Skip to main content

Thank you for visiting nature.com. You are using a browser version with limited support for CSS. To obtain the best experience, we recommend you use a more up to date browser (or turn off compatibility mode in Internet Explorer). In the meantime, to ensure continued support, we are displaying the site without styles and JavaScript.

Visual and linguistic semantic representations are aligned at the border of human visual cortex

Nature Neuroscience volume 24, pages 1628–1636 (2021)Cite this article

Subjects

Abstract

Semantic information in the human brain is organized into multiple networks, but the fine-grain relationships between them are poorly understood. In this study, we compared semantic maps obtained from two functional magnetic resonance imaging experiments in the same participants: one that used silent movies as stimuli and another that used narrative stories. Movies evoked activity from a network of modality-specific, semantically selective areas in visual cortex. Stories evoked activity from another network of semantically selective areas immediately anterior to visual cortex. Remarkably, the pattern of semantic selectivity in these two distinct networks corresponded along the boundary of visual cortex: for visual categories represented posterior to the boundary, the same categories were represented linguistically on the anterior side. These results suggest that these two networks are smoothly joined to form one contiguous map.

This is a preview of subscription content, access via your institution

Relevant articles

Open Access articles citing this article.

Access options

Rent or buy this article

Get just this article for as long as you need it

$39.95

Learn more

Prices may be subject to local taxes which are calculated during checkout

Fig. 1: Voxels with correlated visual and linguistic semantic representations.
Fig. 2: Visual and linguistic representations of semantic concepts known to be well-represented in visual cortex.
Fig. 3: Method for detecting category-specific modality shifts.
Fig. 4: Locations of category-specific modality shifts across cortex.
Fig. 5: Quantitative summary of semantic correspondence across the boundary.
Fig. 6: Alignment of semantic selectivity along the boundary between vision and language.

Data availability

Data are available on Box (https://berkeley.box.com/s/l95gie5xtv56zocsgugmb7fs12nujpog) and at https://gallantlab.org/. All data other than anatomical brain images (as there is concern that anatomical images could violate participant privacy) have been shared. However, we have provided matrices that map from volumetric data to cortical flat maps for visualization purposes.

Code availability

Custom code used for cortical surface-based analyses is available at https://github.com/gallantlab/vl_interface.

References

  1. Barsalou, L. W. Perceptual symbol systems. Behav. Brain Sci. 22, 577–609 (1999).

    Article  CAS  Google Scholar 

  2. Damasio, A. R. The brain binds entities and events by multiregional activation from convergence zones. Neural Comput. 1, 123–132 (1989).

    Article  Google Scholar 

  3. Ralph, M. A. L., Jefferies, E., Patterson, K. & Rogers, T. T. The neural and computational bases of semantic cognition. Nat. Rev. Neurosci. 18, 42–55 (2017).

    Article  CAS  Google Scholar 

  4. Snowden, J. S., Goulding, P. J. & Neary, D. Semantic dementia: a form of circumscribed cerebral atrophy. Behav. Neurol. 2, 167–182 (1989).

  5. Warrington, E. K. The selective impairment of semantic memory. Q. J. Exp. Psychol. 27, 635–657 (1975).

    Article  CAS  Google Scholar 

  6. Wilkins, A. & Moscovitch, M. Selective impairment of semantic memory after temporal lobectomy. Neuropsychologia 16, 73–79 (1978).

    Article  CAS  Google Scholar 

  7. Jefferies, E., Patterson, K., Jones, R. W., Bateman, D. & Lambon Ralph, M. A. A category-specific advantage for numbers in verbal short-term memory: evidence from semantic dementia. Neuropsychologia 42, 639–660 (2004).

    Article  Google Scholar 

  8. Kramer, J. H. et al. Distinctive neuropsychological patterns in frontotemporal dementia, semantic dementia, and Alzheimer disease. Cogn. Behav. Neurol. 16, 211–218 (2003).

    Article  Google Scholar 

  9. Hodges, J. R., Patterson, K., Oxbury, S. & Funnell, E. Semantic dementia. Progressive fluent aphasia with temporal lobe atrophy. Brain 115, 1783–1806 (1992).

    Article  Google Scholar 

  10. Hodges, J. R. et al. The differentiation of semantic dementia and frontal lobe dementia (temporal and frontal variants of frontotemporal dementia) from early Alzheimer’s disease: a comparative neuropsychological study. Neuropsychology 13, 31–40 (1999).

    Article  CAS  Google Scholar 

  11. Damasio, H., Grabowski, T. J., Tranel, D., Hichwa, R. D. & Damasio, A. R. A neural basis for lexical retrieval. Nature 380, 499–505 (1996).

    Article  CAS  Google Scholar 

  12. Damasio, H., Tranel, D., Grabowski, T., Adolphs, R. & Damasio, A. Neural systems behind word and concept retrieval. Cognition 92, 179–229 (2004).

    Article  CAS  Google Scholar 

  13. Devereux, B. J., Clarke, A., Marouchos, A. & Tyler, L. K. Representational similarity analysis reveals commonalities and differences in the semantic processing of words and objects. J. Neurosci. 33, 18906–18916 (2013).

    Article  CAS  Google Scholar 

  14. Fairhall, S. L. & Caramazza, A. Brain regions that represent amodal conceptual knowledge. J. Neurosci. 33, 10552–10558 (2013).

    Article  CAS  Google Scholar 

  15. Kanwisher, N., McDermott, J. & Chun, M. M. The fusiform face area: a module in human extrastriate cortex specialized for face perception. J. Neurosci. 17, 4302–4311 (1997).

    Article  CAS  Google Scholar 

  16. Epstein, R. & Kanwisher, N. A cortical representation of the local visual environment. Nature 392, 598–601 (1998).

    Article  CAS  Google Scholar 

  17. Downing, P. E., Jiang, Y., Shuman, M. & Kanwisher, N. A cortical area selective for visual processing of the human body. Science 293, 2470–2473 (2001).

    Article  CAS  Google Scholar 

  18. Huth, A. G., Nishimoto, S., Vu, A. T. & Gallant, J. L. A continuous semantic space describes the representation of thousands of object and action categories across the human brain. Neuron 76, 1210–1224 (2012).

    Article  CAS  Google Scholar 

  19. Huth, A. G., de Heer, W. A., Griffiths, T. L., Theunissen, F. E. & Gallant, J. L. Natural speech reveals the semantic maps that tile human cerebral cortex. Nature 532, 453–458 (2016).

    Article  Google Scholar 

  20. Deniz, F., Nunez-Elizalde, A. O., Huth, A. G. & Gallant, J. L. The representation of semantic information across human cerebral cortex during listening versus reading is invariant to stimulus modality. J. Neurosci. 39, 7722–7736 (2019).

    Article  CAS  Google Scholar 

  21. Kay, K. N., Naselaris, T., Prenger, R. J. & Gallant, J. L. Identifying natural images from human brain activity. Nature 452, 352–355 (2008).

    Article  CAS  Google Scholar 

  22. Mitchell, T. M. et al. Predicting human brain activity associated with the meanings of nouns. Science 320, 1191–1195 (2008).

    Article  CAS  Google Scholar 

  23. Naselaris, T., Kay, K. N., Nishimoto, S. & Gallant, J. L. Encoding and decoding in fMRI. Neuroimage 56, 400–410 (2011).

    Article  Google Scholar 

  24. Nishimoto, S. et al. Reconstructing visual experiences from brain activity evoked by natural movies. Curr. Biol. 21, 1641–1646 (2011).

    Article  CAS  Google Scholar 

  25. Miller, G. A. WordNet: a lexical database for English. Commun. ACM 38, 39–41 (1995).

    Article  Google Scholar 

  26. Nakamura, K. et al. Functional delineation of the human occipito-temporal areas related to face and scene processing. A PET study. Brain 123, 1903–1912 (2000).

    Article  Google Scholar 

  27. Hasson, U., Harel, M., Levy, I. & Malach, R. Large-scale mirror-symmetry organization of human occipito-temporal object areas. Neuron 37, 1027–1041 (2003).

    Article  CAS  Google Scholar 

  28. Dilks, D. D., Julian, J. B., Paunov, A. M. & Kanwisher, N. The occipital place area is causally and selectively involved in scene perception. J. Neurosci. 33, 1331–6a (2013).

    Article  Google Scholar 

  29. Aguirre, G. K., Zarahn, E. & D’Esposito, M. An area within human ventral cortex sensitive to ‘building’ stimuli: evidence and implications. Neuron 21, 373–383 (1998).

    Article  CAS  Google Scholar 

  30. Ono, M., Kubik, S. & Abernathy, C. D. Atlas of the Cerebral Sulci (Thieme Medical Publishers, 1990).

  31. Friedman, L. & Glover, G. H., Fbirn Consortium. Reducing interscanner variability of activation in a multicenter fMRI study: controlling for signal-to-fluctuation-noise-ratio (SFNR) differences. Neuroimage 33, 471–481 (2006).

    Article  Google Scholar 

  32. Ojemann, J. G. et al. Anatomic localization and quantitative analysis of gradient refocused echo-planar fMRI susceptibility artifacts. Neuroimage 6, 156–167 (1997).

    Article  CAS  Google Scholar 

  33. Van Essen, D. C., Anderson, C. H. & Felleman, D. J. Information processing in the primate visual system: an integrated systems perspective. Science 255, 419–423 (1992).

    Article  Google Scholar 

  34. Modha, D. S. & Singh, R. Network architecture of the long-distance pathways in the macaque brain. Proc. Natl Acad. Sci. USA 107, 13485–13490 (2010).

    Article  CAS  Google Scholar 

  35. Ercsey-Ravasz, M. et al. A predictive network model of cerebral cortical connectivity based on a distance rule. Neuron 80, 184–197 (2013).

    Article  CAS  Google Scholar 

  36. Visser, M., Jefferies, E. & Lambon Ralph, M. A. Semantic processing in the anterior temporal lobes: a meta-analysis of the functional neuroimaging literature. J. Cogn. Neurosci. 22, 1083–1094 (2009).

    Article  Google Scholar 

  37. Lewis, J. W., Talkington, W. J., Puce, A., Engel, L. R. & Frum, C. Cortical networks representing object categories and high-level attributes of familiar real-world action sounds. J. Cogn. Neurosci. 23, 2079–2101 (2011).

    Article  Google Scholar 

  38. Norman-Haignere, S., Kanwisher, N. G. & McDermott, J. H. Distinct cortical pathways for music and speech revealed by hypothesis-free voxel decomposition. Neuron 88, 1281–1296 (2015).

    Article  CAS  Google Scholar 

  39. Levy, I., Hasson, U., Avidan, G., Hendler, T. & Malach, R. Center–periphery organization of human object areas. Nat. Neurosci. 4, 533 (2001).

    Article  CAS  Google Scholar 

  40. Nunez-Elizalde, A. O., Huth, A. G. & Gallant, J. L. Voxelwise encoding models with non-spherical multivariate normal priors. Neuroimage 197, 482–492 (2019).

  41. Gao, J. S., Huth, A. G., Lescroart, M. D. & Gallant, J. L. Pycortex: an interactive surface visualizer for fMRI. Front. Neuroinform. 9, 23 (2015).

    Article  Google Scholar 

Download references

Acknowledgements

We thank J. Nguyen for assistance transcribing and aligning story stimuli and B. Griffin and M.-L. Kieseler for segmenting and flattening cortical surfaces. Funding: This work was supported by grants from the National Science Foundation (NSF) (IIS1208203), the National Eye Institute (EY019684 and EY022454) and the Center for Science of Information, an NSF Science and Technology Center, under grant agreement CCF-0939370. S.F.P. was also supported by the William Orr Dingwall Neurolinguistics Fellowship. A.G.H. was also supported by the William Orr Dingwall Neurolinguistics Fellowship and the Burroughs-Wellcome Fund Career Award at the Scientific Interface.

Author information

Author notes

  1. Alexander G. Huth

    Present address: Departments of Neuroscience & Computer Science, The University of Texas at Austin, Austin TX, USA

  2. Fatma Deniz

    Present address: Department of Software Engineering and Theoretical Computer Science, Technische Universität Berlin, Berlin, Germany

Authors and Affiliations

  1. Helen Wills Neuroscience Institute, University of California, Berkeley, Berkeley, CA, USA

    Sara F. Popham, Alexander G. Huth, Natalia Y. Bilenko, Fatma Deniz, James S. Gao, Anwar O. Nunez-Elizalde & Jack L. Gallant

  2. Department of Psychology, University of California, Berkeley, Berkeley, CA, USA

    Jack L. Gallant

Authors
  1. Sara F. Popham
    View author publications

    You can also search for this author in PubMed Google Scholar

  2. Alexander G. Huth
    View author publications

    You can also search for this author in PubMed Google Scholar

  3. Natalia Y. Bilenko
    View author publications

    You can also search for this author in PubMed Google Scholar

  4. Fatma Deniz
    View author publications

    You can also search for this author in PubMed Google Scholar

  5. James S. Gao
    View author publications

    You can also search for this author in PubMed Google Scholar

  6. Anwar O. Nunez-Elizalde
    View author publications

    You can also search for this author in PubMed Google Scholar

  7. Jack L. Gallant
    View author publications

    You can also search for this author in PubMed Google Scholar

Contributions

S.F.P., A.G.H. and J.L.G. conceptualized the experiment. A.G.H., N.Y.B. and F.D. collected the data. S.F.P., A.G.H., N.Y.B., J.S.G. and A.O.N.-E. contributed to analysis. S.F.P., A.G.H. and J.L.G. wrote the paper.

Corresponding author

Correspondence to Jack L. Gallant.

Ethics declarations

Competing interests

The authors declare no competing financial interests.

Additional information

Peer review information Nature Neuroscience thanks Christopher Baldassano and Johan Carlin for their contribution to the peer review of this work.

Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Extended data

Extended Data Fig. 1 Evaluation of the visual and linguistic semantic models.

Model weights are estimated on the training dataset, then are used to predict brain activity to a held-out dataset. Prediction performance is the correlation of actual and predicted brain activity for each voxel. These performance values are presented simultaneously using a 2-dimensional colormap on the flattened cortex around the occipital pole for each subject. Red voxels are locations where the visual semantic model is performing well, blue voxels are where the linguistic semantic model is performing well, and white voxels are where both models are performing equally well. These maps show where the visual and linguistic networks of the brain abut each other.

Extended Data Fig. 2 Visual and linguistic representations of place concepts.

Identical analysis to Fig. 2a, but for the other 10 subjects. The color of each voxel indicates the representation of place-related information according to the legend at the right. The model weights for vision and language are shown in red and blue, respectively. White borders indicate ROIs found in separate localizer experiments. Three relevant place ROIs are labeled: PPA, OPA, and RSC. Centered on each ROI there is a modality shift gradient that runs from visual semantic categories (red) posterior to linguistic semantic categories (blue) anterior.

Extended Data Fig. 3 Visual and linguistic representations of body part concepts.

Identical analysis to Fig. 2b, but for the other 10 subjects. The color of each voxel indicates the representation of body-related information according to the legend at the right. The model weights for vision and language are shown in red and blue, respectively. White borders indicate ROIs found in separate localizer experiments. The relevant body ROI is labeled: EBA. Centered on each ROI there is a modality shift gradient that runs from visual semantic categories (red) posterior to linguistic semantic categories (blue) anterior.

Extended Data Fig. 4 Visual and linguistic representations of face concepts.

Identical analysis to Fig. 2c, but for the other 10 subjects. The color of each voxel indicates the representation of face-related information according to the legend at the right. The model weights for vision and language are shown in red and blue, respectively. White borders indicate ROIs found in separate localizer experiments. The relevant face ROI is labeled: FFA. Centered on each ROI there is a modality shift gradient that runs from visual semantic categories (red) posterior to linguistic semantic categories (blue) anterior.

Extended Data Fig. 5 Analysis region around the boundary of the occipital lobe.

The thin yellow line indicates the estimated border of the occipital lobe of the brain in each individual subject. This was manually drawn to follow the parieto-occipital sulcus and connect to the preoccipital notch on both ends. The area of the brain which was analyzed in this study was limited to vertices within 50 mm of this border, which is shown in black on each individual’s brain.

Extended Data Fig. 6 Locations of category-specific modality shifts across cortex for alternate parameter set 1.

Identical analysis to Fig. 4, but with an ROI size of 10x25mm. Shown here is the flattened cortex around the occipital pole for one typical subject, along with inflated hemispheres. The modality shift metric calculated at each location near the boundary of the occipital lobe is plotted as an arrow. The arrow color represents the magnitude of the shift. The arrow is directed to show the shift from vision to language. Only locations where the modality shift is statistically significant are shown. Areas of fMRI signal dropout are indicated with hash marks. There are strong modality shifts in a clear ring around visual cortex in the same locations seen in Fig. 4.

Extended Data Fig. 7 Locations of category-specific modality shifts across cortex for alternate parameter set 2.

Identical analysis to Fig. 4, but with an ROI size of 10x10mm. Shown here is the flattened cortex around the occipital pole for one typical subject, along with inflated hemispheres. The modality shift metric calculated at each location near the boundary of the occipital lobe is plotted as an arrow. The arrow color represents the magnitude of the shift. The arrow is directed to show the shift from vision to language. Only locations where the modality shift is statistically significant are shown. Areas of fMRI signal dropout are indicated with hash marks. There are strong modality shifts in a ring around visual cortex in the same locations seen in Fig. 4, though the pattern is more noisy due to the shortened analysis windows.

Supplementary information

Rights and permissions

Reprints and Permissions

About this article

Verify currency and authenticity via CrossMark

Cite this article

Popham, S.F., Huth, A.G., Bilenko, N.Y. et al. Visual and linguistic semantic representations are aligned at the border of human visual cortex. Nat Neurosci 24, 1628–1636 (2021). https://doi.org/10.1038/s41593-021-00921-6

Download citation

  • Received:

  • Accepted:

  • Published:

  • Issue Date:

  • DOI: https://doi.org/10.1038/s41593-021-00921-6

This article is cited by

Search

Advanced search

Quick links

Nature Briefing

Sign up for the Nature Briefing newsletter — what matters in science, free to your inbox daily.

Get the most important science stories of the day, free in your inbox. Sign up for Nature Briefing