OSCR

Brain neuromarkers predict self- and other-related mentalizing across adult, clinical, and developmental samples.

Code ↔ Paper

22 matches between paragraphs of the paper and lines of its authors' code, computed by the harvester (lexical-v1). Click a colored paragraph or line to see its counterpart.

The 22 matches · 3 of them tie a paragraph to a whole file, not to given lines: weak matches, whose lines are not tinted
  1. [1] § Methods › Data analysis › Cross-validated training ↔ scripts/matlab functions/train_mentalizing_classifiers.m, the whole file · a weak match · score 0.74 · support vector machine, fold cross validation, mentalizing related, activated, SVM, MS
  2. [2] § Methods › Data analysis › Region-of-interest (ROI) analyses ↔ scripts/MS_fullAnalysisScript.m, lines 371–426 · score 0.74 · mPFC, ROI classifiers, brain classifiers, MTG, analytic, cerebellum
  3. [3] § Methods › Data analysis › Region-of-interest (ROI) analyses ↔ scripts/MS_fullAnalysisScript.m, lines 371–426 · score 0.74 · mPFC, ROI classifiers, brain classifiers, MTG, analytic, cerebellum
  4. [4] § Results › Local patterns of self- and other-related mentalizing ↔ scripts/matlab functions/testing_ROI_classifiers.m, lines 119–256 · score 0.73 · error bars, training accuracies, standard error, validation accuracies, ROI classifier, heights
  5. [5] § Methods › Data analysis › Cross-validated training ↔ scripts/MS_fullAnalysisScript.m, lines 91–123 · score 0.69 · cross validated training, SVM classifiers, prediction accuracies, weight map, unthresholded, bootstrapping
  6. [6] § Methods › Data analysis › Cross-validated training ↔ scripts/MS_fullAnalysisScript.m, lines 91–123 · score 0.69 · cross validated training, SVM classifiers, prediction accuracies, weight map, unthresholded, bootstrapping
  7. [7] § Results › Cross-validated training results ↔ scripts/matlab functions/train_mentalizing_classifiers.m, the whole file · a weak match · score 0.67 · support vector machines, fold cross validation, Referential Signature, Mentalizing Signature, SVM, MS
  8. [8] § Methods › Data analysis › Cross-validated training ↔ scripts/MS_fullAnalysisScript.m, lines 91–123 · score 0.65 · ridge parameter, cross validated, brain signatures, weight map, folds, MS
  9. [9] § Methods › Data analysis › Cross-validated training ↔ scripts/MS_fullAnalysisScript.m, lines 91–123 · score 0.65 · ridge parameter, cross validated, brain signatures, weight map, folds, MS
  10. [10] § Methods › Data analysis › Testing the signatures in extension tasks ↔ scripts/MS_fullAnalysisScript.m, lines 125–240 · score 0.64 · dot products, Binary prediction, ANOVAs, feedback, attributional, linear
  11. [11] § Methods › Data analysis › Testing the signatures in extension tasks ↔ scripts/MS_fullAnalysisScript.m, lines 125–240 · score 0.64 · dot products, Binary prediction, ANOVAs, feedback, attributional, linear
  12. [12] § Methods › Data analysis › Cross-validated training ↔ scripts/MS_fullAnalysisScript.m, lines 54–89 · score 0.63 · NeuroSynth, social cognition, uniformity, downloaded, union, MS
  13. [13] § Methods › Data analysis › Validation in independent datasets › Associations with age ↔ scripts/MS_fullAnalysisScript.m, lines 242–369 · score 0.62 · model formula, linear mixed, age, sex, variable, fit
  14. [14] § Methods › Data analysis › Validation in independent datasets › Associations with age ↔ scripts/MS_fullAnalysisScript.m, lines 242–369 · score 0.62 · model formula, linear mixed, age, sex, variable, fit
  15. [15] § Methods › Data analysis › Validation in independent datasets ↔ scripts/MS_fullAnalysisScript.m, lines 125–240 · score 0.61 · dot product, accuracy statistics, binary, mentalizing signatures, MS, modeling
  16. [16] § Methods › Data analysis › Validation in independent datasets ↔ scripts/MS_fullAnalysisScript.m, lines 125–240 · score 0.61 · dot product, accuracy statistics, binary, mentalizing signatures, MS, modeling
  17. [17] § Results › Local patterns of self- and other-related mentalizing ↔ scripts/MS_fullAnalysisScript.m, lines 371–426 · score 0.60 · mPFC, ROI classifier, prediction accuracy, surface, cerebellum, NeuroSynth
  18. [18] § Methods › Data analysis › Validation in independent datasets › Group comparisons of pattern expression values ↔ scripts/MS_fullAnalysisScript.m, lines 242–369 · score 0.55 · linear mixed, healthy controls, variable, schizophrenia, fits, models
  19. [19] § Methods › Data analysis › Validation in independent datasets › Group comparisons of pattern expression values ↔ scripts/MS_fullAnalysisScript.m, lines 242–369 · score 0.55 · linear mixed, healthy controls, variable, schizophrenia, fits, models
  20. [20] § Results › Cross-validated training results ↔ scripts/matlab functions/svm_rfe_models.m, the whole file · a weak match · score 0.55 · recursive feature elimination, SVM classifiers, RFE, weight maps, Filled, Voxels
  21. [21] § Results › Leave-one-site-out classifiers trained on all datasets ↔ scripts/matlab functions/testing_ROI_classifiers.m, lines 119–256 · score 0.55 · error bars, standard error, validated accuracy, Mentalizing classifier, ROC, trained
  22. [22] § Results › Cross-validated training results ↔ scripts/MS_fullAnalysisScript.m, lines 371–426 · score 0.55 · mPFC, prediction accuracy, MTG, SMA, FDR, precuneus

Paper

Loaded from Europe PMC by your browser, not stored by OSCR: doi.org · Europe PMC

The paper is loaded when this pane is shown.

The authors' code

MATLAB · 511 lines · 24 KB · MIT · 9 matches

  1. %% Replication Code for "Brain neuromarkers predict self- and other-related mentalizing across adult, clinical, and developmental samples"
  2. % This code reproduces all analyses reported in:
  3. % Açıl et al. https://doi.org/10.1101/2025.03.10.642438
  4. %
  5. % DESCRIPTION:
  6. % This repository contains the full analysis pipeline for the manuscript,
  7. % including but not limited to
  8. % (i) cross-validated training four SVM classifiers as mentalizing signatures
  9. % (ii) validation in independent samples
  10. % (iii) developing ROI classifiers
  11. % (v) generating figures for main text and supplementary material.
  12. %
  13. % REQUIREMENTS:
  14. % MATLAB R2022b or above and the following toolboxes:
  15. % - Canlab Core Toolbox [version number, GitHub link]
  16. % - SPM12
  17. % - Statistics and Machine Learning Toolbox
  18. % - Signal Processing Toolbox
  19. % - DataViz toolbox (by Karvelis) for violin plots: https://github.com/povilaskarvelis/DataViz
  20. % - R Software (required for generating only one figure)
  21. %
  22. % INPUT DATA:
  23. % Single-subject contrast images of the training and testing datasets.
  24. % The training dataset is available at: https://doi.org/10.6084/m9.figshare.29908139
  25. % The validation (testing & extension) datasets are not shared publicly.
  26. % Therefore, only the training sections of the code will work with the
  27. % available images. For access at the validation datasets, please reach out to the
  28. % authors.
  29. %
  30. % OUTPUT:
  31. % Running this code will reproduce all analyses and figures as reported in the paper.
  32. %
  33. % CONTACT:
  34. % For questions or issues, please contact Dorukhan Açıl ([email hidden]) or Leonie Koban ([email hidden]).
  35. %
  36. %% Setup
  37. clear all; close all;
  39. %Set up paths
  40. project_path = './projectFolder';
  41. % This path should contain the input data in /inputData
  43. addpath(genpath('./CanlabCore')) % CanlabCore toolbox
  44. addpath(genpath('./spm12')) % SPM12
  45. rmpath(genpath('./CanlabCore/spm12/external/fieldtrip')) % Remove FieldTrip to avoid function conflicts
  47. cd(project_path)
  48. addpath(project_path)
  50. % NOTE: Adjust ./ and project_path directories based on your repo structure.
  52. signatureNames = {'Self_RS', 'Other_RS', 'MS', 'SvO_RS'}';
  54. %% 1. Load and prepare training data
  56. % Download the training images from the Figshare link above
  57. % and store under /inputData
  58. load('inputData/training_contrast_images.mat')
  59. % This contains the unmasked training fmri data storing 63 images. 21
  60. % subjects x 3 conditions. First 21 images belong to the self condition,
  61. % 22th-42th belong to the other-condition, and 43th-63th belong to the
  62. % control condition
  64. %Rescale using l2norm
  65. training_set_unmasked = rescale(training_set_unmasked, 'l2norm_images');
  67. % Create the social cognition mask and apply it on the training images
  68. % Association and uniformity tests masks from NeuroSynth of three terms
  69. % (mentalizing, self-referential, social) are downloaded on 06/06/2024.
  70. % These are stored in /inputData/Social_Mask
  71. fs = filenames('inputData/Social_Mask/*FDR*.nii');
  73. for f = 1:numel(fs) %computing union of the six masks
  74. if f==1
  75. d{f} = fmri_data(fs{f});
  76. social_mask = d{f};
  77. elseif f>1
  78. d{f} = fmri_data(fs{f});
  79. if f>4
  80. d{f} = threshold(d{f}, [0 100], 'raw-between');
  81. end
  82. social_mask = union(social_mask, d{f});
  83. end
  84. end
  86. %orthviews(social_mask) %to visually inspect
  88. training_set_masked = apply_mask(training_set_unmasked, social_mask);
  89. clear fs f d
  91. %% %% % Training Mentalizing Signatures (i) Self-RS, ii) Other-RS, iii) Mentalizing-RS, iv) Self-vs-Other RS
  92. % Input images are from n=21 participants performing a trait-evaluation task with three conditions: Self, Other, Control.
  93. % 1) Self-Referential Signature (Self-RS) – self-condition vs. other two conditions
  94. % 2) Other-Referential Signature (Other-RS) – other-condition vs. the other two conditions
  95. % 3) Mentalizing Signature (Mentalizing-RS) – mentalizing (self+other) vs. control condition
  96. % 4) Self-vs-Other (SvO-RS) – direct contrast between self- vs. other-related mentalizing
  97. %
  98. % Outputs: Uncorrected weight maps that are effectively the brain signatures of mentalizing,
  99. % statistics of the cross-validated training stage, and bootstrapped
  100. % thresholded images used for visualization
  102. % Train bootstrapped SVM classifiers
  103. % Using 10-fold CV, 5000 bootstraps for weight stability, and .5 ridge parameter for class balancing.
  105. training_stats = train_mentalizing_classifiers(training_set_masked, 1);
  106. % Second argument requires a logical input to enable or disable bootstrapping
  108. % Here define the mentalizing signatures
  109. for i = 1:numel(fieldnames(training_stats))
  110. fn = fieldnames(training_stats);
  111. mentalizing_signatures.(signatureNames{i}) = training_stats.(fn{i}).weight_obj; clear fn
  112. end
  114. % Compute prediction accuracies and visualize the training ROC plots
  115. % The output is displayed in the command window and also stored in the training_results struct.
  117. results.trainingPredictions = training_ROC(training_stats, signatureNames, 21, []); %third argument is the sample size, fourth argument defines the true and false classes
  119. % Visualize the weight maps
  120. %this will plot the unthresholded and thresholded weight maps of the classifiers in separate windows
  121. %requires statistical weight map as an output of bootstrapping
  123. figure_brain(mentalizing_signatures, signatureNames);
  125. %% %% %% Validating Mentalizing Signatures in 6 Independent Datasets
  127. % % Validation datasets:
  128. % Study 2-5 are testing datasets. Study 6-7 are extension datasets.
  129. study_names = {'Study2', 'Study3', 'Study4a', 'Study4b', 'Study5a', 'Study5b', 'Study5c', 'Study6a', 'Study6b', 'Study7'};
  131. % Load validation datasets - Note: these are not shared publicly
  132. validation_sets = {'DAT_Study2.mat', 'DAT_Study3.mat', ...
  133. 'DAT_Study4a.mat', 'DAT_Study4b.mat', ...
  134. 'DAT_Study5a.mat', 'DAT_Study5b.mat', 'DAT_Study5c.mat',...
  135. 'DAT_Study6a.mat', 'DAT_Study6b.mat', 'DAT_Study_7.mat'}; %not publicly available
  136. % The testing and extension datasets are stored in /inputData in standard
  137. % format where conditions are fields of the struct .dat, each of which contains
  138. % single-subject contrast images for the condition.
  140. full_paths = fullfile(['/inputData/', validation_sets]);
  141. %full_paths should store the paths as characters that will be used to load
  142. %the images in next steps. load(full_paths{1}) should load struct called
  143. %'dat' with three fields (per condition) each corresponding to an fmri_data object that
  144. %stores images of the respective condition.
  145. clear validation_sets path_validation_data
  147. % Applying the signatures (weight maps) onto the testing and extension datasets
  148. % Here we compute pattern expression values by taking the dot products of
  149. % four mentalizing classifiers and single-subject condition contrast
  150. % images from each study. Validation images are first rescaled using
  151. % l2-norm, then resampled onto the same image space as the classifiers.
  153. pexp = pattern_expressions(mentalizing_signatures, full_paths, study_names);
  154. % First argument contains the classifiers, second argument must indicate
  155. % the paths to the validation datasets, and the third argument stores
  156. % names for these datasets
  158. % Testing Accuracies using ROC and Accuracy Plots
  160. % i) Testing datasets
  162. studies = {'Study2', 'Study3', 'Study4a', 'Study4b', 'Study5a', 'Study5b', 'Study5c'};
  163. %only Study 2-5 are testing signatures. Beware that the color and shape
  164. %codes are dependent on the study_names, so if they are changed, it might
  165. %throw an error.
  167. results.testingPredictions = testing_signatures(pexp, studies, [], []);
  168. %this will plot ROC and accuracy plots and store accuracy statistics in the
  169. %"testing_results" structs along with average accuracy values
  170. %3rd and 4th arguments define true and false classes which are also defined
  171. %in the function
  174. % ii) Extension datasets
  175. % a) Study 7 in which signatures were tested using linear effects models
  177. results.extension.stat_study7 = rmAnova_extension(pexp.Study7);
  179. % Violin Plots for illustrating pattern expressions by condition
  180. conditions = {'Feedback on Self', 'Feedback on Partner', 'Control Condition'};
  181. plot_data = pexp.Study7;
  182. num_classifiers = 3; % plot in order Self-RS, Other-RS, MS
  183. plot_asterisks = 1; %beware that the function below is not coded to adapt to plot a different dataset
  184. % in that case, turn this to 0
  185. violPlot_pexp(plot_data, conditions, num_classifiers, plot_asterisks);
  187. % b) Study 6 in which signatures predictions are tested using binary prediction tests
  188. % ROC plots are produced. This dataset has only two conditions or a
  189. % 2x2 design, so it is structurally different than other datasets
  191. % Input
  192. %pexp struct with Study6a and Study6b fields must be there
  193. col = [.73 .01 .32; .07 .6 1]; %colors for the lines
  194. maps_to_plot = {'othermap', 'mentmap'}; %this indicates that only Other-RS and MS should be tested.
  196. figure; %Create plots
  197. for w = 1:2
  199. pexp_vals = pexp.Study6a.(maps_to_plot{w});
  201. subplot(2,2,w);
  202. true_con = 'attributional'; false_con = 'factual'; n = numel(pexp.Study6a.(maps_to_plot{1}).attributional);
  204. ROC_6a = roc_plot([pexp_vals.attributional; pexp_vals.factual], [ones(n,1); -ones(n,1)]==1, 'twochoice', 'color', col(1,:)); hold on
  205. title('Attributional vs Factual');
  207. %plot features
  208. ROC_6a.line_handle(1).Marker = 'v'; ROC_6a.line_handle(1).MarkerSize = 6;
  209. ROC_6a.line_handle(2).LineWidth = 2; set(gca, 'FontSize', 12);
  210. acc = num2str(ROC_6a.accuracy, '%.2f');
  211. acc_str = num2str(str2double(acc)*100);
  212. text(0.95, .4, 'Accuracies', 'Units', 'normalized', 'HorizontalAlignment', 'right', 'VerticalAlignment', 'top', 'Color', 'k', 'FontSize', 12, 'FontWeight', 'bold')
  213. text(0.95, .3, ['Social: %', acc_str], 'Units', 'normalized', 'HorizontalAlignment', 'right', 'VerticalAlignment', 'top', 'Color', col(1,:), 'FontSize', 12, 'FontWeight', 'bold')
  215. subplot(2,2,w+2);
  216. pexp_vals = pexp.Study6b.(maps_to_plot{w})
  217. true_con = {'attrib_social', 'attrib_nonsocial'}; false_con = {'fact_social', 'fact_nonsocial'}; n = numel(pexp.Study6b.(maps_to_plot{1}).attrib_social);
  219. ROC_6b1 = roc_plot([pexp_vals.(true_con{1}); pexp_vals.(false_con{1})], [ones(n,1); -ones(n,1)]==1, 'twochoice', 'color', col(1,:)); hold on
  220. ROC_6b2 = roc_plot([pexp_vals.(true_con{2}); pexp_vals.(false_con{2})], [ones(n,1); -ones(n,1)]==1, 'twochoice', 'color', col(2,:)); hold on
  221. title('Attributional vs Factual');
  223. %plot features
  224. ROC_6b1.line_handle(1).Marker = 'v'; ROC_6b2.line_handle(1).Marker = '*'; ROC_6b2.line_handle(2).LineStyle = ':';
  225. ROC_6b1.line_handle(1).MarkerSize = 6; ROC_6b2.line_handle(1).MarkerSize = 6;
  226. ROC_6b1.line_handle(2).LineWidth = 2; ROC_6b2.line_handle(2).LineWidth = 2;
  228. set(gca, 'FontSize', 12);
  229. acc = num2str(ROC_6b1.accuracy, '%.2f'); acc2 = num2str(ROC_6b2.accuracy, '%.2f');
  230. acc_str = num2str(str2double(acc)*100); acc_str2 = num2str(str2double(acc2)*100);
  231. text(0.95, .4, 'Accuracies', 'Units', 'normalized', 'HorizontalAlignment', 'right', 'VerticalAlignment', 'top', 'Color', 'k', 'FontSize', 12, 'FontWeight', 'bold')
  232. text(0.95, .3, ['Social: %', acc_str], 'Units', 'normalized', 'HorizontalAlignment', 'right', 'VerticalAlignment', 'top', 'Color', col(1,:), 'FontSize', 12, 'FontWeight', 'bold')
  233. text(0.95, .2, ['Nonsocial: %', acc_str2], 'Units', 'normalized', 'HorizontalAlignment', 'right', 'VerticalAlignment', 'top', 'Color', col(2,:), 'FontSize', 12, 'FontWeight', 'bold')
  235. end; set(gcf, 'Position', [750 150 530 450]);
  236. results.extension.ROC_6a = ROC_6a; results.extension.ROC_6b1 = ROC_6b1; results.extension.ROC_6b2 = ROC_6b2; clear ROC_6a ROC_6b1 ROC_6b2
  237. %ROC_6a, ROC_6b1, and ROC_6b2 fields store the ROC output for each comparison
  239. clear acc acc2 acc_str acc_str2 ans col conditions false_con maps_to_plot n num_classifiers ...
  240. pexp_vals plot_asterisks plot_data true_con w i
  242. %% %% %% Testing for Individual Differences (Clinical status and Age)
  243. % These analyses test whether clinical status and age are associated with
  244. % the degree of the signatures' separatability of the Self and Other
  245. % conditions.
  247. % i) Testing for sex effects in overall pattern expressions
  249. % Organize datasets - including only testing datasets each with three
  250. % conditions
  251. studies = {'Study2', 'Study3', 'Study4a', 'Study4b', 'Study5a', 'Study5b', 'Study5c'};
  252. combinedData = prepare_dataset(pexp, studies, [], []);
  254. demographics = readtable('inputData/Demographics_ALL.xlsx'); %This is not shared publicly due to data protection
  256. maps = fieldnames(combinedData);
  257. studies = {'Study2', 'Study3', 'Study4', 'Study5'};
  258. indexes = [{1:44}, {45:105}, {106:161}, {162:211}]; %using fixed indexes as they are concatenated in this order
  260. for m = 1:numel(maps) % merge covariates and the pattern fit data
  261. for i = 1:numel(studies)
  262. n = sum(contains(demographics.Study,studies{i}));
  263. combinedData.(maps{m})(indexes{i},7) = ...
  264. demographics(contains(demographics.Study,studies{i}),2); %get Sex info
  265. end
  266. combinedData.(maps{m}).Properties.VariableNames(7) = {'sex'};
  267. combinedData.(maps{m}).sex = categorical(combinedData.(maps{m}).sex);
  268. end
  270. % linear mixed effects model with age and sex as predictor variables
  271. model_formula = 'fit ~ sex + (1|study)'; clear lme
  272. for m = 1:numel(maps) %run the model for each map
  273. results.sexEffect.(maps{m}) = fitlme(combinedData.(maps{m}), model_formula);
  274. end
  277. % ii) Clinical status
  278. % a) Healthy controls vs Schizophrenia patients - LME models
  280. % Organize datasets
  281. studies = {'Study4a', 'Study4b', 'Study5a', 'Study5b'};
  282. ref_group = 'a'; ref_study = '4';
  283. HCvsSZ = prepare_dataset(pexp, studies, ref_group, ref_study);
  285. % linear mixed effects model
  286. model_formula = 'fit ~ group + (1|study)'; %LME for SZ vs HC
  287. maps = fieldnames(HCvsSZ);
  288. for m = 1:numel(maps) %run the model for each map
  289. results.indDiff.HCvsSZ.(maps{m}) = fitlme(HCvsSZ.(maps{m}), model_formula);
  290. disp('-------'); disp('----');
  291. disp(['LME RESULTS FOR: ', maps{m}])
  292. disp(results.indDiff.HCvsSZ.(maps{m})); disp('----'); disp('-------');
  293. end
  295. % Plot pexp distributions
  296. violPlot_pexp2(HCvsSZ)
  299. % b) Healthy controls vs Bipolar patients - t-test
  301. studies = {'Study5a', 'Study5c'};
  302. ref_group = 'a'; ref_study = '5';
  303. output_HCvsBP = prepare_dataset(pexp, studies, ref_group, ref_study);
  305. % two-sample t-test for HC vs BP comparison
  306. for m = 1:numel(maps)
  307. dat = output_HCvsBP.(maps{i});
  308. [results.indDiff.HCvsBP.h results.indDiff.HCvsBP.p results.indDiff.HCvsBP.ci ...
  309. results.indDiff.HCvsBP.stats] = ttest2(dat.fit(dat.group == '1'), dat.fit(dat.group == '0'), 'Vartype', 'unequal');
  310. if h == 1
  311. disp(maps);
  312. disp(p); disp(stats);
  313. end
  314. end
  317. % iii) Age
  319. studies = {'Study2', 'Study3'};
  320. ref_group = 'not needed'; ref_study = '2';
  321. output_dev = prepare_dataset(pexp, studies, ref_group, ref_study);
  323. ages = readtable('inputData/developmentalSamples_age.xlsx'); %load age info
  325. %run LME models
  326. maps = fieldnames(output_dev);
  327. model_formula = 'fit ~ ages + sex + (1|study)';
  328. clc;
  329. for k = 1:numel(maps)
  330. output_dev.(maps{k}).ages = ages{:,1};
  331. output_dev.(maps{k})(find(output_dev.(maps{k}).study == '1'), 8) = ...
  332. demographics(contains(demographics.Study,'Study2'),"Sex");
  333. output_dev.(maps{k})(find(output_dev.(maps{k}).study == '0'), 8) = ...
  334. demographics(contains(demographics.Study,'Study3'),"Sex");
  335. output_dev.(maps{k}).Properties.VariableNames(8) = {'sex'};
  336. results.indDiff.age.(maps{k}) = fitlme(output_dev.(maps{k}), model_formula);
  337. if double(results.indDiff.age.(maps{k}).Coefficients(2,6)) < .05
  338. disp('-----------')
  339. disp(maps{k})
  340. disp(results.indDiff.age.(maps{k}));
  341. end
  342. end
  344. % Plots for the Age effects
  345. % These are produced in R. So we are first exporting the data for R
  346. % access, which we'll use to produce Figures in R via system command,
  347. % and then illustrate them here as images
  349. % Save data for R
  350. dat_table = table(output_dev.selfmap.fit, output_dev.othermap.fit, output_dev.SvOmap.fit, ...
  351. output_dev.selfmap.ages, output_dev.selfmap.study);
  352. dat_table.Properties.VariableNames = {'self_fit', 'other_fit', 'SvO_fit', 'ages', 'sample'};
  353. writetable(dat_table, 'temp/MapFitsAge.xlsx');
  355. %make sure that the R is in the path
  356. addpath(genpath('/Library/Frameworks/R.framework/Resources')); %this is an example path - it should be updated
  357. % This R script should save three figures in temp/Fig_Age. Make sure to
  358. % change the working directory in the R script (figure_Age.R)to the /ProjectFolder
  359. system('/Library/Frameworks/R.framework/Resources/bin/Rscript scripts/figure_Age.R'); %update the path to the R
  361. % Display figures in matlab
  362. img1 = imread('temp/Age_SelfRS.png');
  363. img2 = imread('temp/Age_OtherRS.png');
  364. img3 = imread('temp/Age_SvORS.png');
  365. subplot(1, 3, 1); imshow(img1); subplot(1, 3, 2); imshow(img2); subplot(1, 3, 3); imshow(img3);
  366. set(gcf, 'Position', [1, 299, 1280, 420]);
  368. clear ages ans combinedData dat dat_table demographics i img1 img2 img3 indexes k m maps model_formula n ...
  369. other_fit output_dev output_HCvsBP HCvsSZ ref_group ref_study sample self_fit studies SvO_fit
  371. %% %% %% ROI Analyses
  372. % Training ROI classifiers instead of brain-wide /whole-brain classifiers
  373. % using the same analytic strategy as above
  375. % 1. Extract and create ROI masks from the Neurosynth (NS) mask
  376. source_mask = 'inputData/mentalizing_ass-test_z_FDR_0.01.nii';
  378. % this function will select ROIs with at least 200 voxels, and ask the user
  379. % to assign a name to save this image to the output folder.
  380. % The output is saved in inputData/ROIs
  381. % This step could be skipped because images are provided in inputData/ROIs
  382. extract_ROI(source_mask, training_set_masked)
  384. % 2. Train predictive maps for each ROI
  385. path4ROImasks = 'inputData/ROIs';
  386. results.ROI_training = train_ROI_classifiers(path4ROImasks, training_set_masked);
  387. % MultiClass field contains results from three classifiers (in each column
  388. % of relevant objects): Self classifiers, other classifiers, and
  389. % mentalizing classifiers. ROC field contains training results
  391. % 3. Validate the ROI classifiers in the testing datasets and display
  392. % prediction accuracies both in training and testing datasets
  393. testing_sets = {'DAT_Study2.mat', 'DAT_Study3.mat', ...
  394. 'DAT_Study4a.mat', 'DAT_Study4b.mat', ...
  395. 'DAT_Study5a.mat', 'DAT_Study5b.mat', 'DAT_Study5c.mat'};
  397. results.ROI_testing = testing_ROI_classifiers(results.ROI_training, testing_sets, 1);
  398. % last argument is logical input argument to either make (1) or not make
  399. % the figure (0)
  402. % 4. Display weights of each ROI as Self-vs-Other Classifiers
  403. % beware that, if ROIs are recreated by assigning a user-given name above,
  404. % these names must be updated accordingly
  405. regs_to_plot_weights = {'aMTG_L', 'aMTG_R', 'mPFC', 'TPJ_R', 'TPJ_L', 'Precuneus'};
  406. %SvO classifiers of these regions could predict self vs other both in
  407. %the training and testing datasets - so we are illustrating their weights
  408. regs_to_plot_green = {'Cerebellum_R', 'Cerebellum_L', 'Cerebellum_BL', 'SMA_L'};
  409. %SvO classifiers of these regions weren't successfuly in separating two
  410. %conditions - we are not illustrating their weights
  412. for m = 1:numel(regs_to_plot_weights)
  413. figure; % this surface function may crash in matlab versions after 2022b. It requires a fix in render_on_surface.m function that it refers.
  414. surface_handles = surface(results.ROI_training.stats_ROI_SvO.(regs_to_plot_weights{m}).weight_obj, 'foursurfaces', 'noverbose');
  415. text(0, 1, regs_to_plot_weights{m}, 'FontSize', 20, 'HorizontalAlignment', 'center', 'Interpreter', 'none');
  416. set(gcf,'Position', [5, 208, 807, 420]);
  417. end
  418. for m = 1:numel(regs_to_plot_green)
  419. figure;
  420. reg_col = colormap_tor([0.10 0.82 0.0039], [0.10 0.82 0.0039]);
  421. surface_handles = surface(results.ROI_training.stats_ROI_SvO.(regs_to_plot_green{m}).weight_obj, 'foursurfaces', 'noverbose', 'pos_colormap', reg_col, 'neg_colormap', reg_col);
  422. text(0, 1, regs_to_plot_green{m}, 'FontSize', 20, 'HorizontalAlignment', 'center', 'Interpreter', 'none');
  423. set(gcf,'Position', [5, 208, 807, 420]);
  424. end
  426. clear regs_to_plot_weights regs_to_plot_green
  428. %% %% %% Leave-One-Site-Out (LOSO) Classifiers
  429. % Training leave-one-site-out classifiers combining all datasets that used
  430. % the same or similar task with three conditions: Study 1-5
  432. % 1. Gather data together
  433. testing_sets = {'DAT_Study2.mat', 'DAT_Study3.mat', ...
  434. 'DAT_Study4a.mat', 'DAT_Study4b.mat', ...
  435. 'DAT_Study5a.mat', 'DAT_Study5b.mat', 'DAT_Study5c.mat'};
  436. ref = fmri_data; %reference image to resample onto
  437. conds = {'self','other','control'};
  439. for i = 1:numel(testing_sets) %Merge all data together
  441. load(testing_sets{i});
  442. fn = fieldnames(dat);
  444. for v = 1:numel(fn)
  445. if i == 1 %Create template images
  446. datLOSO.(conds{v}) = fmri_data;
  447. end
  449. dat.(fn{v}) = resample_space(dat.(fn{v}),ref); %Resample space of validation datasets to reference image
  451. datLOSO.(conds{v}) = cat(datLOSO.(conds{v}), dat.(fn{v})); %Concatenate images
  452. end
  453. clear dat
  454. end
  456. training_set_unmasked_LOSO = resample_space(training_set_unmasked,datLOSO.self);
  458. for v = 1:numel(conds) % Include the training dataset
  459. datLOSO.(conds{v}).dat(:,1) = []; %delete the first image which was the reference image
  460. datLOSO.(conds{v}) = rescale(datLOSO.(conds{v}), 'l2norm_images'); %L2norm normalization
  461. datLOSO.(conds{v}) = cat(datLOSO.(conds{v}), get_wh_image(training_set_unmasked_LOSO,(v-1)*21+1:21*v));
  462. end
  464. LOSO_set = cat(datLOSO.self, datLOSO.other, datLOSO.control); clear datLOSO ref
  465. LOSO_set.removed_images = zeros(696,1); % correction to prevent crashes
  466. LOSO_set = apply_mask(LOSO_set,social_mask); %apply the social cognition mask
  468. % 2. Train SVM classifiers using 5 folds with each study in a single fold
  469. sites = {'Study2', 'Study3', 'Study4', 'Study5', 'Study1'}; % in the order that they appear in LOSO_set dataset
  470. sampleSizes = [44, 61, 56, 50, 21];
  472. results.LOSO_classifiers = train_LOSO_classifiers(LOSO_set, sites, sampleSizes);
  473. % Second argument lists the studies in the order they were concatenated in
  474. % the LOSO_set fmri_dataset. Third argument defines the sample size of each study
  476. clear conds fn i output_folder path4ROImasks v sampleSizes sites training_set_unmasked_LOSO
  478. %% %% %% Supplementary Tables and Figures
  480. % i) Label tables that show significant clusters in the thresholded maps of
  481. % mentalizing signatures
  483. cluster_labels = get_cluster_labels(training_stats, signatureNames);
  485. % ii) Whole-brain results - Train and test new classifiers using the
  486. % unmasked training dataset, or in other words that masked with whole-brain
  487. % gray matter mask.
  489. clasf_wBrain = train_mentalizing_classifiers(training_set_unmasked, 1);
  491. %Training ROCs
  492. results.wBrain_training = training_ROC(clasf_wBrain, signatureNames, 21, []);
  494. %Test the whole brain classifiers in the testing datasets
  495. pexp_wB = pattern_expressions(clasf_wBrain, full_paths, study_names);
  497. % Testing state ROC and Accuracy Plots
  498. study_names = {'Study2', 'Study3', 'Study4a', 'Study4b', 'Study5a', 'Study5b', 'Study5c'};
  499. results.wBrain_testing = testing_signatures(pexp_wB, study_names, [], []);
  500. %this will plot ROC and accuracy plots and store accuracy statistics in the
  501. %"testing_results" structs along with average accuracy values
  503. % Brain images
  504. figure_brain(clasf_wBrain, signatureNames);
  506. % ii) RFE images - Compute new SVM models using recursive feature
  507. % elimination with final n = 5000
  508. map_names = {'Self', 'Other', 'Mentalizing', 'SelfvsOther'};
  509. results.svm_rfe = svm_rfe_models(training_set_masked, map_names);

MS_fullAnalysisScript.m at commit d9ec388, under MIT · at the source

Overview

Authors: Dorukhan Açıl1,2,3, Jessica R Andrews-Hanna4,5, Marina López-Solà6,7, Mariët van Buuren8, Lydia Krabbendam8, Liwen Zhang9, Lisette van der Meer10,11, Paola Fuentes-Claramonte12,13, Edith Pomarol-Clotet12,13, Raymond Salvador12,13, Martin Debbané14,15, Pascal Vrticka16, Patrik Vuilleumier17,18, David A Sbarra4, Andrea M Coppola4, Anita Tusche19,20, Lars O White3, Tor D Wager21, Leonie Koban22,23
ORCID iDs: Dorukhan Açıl, Jessica R Andrews-Hanna, Mariët van Buuren, Lydia Krabbendam, Liwen Zhang, Paola Fuentes-Claramonte, Martin Debbané, Pascal Vrticka, Patrik Vuilleumier, Andrea M Coppola, Anita Tusche, Tor D Wager, Leonie Koban
23 affiliations
  1. Max Planck Institute for Human Cognitive and Brain Sciences, Leipzig, Germany
  2. Department of Child and Adolescent Psychiatry, Psychotherapy, and Psychosomatics, Leipzig University, Leipzig, Germany
  3. Department of Clinical Child and Adolescent Psychology and Psychotherapy, University of Bremen, Bremen, Germany
  4. Department of Psychology, University of Arizona, Tucson, AZ USA
  5. Cognitive Science, University of Arizona, Tucson, AZ USA
  6. Department of Medicine, School of Medicine and Health Sciences, Institute of Neurosciences, University of Barcelona, Barcelona, Spain
  7. Institut d’Investigacions Biomèdiques August Pi i Sunyer (IDIBAPS), Barcelona, Spain
  8. Department of Clinical, Neuro and Developmental Psychology, Faculty of Behavioral and Movement Sciences, Institute for Brain and Behavior Amsterdam, Vrije Universiteit Amsterdam, Amsterdam, The Netherlands
  9. Institute for Medical Imaging Technology, Ruijin Hospital, Shanghai Jiao Tong University School of Medicine, Shanghai, China
  10. Department of Clinical and Developmental Neuropsychology, University of Groningen, Groningen, The Netherlands
  11. Department of Psychiatric Rehabilitation, Lentis Zuidlaren, Zuidlaren, The Netherlands
  12. FIDMAG Germanes Hospitalàries Research Foundation, Barcelona, Spain
  13. Centro de Investigación Biomédica en Red de Salud Mental (CIBERSAM) ISCIII, Barcelona, Spain
  14. Developmental Clinical Psychology Research Unit, Faculty of Psychology and Educational Sciences, University of Geneva, Geneva, Switzerland
  15. Research Department of Clinical, Educational and Health Psychology, University College London, London, UK
  16. Social Neuroscience of Human Attachment (SoNeAt) Lab, Department of Psychology, University of Essex, Colchester, UK
  17. Laboratory of Behavioural Neurology and Imaging of Cognition, Department of Neuroscience, University Medical Center, University of Geneva, Geneva, Switzerland
  18. Swiss Center for Affective Sciences, University of Geneva, Geneva, Switzerland
  19. Department of Psychology, Queen’s University, Kingston, ON Canada
  20. Center for Neuroscience Studies, Queen’s University, Kingston, ON Canada
  21. Department of Psychological and Brain Sciences, Dartmouth College, Hanover, NH USA
  22. Lyon Neuroscience Research Center (CRNL), CNRS, Inserm, Université Claude Bernard Lyon 1, Bron, France
  23. Le Vinatier Psychiatrie Universitaire Lyon Métropole, Bron, France
Institutions: University of Bremen (Germany); Max Planck Institute for Human Cognitive and Brain Sciences (Germany); Leipzig University (Germany); University of Arizona (United States); Consorci Institut D'Investigacions Biomediques August Pi I Sunyer (Spain); Universitat de Barcelona (Spain); Vrije Universiteit Amsterdam (Netherlands); Shanghai Jiao Tong University (China); Ruijin Hospital (China); University of Groningen (Netherlands); Lentis (Netherlands); Instituto de Salud Carlos III (Spain); Centro de Investigación Biomédica en Red de Salud Mental (Spain); Fidmag Sisters Hospitallers (Spain); University of Geneva (Switzerland); University College London (United Kingdom); University of Essex (United Kingdom); Queen's University (Canada); Dartmouth College (United States); Lyon 1 Université (France); Centre National de la Recherche Scientifique (France); Inserm (France); Centre de Recherche en Neurosciences de Lyon (France)
Journal: Nature communications, volume 17, issue 1, article 7229
Dates: received 20 March 2025; accepted 20 May 2026; published online 6 June 2026
Type: Research article · Language: English
License: CC BY
Identifiers: DOI 10.1038/s41467-026-73945-w · PMID 42251044 · PMCID PMC13396484 · OpenAlex W4408404455
Open access: gold, a free copy (OpenAlex)
Preprint: doi:10.1101/2025.03.10.642438
Status: code verified
Categories: human (organism), schizophrenia / psychosis (population)
Methods: Connectivity, Statistics, Machine learning, fMRI & imaging, Spectral & time-frequency
Keywords: Social neuroscience, Neural decoding
MeSH: Brain*, Mentalization*, Theory of Mind*, Adolescent, Adult, Biomarkers, Brain Mapping, Female, Humans, Machine Learning, Magnetic Resonance Imaging, Male, Middle Aged, Schizophrenia, Young Adult (* major topic)
Topic: Personality Disorders and Psychopathology (Clinical Psychology, Psychology), according to OpenAlex
Funding: European Research Council (648082, 101041087); NIMH NIH HHS (R37 MH076136, R01 MH116026, R01 MH125414, P50 MH094258); NIBIB NIH HHS (R01 EB026549)
Citations: cited by 1 paper (Europe PMC); 96 references in the paper

Abstract

Human social interactions rely on the ability to reflect on one’s own and others’ internal states and traits—a process known as mentalizing. Impaired or altered mentalizing is a hallmark of multiple psychiatric and neurodevelopmental conditions. Yet, replicable and easily testable brain markers of mentalizing have so far been lacking. Here, we apply an interpretable machine learning approach to multiple datasets (total n = 390) to train and validate fMRI brain signatures that predict i) mentalizing about the self, ii) mentalizing about another person, and iii) both types of mentalizing. Self-mentalizing and other-mentalizing classifiers had positive weights in anterior/medial and posterior/lateral brain areas, respectively, with accuracy rates of 82% and 77% for out-of-sample prediction. The classifier trained across both types of mentalizing showed 98% predictive accuracy and separated (mental) attributional from factual inferences. Classifier patterns revealed better self/other separation in healthy adults compared to individuals with schizophrenia and with increasing age in adolescence. Together, our findings reveal consistent and separable neural patterns subserving trait-based mentalizing about self and others—present at least from the age of adolescence and functionally altered in severe neuropsychiatric disorders. These mentalizing signatures hold promise as candidate neuromarkers of social-cognitive processes in different contexts and clinical conditions.

Reproduced under the paper's license (CC BY), from the paper cited above.

Repositories

Its files are read in the Code ↔ Paper reader above, with 22 matches between paragraphs and lines of code.

ldmk/2025_MentalizingSignatures

License: MIT
State: the link answers, verified on 27 September 2026
Evidence: files inventoried
Commit: d9ec38844cff0d3031358e667ac7f22e8ef8927e, 13 May 2026
Languages: MATLAB (17), R (1)
Size: 57 files, 18 scripts
Software Heritage: not archived
Found in: “Code availability”
Holds: README, license file
Not found: CITATION.cff, environment file, tests, continuous integration, documentation
Tools: Statistics and Machine Learning Toolbox (3 files), SPM (1 file)
Availability: 1 check, the latest on 27 September 2026: the link answers
  • 27 September 2026: the link answers
20 files
At the source: github.com/ldmk/2025_MentalizingSignatures

Zenodo 19461625

License: MIT
State: the link answers, verified on 27 September 2026
Evidence: files inventoried
Size: 1 file
Software Heritage: not checked
Found in: the references
Not found: README, license file, CITATION.cff, environment file, tests, continuous integration, documentation
Tools: Statistics and Machine Learning Toolbox (3 files), SPM (1 file)
Availability: 1 check, the latest on 27 September 2026: the link answers (HTTP 200)
  • 27 September 2026: the link answers (HTTP 200)
20 files
At the source:

Code availability

Mentalizing signatures for use in future studies, as well as custom MATLAB scripts for analyses are available at95: https://github.com/ldmk/2025_MentalizingSignatures.

Reproduced under the paper's license (CC BY), from the paper cited above.

Tracing map

Proposed by the machine: these links were found in the paper and verified at the source, without human review. The map will receive a Zenodo DOI once one of the paper's authors has validated it with their ORCID.

What the map holds:

  • 2 repositories of the authors' code, each at its verified commit, with its license and how the link was found in the paper;
  • 36 scripts, each with its path and the digest of its content;
  • 22 matches between paragraphs of the paper and lines of the code (method lexical-v1);
  • neither the text of the paper nor the code itself.

Its JSON (tracing-map.json) is deposited on Zenodo with its DOI once the map is validated.

Data

Datasets cited

Data availability

Source data for all figures and single-subject contrast images of the training dataset (Study 1) are available at94 10.6084/m9.figshare.29908139. Contrast images from Study 4a can be accessed via OpenNeuro (https://openneuro.org/datasets/ds001618/versions/1.0.1), and data from Study 6 are available through the NIHM Archive (https://nda.nih.gov/edit_collection.html?id=2643). Imaging data from five contributing studies (Studies 2, 3, 4b, 5, and 7) are not shared online for privacy reasons; however, deidentified data from these studies can be made available in line with applicable data protection and privacy regulations upon request to the corresponding authors.

Reproduced under the paper's license (CC BY), from the paper cited above.

Versions

The history of this record: each version stored by the harvester or made by a correction of its authors or of the maintainers of its code, and what changed in its facts. The texts of the paper (its abstract, its availability statements) are not part of it; versions that changed only those are not listed.

Version 1, 27 September 2026: the first record

Recorded: type, language, journal, volume, issue, pages, dates, 19 authors, 2 keywords, 15 MeSH terms, 3 funders, 83 references.

Cite

This paper

Açıl, D., Andrews-Hanna, J. R., López-Solà, M., van Buuren, M., Krabbendam, L., Zhang, L., van der Meer, L., Fuentes-Claramonte, P., Pomarol-Clotet, E., Salvador, R., Debbané, M., Vrticka, P., Vuilleumier, P., Sbarra, D. A., Coppola, A. M., Tusche, A., White, L. O., Wager, T. D., & Koban, L. (2026). Brain neuromarkers predict self- and other-related mentalizing across adult, clinical, and developmental samples. Nature communications, 17(1), 7229. https://doi.org/10.1038/s41467-026-73945-w

BibTeX

@article{acl2026brain,
author = {Açıl, Dorukhan and Andrews-Hanna, Jessica R and López-Solà, Marina and van Buuren, Mariët and Krabbendam, Lydia and Zhang, Liwen and van der Meer, Lisette and Fuentes-Claramonte, Paola and Pomarol-Clotet, Edith and Salvador, Raymond and Debbané, Martin and Vrticka, Pascal and Vuilleumier, Patrik and Sbarra, David A and Coppola, Andrea M and Tusche, Anita and White, Lars O and Wager, Tor D and Koban, Leonie},
title = {{Brain neuromarkers predict self- and other-related mentalizing across adult, clinical, and developmental samples}},
journal = {Nature communications},
year = {2026},
month = jun,
volume = {17},
number = {1},
pages = {7229},
publisher = {Nature Publishing Group},
issn = {2041-1723},
doi = {10.1038/s41467-026-73945-w},
url = {https://doi.org/10.1038/s41467-026-73945-w},
pmid = {42251044},
pmcid = {PMC13396484}
}

RIS

TY - JOUR
AU - Açıl, Dorukhan
AU - Andrews-Hanna, Jessica R
AU - López-Solà, Marina
AU - van Buuren, Mariët
AU - Krabbendam, Lydia
AU - Zhang, Liwen
AU - van der Meer, Lisette
AU - Fuentes-Claramonte, Paola
AU - Pomarol-Clotet, Edith
AU - Salvador, Raymond
AU - Debbané, Martin
AU - Vrticka, Pascal
AU - Vuilleumier, Patrik
AU - Sbarra, David A
AU - Coppola, Andrea M
AU - Tusche, Anita
AU - White, Lars O
AU - Wager, Tor D
AU - Koban, Leonie
TI - Brain neuromarkers predict self- and other-related mentalizing across adult, clinical, and developmental samples
T2 - Nature communications
J2 - Nat Commun
PY - 2026
DA - 2026/06/06
VL - 17
IS - 1
SP - 7229
SN - 2041-1723
PB - Nature Publishing Group
DO - 10.1038/s41467-026-73945-w
UR - https://doi.org/10.1038/s41467-026-73945-w
LA - en
ER -

CSL-JSON

{
"id": "10.1038/s41467-026-73945-w",
"type": "article-journal",
"title": "Brain neuromarkers predict self- and other-related mentalizing across adult, clinical, and developmental samples",
"container-title": "Nature communications",
"author": [
{
"family": "Açıl",
"given": "Dorukhan"
},
{
"family": "Andrews-Hanna",
"given": "Jessica R"
},
{
"family": "López-Solà",
"given": "Marina"
},
{
"family": "van Buuren",
"given": "Mariët"
},
{
"family": "Krabbendam",
"given": "Lydia"
},
{
"family": "Zhang",
"given": "Liwen"
},
{
"family": "van der Meer",
"given": "Lisette"
},
{
"family": "Fuentes-Claramonte",
"given": "Paola"
},
{
"family": "Pomarol-Clotet",
"given": "Edith"
},
{
"family": "Salvador",
"given": "Raymond"
},
{
"family": "Debbané",
"given": "Martin"
},
{
"family": "Vrticka",
"given": "Pascal"
},
{
"family": "Vuilleumier",
"given": "Patrik"
},
{
"family": "Sbarra",
"given": "David A"
},
{
"family": "Coppola",
"given": "Andrea M"
},
{
"family": "Tusche",
"given": "Anita"
},
{
"family": "White",
"given": "Lars O"
},
{
"family": "Wager",
"given": "Tor D"
},
{
"family": "Koban",
"given": "Leonie"
}
],
"container-title-short": "Nat Commun",
"volume": "17",
"issue": "1",
"page": "7229",
"DOI": "10.1038/s41467-026-73945-w",
"PMID": "42251044",
"PMCID": "PMC13396484",
"ISSN": "2041-1723",
"publisher": "Nature Publishing Group",
"URL": "https://doi.org/10.1038/s41467-026-73945-w",
"language": "en",
"issued": {
"date-parts": [
[
2026,
6,
6
]
]
}
}

The tracing map gets a citation of its own once an author has validated it and it has a DOI.

Similar papers

The papers with a page that share the most with this one: the tools found in their code, their categories, datasets, cited references and authors, the rarest counting most.

[1] doi:10.3758/s13415-026-01425-1 [code]
Mindful minds: How group identity shapes brain and behavior in social decision-making.
Journal: Cognitive, affective & behavioral neuroscience
In common: SPM, 2 references, 2 authors
[2] doi:10.1038/s41467-026-71568-9 [code]
Convergent and selective representations of pain, appetitive processes, aversive processes, and cognitive control in the insula.
Journal: Nature communications
In common: SPM, Statistics and Machine Learning Toolbox, 5 references, author Tor D. Wager
[3] doi:10.1038/s41467-026-74565-0 [code]
The functional neurobiology of dispositions towards negative emotions.
Journal: Nature communications
In common: SPM, Statistics and Machine Learning Toolbox, 4 references, author Tor D. Wager
[4] doi:10.1038/s41593-026-02219-x [code]
A neural signature of adaptive mentalization.
Journal: Nature neuroscience
In common: SPM, Statistics and Machine Learning Toolbox, 7 references
[5] doi:10.1038/s41467-026-71151-2 [code]
Common and distinct neural correlates of social interaction processing and theory of mind in narratives.
Journal: Nature communications
In common: SPM, Statistics and Machine Learning Toolbox, 3 references, author Tor D. Wager
[6] doi:10.1038/s41467-026-74743-0 [code]
Meta-analytic evidence for distinct neural correlates of conditioned versus verbally induced placebo analgesia.
Journal: Nature communications
In common: SPM, Statistics and Machine Learning Toolbox, 1 reference, author Tor D. Wager
[7] doi:10.1038/s41467-026-74331-2 [code]
Trial-by-trial fMRI-neurofeedback dissociates fusiform and occipital contributions to face detection and recognition.
Journal: Nature communications
In common: SPM, Statistics and Machine Learning Toolbox, author Patrik Vuilleumier
[8] doi:10.3389/fncir.2026.1741762 [code]
Task-dependent increases and decreases of BOLD signal in theory of mind brain regions during strategic social interaction.
Journal: Frontiers in neural circuits
In common: SPM, Statistics and Machine Learning Toolbox, 3 references
[9] doi:10.1038/s41467-026-73153-6 [code]
Latent neural architecture organising shared aesthetic evaluations of visual artworks.
Journal: Nature communications
In common: Statistics and Machine Learning Toolbox, 4 references
[10] doi:10.1038/s41380-026-03547-x [code]
Multiscale characterization of cortical signatures in positive and negative schizotypy: a worldwide ENIGMA study.
Journal: Molecular psychiatry
In common: schizophrenia / psychosis, author Martin Debbané

Contribute

The authors of this paper can claim it, correct its record and validate its tracing map, and the maintainers of its code (its owner, or a public member of its organization) correct what it says of their repository; anyone signed in can ask for its removal. Every request goes to OSCR's own machine, which answers it; your account page follows them.

Sign in with ORCID to claim this paper as one of its authors, correct its record or validate its tracing map: when the paper's metadata lists your ORCID iD, you are recognized at once. Maintainers of its code: sign in with GitHub, then claim the repository on your account page.

Request its removal

To ask OSCR to remove this record, the copies of its authors' scripts or its tracing map, use the removal request page: signed in, you say who you are, what to remove and why, then review and confirm the request. Published rules decide every request (how).

Discussion, reproductions, activity

Discussion: questions and error reports about this paper and its code, from signed-in readers and its authors. It opens with sign-in.

Reproductions: reports from readers who ran the authors' code: what they reproduced, with which environment, commit and data. It opens with sign-in.

Activity: what happens around this paper: new versions of its record, its map's validation, discussions and reproductions. It opens with sign-in.