Data availability
The data supporting the findings of this study are publicly available at GitHub (https://github.com/lbcn-stanford/Nature2026). These data include masks for the four insular regions in MNI152 ICBM 2009c nonlinear asymmetric space; electrophysiological connectivity data for intra-insular and insula–cortical connectivity; functional electrical stimulation data; GradCPT data for both regular and reversed task conditions; and sensory-evoked potential data from the Italy cohort. Electrode coordinates in each dataset are provided in MNI152 ICBM 2009c nonlinear asymmetric space. All other patient data including EEG recordings can be shared with academic and research teams by direct request from the corresponding author.
References
Wager, T. D. et al. An fMRI-based neurologic signature of physical pain. N. Engl. J. Med. 368, 1388–1397 https://doi.org/10.1056/NEJMoa1204471 (2013).
Ostrowsky, K. et al. Representation of pain and somatic sensation in the human insula: a study of responses to direct electrical cortical stimulation. Cereb. Cortex 12, 376–385 https://doi.org/10.1093/cercor/12.4.376 (2002).
Garfinkel, S. N. & Critchley, H. D. Interoception, emotion and brain: new insights link internal physiology to social behaviour. Commentary on: “Anterior insular cortex mediates bodily sensibility and social anxiety” by Terasawa et al. (2012). Soc. Cogn. Affect. Neurosci. 8, 231–234 https://doi.org/10.1093/scan/nss140 (2013).
Seth, A. K. Interoceptive inference, emotion, and the embodied self. Trends Cogn. Sci. 17, 565–573 https://doi.org/10.1016/j.tics.2013.09.007 (2013).
Craig, A. D. Significance of the insula for the evolution of human awareness of feelings from the body. Ann. N. Y. Acad. Sci. 1225, 72–82 https://doi.org/10.1111/j.1749-6632.2011.05990.x (2011).
Singer, T., Critchley, H. D. & Preuschoff, K. A common role of insula in feelings, empathy and uncertainty. Trends Cogn. Sci. 13, 334–340 https://doi.org/10.1016/j.tics.2009.05.001 (2009).
Quabs, J. et al. Cytoarchitecture, probability maps and segregation of the human insula. Neuroimage 260, 119453 https://doi.org/10.1016/j.neuroimage.2022.119453 (2022).
Kurth, F. et al. Cytoarchitecture and probabilistic maps of the human posterior insular cortex. Cereb. Cortex 20, 1448–1461 https://doi.org/10.1093/cercor/bhp208 (2010).
Grodzinsky, Y. et al. Logical negation mapped onto the brain. Brain Struct. Funct. 225, 19–31 https://doi.org/10.1007/s00429-019-01975-w (2020).
Hein, M., Mohlberg, H., Bludau, S., Quabs, J. & Amunts, K. Probabilistic cytoarchitectonic map of Areas Ia2, Ia3, Id8, Id9, Id10 (v4.0). EBRAINS https://doi.org/10.25493/7QE6-PBE (2021).
Sherrington, C. S. The Integrative Action of the Nervous System 317 (Yale Univ. Press, 1906).
Craig, A. D. An ascending general homeostatic afferent pathway originating in lamina I. Prog. Brain Res. 107, 225–242 (1996).
Cameron, O. G. Interoception: the inside story—a model for psychosomatic processes. Psychosom. Med. 63, 697–710 https://doi.org/10.1097/00006842-200109000-00001 (2001).
Critchley, H. D., Wiens, S., Rotshtein, P., Ohman, A. & Dolan, R. J. Neural systems supporting interoceptive awareness. Nat. Neurosci. 7, 189–195 https://doi.org/10.1038/nn1176 (2004).
Namkung, H., Kim, S. H. & Sawa, A. The insula: an underestimated brain area in clinical neuroscience, psychiatry, and neurology. Trends Neurosci. 40, 200–207 https://doi.org/10.1016/j.tins.2017.02.002 (2017).
Uddin, L. Q., Nomi, J. S., Hébert-Seropian, B., Ghaziri, J. & Boucher, O. Structure and function of the human insula. J. Clin. Neurophysiol. 34, 300–306 https://doi.org/10.1097/WNP.0000000000000377 (2017).
Quabs, J., Bittner, N. & Caspers, S. Structural connectivity differences reflect microstructural heterogeneity of the human insular cortex. Hum. Brain Mapp. 46, e70231 https://doi.org/10.1002/hbm.70231 (2025).
Craig, A. D. How do you feel–now? The anterior insula and human awareness. Nat. Rev. Neurosci. 10, 59–70 https://doi.org/10.1038/nrn2555 (2009).
Damasio, A., Damasio, H. & Tranel, D. Persistence of feelings and sentience after bilateral damage of the insula. Cereb. Cortex 23, 833–846 https://doi.org/10.1093/cercor/bhs077 (2013).
Barrett, L. F., Mesquita, B., Ochsner, K. N. & Gross, J. J. The experience of emotion. Annu. Rev. Psychol. 58, 373–403 https://doi.org/10.1146/annurev.psych.58.110405.085709 (2007).
Barrett, L. F. & Simmons, W. K. Interoceptive predictions in the brain. Nat. Rev. Neurosci. 16, 419–429 https://doi.org/10.1038/nrn3950 (2015).
Paulus, M. P. & Stein, M. B. An insular view of anxiety. Biol. Psychiatry 60, 383–387 https://doi.org/10.1016/j.biopsych.2006.03.042 (2006).
Afif, A., Chabardes, S., Minotti, L., Kahane, P. & Hoffmann, D. Safety and usefulness of insular depth electrodes implanted via an oblique approach in patients with epilepsy. Neurosurgery 62, ONS471–ONS479 https://doi.org/10.1227/01.neu.0000326037.62337.80 (2008).
Avanzini, P., Pelliccia, V., Lo Russo, G., Orban, G. A. & Rizzolatti, G. Multiple time courses of somatosensory responses in human cortex. Neuroimage 169, 212–226 https://doi.org/10.1016/j.neuroimage.2017.12.037 (2018).
Seeley, W. W. The Salience Network: a neural system for perceiving and responding to homeostatic demands. J. Neurosci. 39, 9878–9882 https://doi.org/10.1523/JNEUROSCI.1138-17.2019 (2019).
Badke D’Andrea, C. et al. Action-mode subnetworks for decision-making, action control, and feedback. Proc. Natl Acad. Sci. USA 122, e2502021122 https://doi.org/10.1073/pnas.2502021122 (2025).
Rosenberg, M., Noonan, S., DeGutis, J. & Esterman, M. Sustaining visual attention in the face of distraction: a novel gradual-onset continuous performance task. Atten. Percept. Psychophys. 75, 426–439 https://doi.org/10.3758/s13414-012-0413-x (2013).
Lyu, D. et al. Mapping human thalamocortical connectivity with electrical stimulation and recording. Nat. Neurosci. 28, 1797–1809 https://doi.org/10.1038/s41593-025-02009-x (2025).
Almashaikhi, T. et al. Functional connectivity of insular efferences. Hum. Brain Mapp. 35, 5279–5294 https://doi.org/10.1002/hbm.22549 (2014).
Almashaikhi, T. et al. Intrainsular functional connectivity in human. Hum. Brain Mapp. 35, 2779–2788 https://doi.org/10.1002/hbm.22366 (2014).
Matsumoto, R. et al. Functional connectivity in the human language system: a cortico-cortical evoked potential study. Brain 127, 2316–2330 (2004).
Keller, C. J. et al. Mapping human brain networks with cortico-cortical evoked potentials. Philos. Trans. R. Soc. Lond. B 369, 20130528 https://doi.org/10.1098/rstb.2013.0528 (2014).
Russo, S. et al. Thalamic feedback shapes brain responses evoked by cortical stimulation in mice and humans. Nat. Commun. 16, 3627 https://doi.org/10.1038/s41467-025-58717-2 (2025).
Fan, L. et al. The Human Brainnetome Atlas: a new brain atlas based on connectional architecture. Cereb. Cortex 26, 3508–3526 https://doi.org/10.1093/cercor/bhw157 (2016).
Penfield, W. & Faulk, M. E. Jr. The insula; further observations on its function. Brain 78, 445–470 https://doi.org/10.1093/brain/78.4.445 (1955).
Mesulam, M. & Mufson, E. J. in Association and Auditory Cortices (eds Peters, A. & Jones, E. G.) 179–226 (Springer, 1985).
Kwon, M. et al. Convergent and selective representations of pain, appetitive processes, aversive processes, and cognitive control in the insula. Nat. Commun. https://doi.org/10.1038/s41467-026-71568-9 (2026).
Amunts, K., Mohlberg, H., Bludau, S. & Zilles, K. Julich-Brain: a 3D probabilistic atlas of the human brain’s cytoarchitecture. Science 369, 988–992 https://doi.org/10.1126/science.abb4588 (2020).
Vestergaard, M., Carta, M., Guney, G. & Poulet, J. F. A. The cellular coding of temperature in the mammalian cortex. Nature 614, 725–731 https://doi.org/10.1038/s41586-023-05705-5 (2023).
Henderson, L. A., Gandevia, S. C. & Macefield, V. G. Somatotopic organization of the processing of muscle and cutaneous pain in the left and right insula cortex: a single-trial fMRI study. Pain 128, 20–30 https://doi.org/10.1016/j.pain.2006.08.013 (2007).
Brooks, J. C., Zambreanu, L., Godinez, A., Craig, A. D. & Tracey, I. Somatotopic organisation of the human insula to painful heat studied with high resolution functional imaging. Neuroimage 27, 201–209 https://doi.org/10.1016/j.neuroimage.2005.03.041 (2005).
Mazzola, L., Isnard, J., Peyron, R., Guenot, M. & Mauguiere, F. Somatotopic organization of pain responses to direct electrical stimulation of the human insular cortex. Pain 146, 99–104 https://doi.org/10.1016/j.pain.2009.07.014 (2009).
Mesulam, M. M. From sensation to cognition. Brain 121, 1013–1052 (1998).
Fox, K. C. R. et al. Intrinsic network architecture predicts the effects elicited by intracranial electrical stimulation of the human brain. Nat. Hum. Behav. 4, 1039–1052 https://doi.org/10.1038/s41562-020-0910-1 (2020).
Sterling, P. Allostasis: a model of predictive regulation. Physiol. Behav. 106, 5–15 https://doi.org/10.1016/j.physbeh.2011.06.004 (2012).
Pezzulo, G., Rigoli, F. & Friston, K. Active inference, homeostatic regulation and adaptive behavioural control. Prog. Neurobiol. 134, 17–35 https://doi.org/10.1016/j.pneurobio.2015.09.001 (2015).
Dosenbach, N. U. F., Raichle, M. E. & Gordon, E. M. The brain’s action-mode network. Nat. Rev. Neurosci. 26, 158–168 https://doi.org/10.1038/s41583-024-00895-x (2025).
Reiss, S., Peterson, R. A., Gursky, D. M. & McNally, R. J. Anxiety sensitivity, anxiety frequency and the prediction of fearfulness. Behav. Res. Ther. 24, 1–8 https://doi.org/10.1016/0005-7967(86)90143-9 (1986).
Lusk, Z. et al. Combining clinical evaluations and neuroscience research in the human intracranial electroencephalography practice: 15-year cohort study. J. Cogn. Neurosci. 37, 2108–2125 https://doi.org/10.1162/JOCN.a.53 (2025).
Duong, A. et al. Topographical map of subjective states evoked by focal seizures and electrical stimulation of the human insula. Epilepsia 66, 2766–2778 https://doi.org/10.1111/epi.18433 (2025).
Parvizi, J. & Kastner, S. Promises and limitations of human intracranial electroencephalography. Nat. Neurosci. 21, 474–483 https://doi.org/10.1038/s41593-018-0108-2 (2018).
Liu, S. & Parvizi, J. Cognitive refractory state caused by spontaneous epileptic high-frequency oscillations in the human brain. Sci. Transl. Med. 11, eaax7830 https://doi.org/10.1126/scitranslmed.aax7830 (2019).
Groppe, D. M. et al. iELVis: An open source MATLAB toolbox for localizing and visualizing human intracranial electrode data. J. Neurosci. Methods 281, 40–48 https://doi.org/10.1016/j.jneumeth.2017.01.022 (2017).
Fischl, B. FreeSurfer. Neuroimage 62, 774–781 (2012).
Jenkinson, M., Bannister, P., Brady, M. & Smith, S. Improved optimization for the robust and accurate linear registration and motion correction of brain images. Neuroimage. 17, 825–841 (2002).
Papademetris, X. et al. BioImage Suite: an integrated medical image analysis suite: an update. Insight J. 2006, 209 (2006).
Kurth, F., Zilles, K., Fox, P. T., Laird, A. R. & Eickhoff, S. B. A link between the systems: functional differentiation and integration within the human insula revealed by meta-analysis. Brain Struct. Funct. 214, 519–534 https://doi.org/10.1007/s00429-010-0255-z (2010).
Esterman, M., Noonan, S. K., Rosenberg, M. & Degutis, J. In the zone or zoning out? Tracking behavioral and neural fluctuations during sustained attention. Cereb. Cortex 23, 2712–2723 https://doi.org/10.1093/cercor/bhs261 (2013).
Kucyi, A. et al. Electrophysiological dynamics of antagonistic brain networks reflect attentional fluctuations. Nat. Commun. 11, 325 https://doi.org/10.1038/s41467-019-14166-2 (2020).
Esterman, M., Chiu, Y. C., Tamber-Rosenau, B. J. & Yantis, S. Decoding cognitive control in human parietal cortex. Proc. Natl Acad. Sci. USA 106, 17974–17979 https://doi.org/10.1073/pnas.0903593106 (2009).
Wager, T. D. et al. Common and unique components of response inhibition revealed by fMRI. Neuroimage 27, 323–340 https://doi.org/10.1016/j.neuroimage.2005.01.054 (2005).
Sylvester, C. Y. et al. Switching attention and resolving interference: fMRI measures of executive functions. Neuropsychologia 41, 357–370 https://doi.org/10.1016/s0028-3932(02)00167-7 (2003).
Thompson-Schill, S. L., Ramscar, M. & Chrysikou, E. G. Cognition without control: when a little frontal lobe goes a long way. Curr. Dir. Psychol. Sci. 18, 259–263 https://doi.org/10.1111/j.1467-8721.2009.01648.x (2009).
Aron, A. R., Robbins, T. W. & Poldrack, R. A. Inhibition and the right inferior frontal cortex. Trends Cogn. Sci. 8, 170–177 (2004).
Acknowledgements
We thank our patients for volunteering their time; the staff at EEG technologies at Stanford Health for ensuring smooth integration of clinical and research workflows; the members of the Laboratory of Behavioural and Cognitive Neuroscience for their engaging discussions during the early drafting of the manuscript; and K. Amunts for her continuous support and discussions regarding area Id4.
Funding
This work was supported by research grant R01NS137650 to J.P.; R21NS140918 to A.G.R.; and a gift from G. Casella to H.V.
Author information
Authors and Affiliations
Laboratory of Behavioral and Cognitive Neuroscience, Stanford University, Palo Alto, CA, USA
Julian Quabs, Sofia Pantis, Guanpeng Chen, Weichen Huang, Eileen Ma, Dian Lyu, Chloe Wang & Josef Parvizi
Department of Neurology and Neurological Sciences, Stanford School of Medicine, Palo Alto, CA, USA
Julian Quabs, Sofia Pantis, Guanpeng Chen, Weichen Huang, Dian Lyu, Hannes Vogel & Josef Parvizi
Department of Pathology, Stanford University, Stanford, CA, USA
Julian Quabs & Hannes Vogel
Institute for Anatomy I, Medical Faculty and University Hospital, Heinrich Heine University of Düsseldorf, Düsseldorf, Germany
Julian Quabs & Svenja Caspers
Institute of Neuroscience and Medicine (INM-1), Research Centre Jülich, Jülich, Germany
Julian Quabs & Svenja Caspers
Istituto di Neuroscienze, Consiglio Nazionale delle Ricerche, Parma, Italy
Maria del Vecchio & Pietro Avanzini
Department of Neurosurgery, Stanford School of Medicine, Palo Alto, CA, USA
Vivek Buch, Ashwin G. Ramayya, Hannes Vogel & Josef Parvizi
Authors
- Julian Quabs
- Sofia Pantis
- Guanpeng Chen
- Weichen Huang
- Eileen Ma
- Maria del Vecchio
- Dian Lyu
- Chloe Wang
- Pietro Avanzini
- Vivek Buch
- Ashwin G. Ramayya
- Svenja Caspers
- Hannes Vogel
- Josef Parvizi
Contributions
J.P. conceived and designed the study, performed all intracranial electrical stimulation mapping procedures at the bedside, wrote the first draft of the manuscript, replied to reviewers and did revisions, and supervised all aspects of the work. J.Q. developed the cytoarchitectonic atlas of the human insula, performed electrode localization and microanatomical labelling, developed the functional clustering and connectivity analysis code, performed most of the analyses, designed and prepared the figures, wrote the manuscript and contributed to its revision. S.C. co-developed the cytoarchitectonic atlas and contributed to anatomical analyses and manuscript revision. D.L. developed and implemented the CCEP methodology and contributed to connectivity analyses. S.P. collected the GradCPT electrophysiology data. G.C contributed to task-based analyses. W.H., E.M. and C.W. contributed to data processing and analysis. M.d.V. and P.A. designed, collected and analysed the somatosensory evoked potential data in the Italian cohort. V.B. and A.G.R. performed neurosurgical electrode implantation procedures and contributed to clinical oversight. H.V. provided clinical and pathological expertise and contributed to study oversight. J.P., A.G.R. and H.V. acquired funding. All of the authors reviewed and approved the final manuscript.
Corresponding author
Correspondence to Josef Parvizi.
Ethics declarations
Competing interests
The authors declare no competing interests.
Peer review
Peer review information
Nature thanks the anonymous reviewers for their contribution to the peer review of this work. Peer reviewer reports are available.
Additional information
Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Extended data figures and tables
Extended Data Fig. 1 Multidimensional scaling of functional response profiles across insular cytoarchitectonic areas.
a, Three-dimensional multidimensional scaling (MDS) visualization of the functional similarity between the cytoarchitectonic areas of the human insula based on their electrically evoked response profiles (somatosensory, pain/temperature, visceral, and no response). Each point represents one cytoarchitectonic area, with distances between points reflecting dissimilarity in functional response profiles. The low stress value (<0.01) indicates a good representation of the original dissimilarities in three-dimensional MDS space. Areas are grouped according to the four functional regions identified by unsupervised k-means clustering (see Methods: Functional Clustering of Insular Areas).
Extended Data Fig. 2 Electrical stimulation parameters across response categories.
Distribution of stimulation current amplitudes (milliamperes, mA) for a) all responsive sites, b) all unresponsive sites, c) responsive sites excluding pain and temperature responses, and d) pain and temperature-responsive sites. The small bars between the larger bars represent the relatively few instances in which currents of x + 0.5 mA were used. Stimulation amplitudes were comparable across response categories, indicating that differences in subjective responses were unlikely to be explained by systematic differences in the stimulation current applied.
Extended Data Fig. 3 Transition between somatosensory and pain/thermal sensations follows granular architecture.
a, In four participants with oblique electrodes traversing the posterior insula, we observed a sharp change in the evoked experience when crossing the border from the ventral posterior dysgranular Region II (somatosensory responses) to the dorsal posterior granular Region I (pain and temperature responses). Notably, the few somatosensory responses observed in Region I (e.g., in participant 1) originated from area Id4 in the anterior insula, which consistently evoked sensory sensations localized to the oral cavity. Slight discrepancies between individual participant borders and the group-averaged cytoarchitectonic boundaries are expected, as the atlas represents the highest-probability location of borders, whereas individual participant anatomy may vary. Of note, electrodes localized close to the anterior insular pole remained unresponsive to stimulation.
Extended Data Fig. 4 Somatotopic organization of the insula assessed by somatosensory evoked potentials.
a) Somatosensory evoked potentials (SEPs) were recorded from 218 insular contacts in 37 participants from an Italian cohort following peripheral stimulation of the mandibular (blue), median (red) and tibial (salmon) nerves (see Methods: Somatosensory Evoked Potentials (SEPs)). Electrodes showing significant responses to peripheral nerve stimulation are displayed according to the stimulated nerve; grey electrodes indicate recording sites without a significant response. Significant responses to stimulation of all three body parts were observed only within granular Region I. b) Radar chart comparing the regional distribution of single-body-part responses between the peripheral nerve stimulation and direct cortical stimulation cohorts. The proportion of contacts showing single-body-part responses was highest in Region I in both cohorts (37/92 contacts, 40.22%, peripheral nerve stimulation; 95/215 contacts, 44.19%, direct cortical stimulation), followed by Region II (12/52 contacts, 23.08%; 38/194 contacts, 19.59%, respectively) and the anterior insula (11/74 contacts, 14.86%; 1/272 contacts, 0.37%, respectively). c) Comparison of anterior–posterior (Y) coordinates of single-body-part responses within Region I between the direct cortical stimulation and peripheral nerve stimulation cohorts. For direct cortical stimulation, n = 18 contacts for mouth, n = 43 contacts for arm and n = 21 contacts for leg responses; for peripheral nerve stimulation, n = 1 contact for trigeminal, n = 30 contacts for median and n = 6 contacts for tibial nerve responses. Boxplots show lower-extremity responses at more posterior locations, upper-extremity responses at intermediate locations and mouth-related responses at more anterior locations. Centre lines indicate medians (50th percentile), box bounds indicate the 25th and 75th percentiles, and whiskers extend to the most extreme observations within 1.5× the interquartile range; observations beyond the whiskers are plotted individually as outliers.
Extended Data Fig. 5 Beck Anxiety Inventory scores according to electrical stimulation responses.
a, Beck Anxiety Inventory (BAI) scores were assessed in participants undergoing electrical stimulation of Region III, where visceral and anxiety responses were evoked. Twenty-four participants were categorized into three groups based on their responses to intracranial electrical stimulation: Anxiety (stimulation evoked anxiety symptoms), Visceral (stimulation evoked visceral sensations without anxiety), and No Response (stimulation evoked no subjective symptoms). BAI scores are separated into cognitive and visceral/somatic components. The Anxiety group exhibited higher scores than the Visceral and No Response groups. Notably, four participants in the Anxiety group showed an asymmetry between cognitive and visceral/somatic scores (cognitive > visceral/somatic or visceral/somatic > cognitive), indicating differences in the relative contribution of the two symptom dimensions.
Extended Data Fig. 6 Anterior insular pole activation during the reversed salience task.
a, To assess whether activation of the anterior insular pole during salience detection and changes in action mode was independent of the required motor response, the task condition was reversed. In the original condition, participants responded to frequently presented city stimuli and withheld responses to infrequent target mountain stimuli. In the reversed condition, a subset of six participants (100 electrodes) withheld responses to frequent city stimuli and responded to infrequent mountain stimuli. The reversed task produced a comparable activation pattern, with predominant activation in and around Region IV.
Extended Data Fig. 7 Statistical validation of response profiles in posterior insular Regions I and II.
To test the anatomical specificity of responses, we compared each region’s response against chance level and then compared regions with significantly greater-than-chance responses against all other regions (see Methods: Validation of Regional Response Profiles). To ensure substantial regional coverage and avoid estimating regional response profiles from sparsely sampled regions, only subjects contributing ≥3 electrode pairs within a given region were included in the analyses. Cortical-stimulation analyses comprised Region I, n = 77.5 pairs of electrodes from 11 participants; Region II, n = 75 pairs of electrodes from 14 participants; Region III, n = 65.5 pairs of electrodes from 13 participants; and Region IV, n = 29.5 pairs of electrodes from 7 participants. One-sided binomial GEE accounted for within-participant clustering; Firth logistic regression was used for complete separation. χ2(1) denotes the chi-square statistic with 1 degree of freedom, OR the odds ratio and LCB the one-sided 95% lower confidence bound. P values were Benjamini–Hochberg FDR-adjusted. Chance, population and observed regional frequencies are shown in blue, dotted green and red, respectively. The statistical validation of response profiles in posterior insular Regions I and II is shown. a) Pain and temperature responses were enriched in Region I relative to chance and all other regions (vs. chance: χ2(1) = 10.58, OR = 4.93, 95% LCB = 2.20, Padj = 0.002396; vs. Region II: χ2(1) = 23.70, OR = 6.48, 95% LCB = 3.45, Padj = 1.126 × 10−6; vs. Region III: χ2(1) = 118.21, OR = 266.40, 95% LCB = 25.30, Padj = 6.224 × 10−27; vs. Region IV: χ2(1) = 63.93, OR = 120.54, 95% LCB = 11.26, Padj = 2.213 × 10−15). b) Region II demonstrated enrichment of somatosensory responses relative to chance and Regions III and IV (vs. chance: χ2(1) = 4.53, OR = 2.56, 95% LCB = 1.24, Padj = 0.04377; vs. Region I: χ2(1) = 0.94, OR = 1.81, 95% LCB = 0.66, Padj = 0.1656; vs. Region III: χ2(1) = 98.56, OR = 206.98, 95% LCB = 19.64, Padj = 9.482 × 10−23; vs. Region IV: χ2(1) = 53.00, OR = 93.65, 95% LCB = 8.74, Padj = 4.998 × 10−13). c) Single-body-part responses were similarly enriched in Region I relative to chance and all other regions (vs. chance: χ2(1) = 16.49, OR = 3.64, 95% LCB = 2.15, Padj = 0.0002564; vs. Region II: χ2(1) = 7.94, OR = 2.95, 95% LCB = 1.57, Padj = 0.002639; vs. Region III: χ2(1) = 122.21, OR = 280.42, 95% LCB = 26.63, Padj = 1.248 × 10−27; vs. Region IV: χ2(1) = 66.21, OR = 126.88, 95% LCB = 11.86, Padj = 8.090 × 10−16). d) Multiple-body-part responses were enriched in Region II relative to chance and all other regions (vs. chance: χ2(1) = 7.18, OR = 2.93, 95% LCB = 1.51, Padj = 0.01108; vs. Region I: χ2(1) = 3.50, OR = 3.61, 95% LCB = 1.17, Padj = 0.03206; vs. Region III: χ2(1) = 20.54, OR = 34.48, 95% LCB = 3.20, Padj = 5.397 × 10−6; vs. Region IV: χ2(1) = 10.12, OR = 15.60, 95% LCB = 1.43, Padj = 0.001037).
Extended Data Fig. 8
To test the anatomical specificity of responses, we compared each region’s response against chance level and then compared regions with significantly greater-than-chance responses against all other regions (see Methods: Validation of Regional Response Profiles). To ensure substantial regional coverage and avoid estimating regional response profiles from sparsely sampled regions, only subjects contributing ≥3 electrode pairs within a given region were included in the analyses. Cortical-stimulation analyses comprised Region I, n = 77.5 pairs of electrodes from 11 participants; Region II, n = 75 pairs of electrodes from 14 participants; Region III, n = 65.5 pairs of electrodes from 13 participants; and Region IV, n = 29.5 pairs of electrodes from 7 participants. One-sided binomial GEE accounted for within-participant clustering; Firth logistic regression was used for complete separation. χ2(1) denotes the chi-square statistic with 1 degree of freedom, OR the odds ratio and LCB the one-sided 95% lower confidence bound. P values were Benjamini–Hochberg FDR-adjusted. Chance, population and observed regional frequencies are shown in blue, dotted green and red, respectively. Statistical validation of response profiles in anterior insular Regions III and IV is shown. a) Region III was characterized by elevated visceral responses relative to chance and all other regions (vs. chance: χ2(1) = 13.21, OR = 4.90, 95% LCB = 2.39, Padj = 0.0007308; vs. Region I: χ2(1) = 51.06, OR = 97.48, 95% LCB = 9.19, Padj = 1.195 × 10−12; vs. Region II: χ2(1) = 8.25, OR = 9.39, 95% LCB = 2.60, Padj = 0.002327; vs. Region IV: χ2(1) = 23.89, OR = 37.30, 95% LCB = 3.45, Padj = 1.112 × 10−6) and b) anxiety-related responses (vs. chance: χ2(1) = 7.88, OR = 5.22, 95% LCB = 1.98, Padj = 0.008773; vs. Region I: χ2(1) = 19.28, OR = 35.43, 95% LCB = 3.26, Padj = 9.692 × 10−6; vs. Region II: χ2(1) = 18.81, OR = 34.29, 95% LCB = 3.16, Padj = 1.156 × 10−5; vs. Region IV: χ2(1) = 8.35, OR = 13.56, 95% LCB = 1.23, Padj = 0.002313). c) Region IV displayed an elevated rate of non-responsive contact sites relative to chance and all other regions (vs. chance: χ2(1) = 83.01, OR = 130.60, 95% LCB = 12.38, Padj = 8.569 × 10−19; vs. Region I: χ2(1) = 133.67, OR = 470.33, 95% LCB = 43.59, Padj = 7.735 × 10−30; vs. Region II: χ2(1) = 80.38, OR = 173.18, 95% LCB = 16.17, Padj = 7.402 × 10−19; vs. Region III: χ2(1) = 38.65, OR = 64.72, 95% LCB = 6.02, Padj = 6.085 × 10−10). d) Activation during the salience detection task was also elevated in Region IV relative to chance and all other regions (vs. chance: χ2(1) = 13.94, OR = 4.82, 95% LCB = 2.41, Padj = 0.0006621; vs. Region I: χ2(1) = 11.25, OR = 15.03, 95% LCB = 3.98, Padj = 0.0005967; vs. Region II: χ2(1) = 9.99, OR = 29.13, 95% LCB = 5.04, Padj = 0.001051; vs. Region III: χ2(1) = 9.69, OR = 3.21, 95% LCB = 1.73, Padj = 0.001167).
Extended Data Fig. 9 Connectivity among insular Regions I–III.
Causal electrophysiological connectivity was assessed using the same cerebro-cerebral evoked potential (CCEP) analyses as in Fig. 4 (see Methods: CCEPs: Within-Insula Connectivity). Connectivity strength was quantified by the F1 feature, reflecting the early high-frequency, phase-locked response to stimulation. For each regional comparison, flat maps show electrode-level connectivity, with each recording electrode colour-coded by its mean F1 connectivity from all stimulation sites in the other region within the same participant. Participant-level plots show mean connectivity in both directions, with the x- and y-axes representing mean F1 connectivity strength in the two opposing directions, respectively. The identity line represents equal bidirectional connectivity, and deviations indicate directional asymmetry. Pair-level spectrotemporal contrasts show significant time–frequency clusters in which connectivity was stronger in one direction than the other. a–c) Connectivity between Regions III and I (n = 254 matched bidirectional CCEP pairs from 30 participants) showed no significant directional asymmetry in F1 connectivity at either the matched-pair (two-sided paired permutation test, 10,000 permutations, BH-FDR-adjusted P = 0.656) or participant level (PFDR = 0.7756), while spectrotemporal analysis revealed stronger late low-frequency connectivity from Region I toward Region III. d–f) Connectivity between Regions III and II (n = 144 matched bidirectional CCEP pairs from 30 participants) showed stronger F1 connectivity from Region III to Region II, significant at both the matched-pair (PFDR < 0.0001) and participant levels (PFDR = 0.0018). Conversely, spectrotemporal analysis revealed stronger late low-frequency connectivity from Region II toward Region III. g–i) Connectivity between Regions II and I (n = 186 matched bidirectional CCEP pairs from 26 participants) showed no significant directional asymmetry in F1 connectivity at either the matched-pair (PFDR = 0.4256) or participant level (PFDR = 0.7756). Cohen’s d effect sizes and 95% confidence intervals for these directional F1 connectivity comparisons are shown in Fig. 4k. Together, these analyses demonstrate that connectivity among Regions I–III depends on both direction and the temporal and spectral characteristics of the evoked response, with posterior Regions I and II showing a potential late low-frequency influence toward Region III.
Rights and permissions
Springer Nature or its licensor (e.g. a society or other partner) holds exclusive rights to this article under a publishing agreement with the author(s) or other rightsholder(s); author self-archiving of the accepted manuscript version of this article is solely governed by the terms of such publishing agreement and applicable law.
About this article
Cite this article
Quabs, J., Pantis, S., Chen, G. et al. Functional architecture of the human insula. Nature (2026). https://doi.org/10.1038/s41586-026-11175-2
Received:
Accepted:
Published:
Version of record:
DOI: https://doi.org/10.1038/s41586-026-11175-2