver National Institute of Child Health and Human Development (K99 HD-057522)
Topics
  • Life Sciences › Neuroscience › Cognitive Neuroscience › Neurobiology of Language and Bilingualism
  • Social Sciences › Psychology › Developmental and Educational Psychology › Reading and Literacy Development
  • Social Sciences › Psychology › Developmental and Educational Psychology › Language Development and Disorders
Keywords
language lateralization, functional specificity, language regions, arithmetic, working memory, cognitive control, music processing, domain specificity
MeSH terms
Brain, Female, Humans, Linguistics, Magnetic Resonance Imaging, Male, Speech

About these checks

Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.

Claims read by deepseek-v4.1-flash under definition version 3 (extract-v3) on 2026-10-07.

Classified by jev-1.13-20260917 (gold-guideline-v0.2) on 2026-10-07.

Statistics and arithmetic are recomputed, not judged.

Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.

What counts as a claim ›How we measure confidence ›How we check this ›

Are you an author, or do you see a problem with this page? Request a review