OSCR

Neural substrates of ambiguity and beauty in haiku poetry: an fMRI study.

Code ↔ Paper

4 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 4 matches
  1. [1] § Materials and methods › fMRI data analysis › Psychophysiological interaction (PPI) analysis ↔ Matlab_script/run_ppi_firstlevel_batch_manualPPI.m, lines 102–143 · score 0.57 · PPI regressor, Tikhonov, GLM, signals, VOI, HRF
  2. [2] § Results › Brain results ↔ Matlab_script/mvpa_PCC_ambiguity.m, lines 157–205 · score 0.55 · linear SVM, cross validation, components, MVPA, PCC, ambiguity
  3. [3] § Results › Brain results ↔ Matlab_script/mvpa_LIPL_ambiguity.m, lines 157–204 · score 0.54 · linear SVM, cross validation, components, MVPA, ambiguity
  4. [4] § Materials and methods › fMRI data analysis › Multivoxel pattern analysis (MVPA) ↔ Matlab_script/mvpa_LIPL_ambiguity.m, lines 1–6 · score 0.50 · cross validation, left IPL, MVPA, SVMs, classified, ambiguity

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 · 414 lines · 16 KB · no license · 2 matches

  1. %% MVPA Step 3: LIPL ROI × Ambiguity Classification
  2. % 目的: Left IPLが曖昧度（low vs high）を分類できるかを検証
  3. % 方法: Leave-one-trial-out cross-validation + PCA + 線形SVM
  4. % Statistics Toolbox不要
  6. clear; clc;
  8. %% パラメータ設定
  9. TR = 2; % 秒
  10. peak_trs_offset = [2, 3, 4]; % onset後の秒数（4-10秒）
  11. roi_radius = 6; % mm
  12. roi_coord_mni = [-60, -42, 40]; % Left IPL
  13. n_pca_components = 10; % PCA次元数
  15. % パス設定
  16. base_dir = 'D:/mri_toolbox';
  17. csv_path = fullfile(base_dir, 'mri_time', 'beauty_time.csv');
  18. output_dir = fullfile(base_dir, 'MVPA_Results', 'LIPL_Ambiguity');
  19. if ~exist(output_dir, 'dir'), mkdir(output_dir); end
  21. % 被験者リスト
  22. all_ids = setdiff(1:44, [1, 2, 7, 28, 33]); % 39名
  24. %% CSVデータ読み込み
  25. fprintf('CSVデータ読み込み中...\n');
  26. opts = detectImportOptions(csv_path, 'Encoding', 'UTF-8');
  27. behavior_data = readtable(csv_path, opts);
  28. fprintf('✓ CSV読み込み完了\n\n');
  30. %% セッション番号マッピング
  31. session_map = containers.Map('KeyType', 'double', 'ValueType', 'double');
  32. session_map(9) = 6; session_map(43) = 7; session_map(6) = 6;
  33. session_map(37) = 6; session_map(18) = 7;
  35. %% 全被験者でループ
  36. results = struct();
  37. success_count = 0;
  38. fail_count = 0;
  40. for idx = 1:length(all_ids)
  41. sub_id = all_ids(idx);
  43. fprintf('========================================\n');
  44. fprintf('Processing Sub%02d (%d/%d)\n', sub_id, idx, length(all_ids));
  45. fprintf('========================================\n');
  47. try
  48. %% 1. 機能画像のパス取得
  49. if isKey(session_map, sub_id)
  50. session = session_map(sub_id);
  51. else
  52. session = 5;
  53. end
  55. func_dir = sprintf('%s/HAIKU_%02d/%d_mb_ep_bold_mb4_sl76_tr2_task2/', ...
  56. base_dir, sub_id, session);
  57. func_files = spm_select('FPList', func_dir, '^s.*\.nii$');
  59. if isempty(func_files)
  60. error('機能画像が見つかりません');
  61. end
  63. fprintf('✓ 機能画像: %d ボリューム\n', size(func_files, 1));
  65. %% 2. 被験者の行動データ抽出
  66. sub_data = behavior_data(strcmp(strtrim(string(behavior_data.ID)), num2str(sub_id)) | ...
  67. strcmp(strtrim(string(behavior_data.ID)), sprintf('%02d', sub_id)), :);
  69. if isempty(sub_data)
  70. error('CSVにデータが見つかりません');
  71. end
  73. n_trials = height(sub_data);
  74. fprintf('✓ 試行数: %d\n', n_trials);
  76. % Ambiguityラベル（1=low, 2=high）
  77. ambiguity_labels = sub_data.condition;
  79. % 各条件の試行数を確認
  80. n_low = sum(ambiguity_labels == 1);
  81. n_high = sum(ambiguity_labels == 2);
  82. fprintf(' Low ambiguity: %d trials\n', n_low);
  83. fprintf(' High ambiguity: %d trials\n', n_high);
  85. if n_low < 2 || n_high < 2
  86. error('各条件に最低2試行必要です');
  87. end
  89. %% 3. ROIマスク作成
  90. V = spm_vol(deblank(func_files(1,:)));
  92. % MNI → voxel座標変換
  93. vox_coord = round(inv(V.mat) * [roi_coord_mni'; 1]);
  94. vox_coord = vox_coord(1:3);
  96. fprintf('✓ ROI中心 (MNI): [%d, %d, %d]\n', roi_coord_mni);
  97. fprintf('✓ ROI中心 (vox): [%d, %d, %d]\n', vox_coord);
  99. % 6mm球内のボクセルを特定
  100. [X, Y, Z] = ndgrid(1:V.dim(1), 1:V.dim(2), 1:V.dim(3));
  101. voxel_coords = [X(:), Y(:), Z(:)];
  103. % 各ボクセルのMNI座標を計算
  104. mni_coords = V.mat * [voxel_coords'; ones(1, size(voxel_coords, 1))];
  105. mni_coords = mni_coords(1:3, :)';
  107. % ROI中心からの距離
  108. distances = sqrt(sum((mni_coords - repmat(roi_coord_mni, size(mni_coords, 1), 1)).^2, 2));
  109. roi_mask = distances <= roi_radius;
  110. n_voxels = sum(roi_mask);
  112. fprintf('✓ ROI内ボクセル数: %d\n', n_voxels);
  114. %% 4. 各試行のBOLD信号を抽出（複数TRの平均）
  115. X_data = zeros(n_trials, n_voxels); % 試行 × ボクセル
  117. for trial = 1:n_trials
  118. onset_sec = sub_data.time_b(trial);
  120. % 複数TRを平均
  121. trial_signal = zeros(length(peak_trs_offset), n_voxels);
  123. for t_idx = 1:length(peak_trs_offset)
  124. peak_sec = onset_sec + peak_trs_offset(t_idx);
  125. peak_tr = round(peak_sec / TR) + 1;
  127. if peak_tr > size(func_files, 1)
  128. peak_tr = size(func_files, 1);
  129. end
  131. vol = spm_read_vols(spm_vol(deblank(func_files(peak_tr,:))));
  132. roi_voxels = vol(roi_mask);
  133. trial_signal(t_idx, :) = roi_voxels(:)';
  134. end
  136. % 複数TRの平均
  137. X_data(trial, :) = mean(trial_signal, 1);
  138. end
  140. fprintf('✓ BOLD信号抽出完了: [%d trials × %d voxels]\n', n_trials, n_voxels);
  142. %% 5. PCAで次元削減
  143. n_components = min(n_pca_components, n_trials - 2);
  144. X_centered = X_data - mean(X_data, 1);
  146. % SVDでPCA
  147. [U, S, V_pca] = svd(X_centered', 'econ');
  148. pca_components = U(:, 1:n_components);
  150. % 説明分散
  151. explained_var = diag(S).^2 / sum(diag(S).^2);
  152. cumsum_var = cumsum(explained_var(1:n_components));
  154. fprintf('✓ PCAで%dボクセル → %d次元に削減\n', n_voxels, n_components);
  155. fprintf(' 累積説明分散: %.1f%%\n', cumsum_var(end) * 100);
  157. %% 6. Leave-One-Trial-Out Cross-Validation（線形SVM）
  158. predictions = zeros(n_trials, 1);
  159. decision_values = zeros(n_trials, 1);
  161. for trial = 1:n_trials
  162. % 訓練データとテストデータを分割
  163. train_idx = setdiff(1:n_trials, trial);
  164. test_idx = trial;
  166. X_train = X_data(train_idx, :);
  167. y_train = ambiguity_labels(train_idx);
  168. X_test = X_data(test_idx, :);
  170. % 訓練データで中心化
  171. train_mean = mean(X_train, 1);
  173. % PCA変換
  174. X_train_centered = X_train - train_mean;
  175. X_test_centered = X_test - train_mean;
  177. X_train_pca = X_train_centered * pca_components;
  178. X_test_pca = X_test_centered * pca_components;
  180. % 線形SVM（手動実装）
  181. % 2クラス分類: y = 1 or 2 → -1 or +1 に変換
  182. y_train_svm = y_train;
  183. y_train_svm(y_train == 1) = -1; % Low ambiguity
  184. y_train_svm(y_train == 2) = +1; % High ambiguity
  186. % Linear SVM (simplified version using ridge regression)
  187. lambda = 1.0;
  188. XtX = X_train_pca' * X_train_pca;
  189. XtX_reg = XtX + lambda * eye(size(XtX));
  190. w = XtX_reg \ (X_train_pca' * y_train_svm);
  192. % 予測（決定値）
  193. decision_val = X_test_pca * w;
  194. decision_values(trial) = decision_val;
  196. % 分類（符号で判定）
  197. if decision_val >= 0
  198. predictions(trial) = 2; % High ambiguity
  199. else
  200. predictions(trial) = 1; % Low ambiguity
  201. end
  202. end
  204. fprintf('✓ Cross-validation完了\n');
  206. %% 7. 精度計算
  207. y_true = ambiguity_labels;
  208. y_pred = predictions;
  210. % 正解率（Accuracy）
  211. accuracy = sum(y_true == y_pred) / n_trials;
  213. % 各条件の正解率
  214. low_idx = (y_true == 1);
  215. high_idx = (y_true == 2);
  217. accuracy_low = sum(y_true(low_idx) == y_pred(low_idx)) / sum(low_idx);
  218. accuracy_high = sum(y_true(high_idx) == y_pred(high_idx)) / sum(high_idx);
  220. % Balanced accuracy
  221. balanced_accuracy = (accuracy_low + accuracy_high) / 2;
  223. % Confusion matrix
  224. tp = sum(y_true == 2 & y_pred == 2);
  225. tn = sum(y_true == 1 & y_pred == 1);
  226. fp = sum(y_true == 1 & y_pred == 2);
  227. fn = sum(y_true == 2 & y_pred == 1);
  229. % Sensitivity & Specificity
  230. if (tp + fn) > 0
  231. sensitivity = tp / (tp + fn);
  232. else
  233. sensitivity = NaN;
  234. end
  236. if (tn + fp) > 0
  237. specificity = tn / (tn + fp);
  238. else
  239. specificity = NaN;
  240. end
  242. % Binomial test p-value
  243. n_correct = sum(y_true == y_pred);
  244. p_binomial = binomial_test(n_correct, n_trials, 0.5);
  246. fprintf('\n--- デコーディング結果 ---\n');
  247. fprintf('Accuracy: %.2f%% (%d/%d)\n', accuracy * 100, sum(y_true == y_pred), n_trials);
  248. fprintf('Balanced Accuracy: %.2f%%\n', balanced_accuracy * 100);
  249. fprintf(' Low ambiguity正解率: %.2f%% (%d/%d)\n', accuracy_low * 100, sum(y_true(low_idx) == y_pred(low_idx)), sum(low_idx));
  250. fprintf(' High ambiguity正解率: %.2f%% (%d/%d)\n', accuracy_high * 100, sum(y_true(high_idx) == y_pred(high_idx)), sum(high_idx));
  251. fprintf('Sensitivity: %.2f%%\n', sensitivity * 100);
  252. fprintf('Specificity: %.2f%%\n', specificity * 100);
  253. fprintf('p = %.4f (binomial test vs chance)\n', p_binomial);
  255. fprintf('\nConfusion Matrix:\n');
  256. fprintf(' Predicted Low Predicted High\n');
  257. fprintf('Actual Low %3d %3d\n', tn, fp);
  258. fprintf('Actual High %3d %3d\n', fn, tp);
  260. %% 8. 結果を保存
  261. results(idx).sub_id = sub_id;
  262. results(idx).n_trials = n_trials;
  263. results(idx).n_low = n_low;
  264. results(idx).n_high = n_high;
  265. results(idx).n_voxels = n_voxels;
  266. results(idx).n_components = n_components;
  267. results(idx).explained_var = cumsum_var(end);
  268. results(idx).accuracy = accuracy;
  269. results(idx).balanced_accuracy = balanced_accuracy;
  270. results(idx).accuracy_low = accuracy_low;
  271. results(idx).accuracy_high = accuracy_high;
  272. results(idx).sensitivity = sensitivity;
  273. results(idx).specificity = specificity;
  274. results(idx).p_value = p_binomial;
  275. results(idx).confusion_matrix = [tn, fp; fn, tp];
  276. results(idx).predictions = predictions;
  277. results(idx).true_labels = y_true;
  278. results(idx).decision_values = decision_values;
  280. % 個別ファイルに保存
  281. sub_result = results(idx);
  282. save(fullfile(output_dir, sprintf('sub%02d_result.mat', sub_id)), 'sub_result');
  284. fprintf('✓ 保存完了\n\n');
  285. success_count = success_count + 1;
  287. catch ME
  288. fprintf('✗ エラー: %s\n', ME.message);
  289. if ~isempty(ME.stack)
  290. fprintf(' スタック: %s (line %d)\n\n', ME.stack(1).name, ME.stack(1).line);
  291. end
  292. fail_count = fail_count + 1;
  293. results(idx).sub_id = sub_id;
  294. results(idx).error = ME.message;
  295. end
  296. end
  298. %% 9. グループレベル集計
  299. fprintf('========================================\n');
  300. fprintf('全被験者処理完了\n');
  301. fprintf('========================================\n');
  302. fprintf('成功: %d / %d\n', success_count, length(all_ids));
  303. fprintf('失敗: %d / %d\n\n', fail_count, length(all_ids));
  305. % 成功した被験者の結果を集計
  306. valid_idx = arrayfun(@(x) isfield(x, 'accuracy'), results);
  307. valid_results = results(valid_idx);
  309. if isempty(valid_results)
  310. fprintf('有効な結果がありません\n');
  311. return;
  312. end
  314. all_accuracies = [valid_results.accuracy];
  315. all_balanced_accuracies = [valid_results.balanced_accuracy];
  316. all_p_values = [valid_results.p_value];
  318. fprintf('--- グループレベル統計 ---\n');
  319. fprintf('有効被験者数: %d\n', length(all_accuracies));
  321. fprintf('\n【Accuracy】\n');
  322. fprintf('平均: %.2f%% (SD = %.2f%%)\n', mean(all_accuracies) * 100, std(all_accuracies) * 100);
  323. fprintf('中央値: %.2f%%\n', median(all_accuracies) * 100);
  324. fprintf('範囲: [%.2f%%, %.2f%%]\n', min(all_accuracies) * 100, max(all_accuracies) * 100);
  326. fprintf('\n【Balanced Accuracy】\n');
  327. fprintf('平均: %.2f%% (SD = %.2f%%)\n', mean(all_balanced_accuracies) * 100, std(all_balanced_accuracies) * 100);
  328. fprintf('中央値: %.2f%%\n', median(all_balanced_accuracies) * 100);
  330. % 分布
  331. fprintf('\n【分布】\n');
  332. fprintf('Accuracy > 60%%: %d名 (%.1f%%)\n', sum(all_accuracies > 0.6), sum(all_accuracies > 0.6)/length(all_accuracies)*100);
  333. fprintf('Accuracy > 55%%: %d名 (%.1f%%)\n', sum(all_accuracies > 0.55), sum(all_accuracies > 0.55)/length(all_accuracies)*100);
  334. fprintf('Accuracy > 50%%: %d名 (%.1f%%)\n', sum(all_accuracies > 0.5), sum(all_accuracies > 0.5)/length(all_accuracies)*100);
  335. fprintf('Accuracy < 50%%: %d名 (%.1f%%)\n', sum(all_accuracies < 0.5), sum(all_accuracies < 0.5)/length(all_accuracies)*100);
  337. % 有意な被験者数
  338. sig_count = sum(all_p_values < 0.05);
  339. fprintf('\n有意な被験者数 (p<0.05): %d / %d (%.1f%%)\n', sig_count, length(all_p_values), sig_count/length(all_p_values)*100);
  341. % One-sample t-test
  342. mean_acc = mean(all_accuracies);
  343. std_acc = std(all_accuracies);
  344. n = length(all_accuracies);
  345. chance_level = 0.5;
  347. t_group = (mean_acc - chance_level) / (std_acc / sqrt(n));
  349. fprintf('\n【グループレベル検定】\n');
  350. fprintf('平均 Accuracy vs Chance (50%%):\n');
  351. fprintf('t(%d) = %.3f\n', n-1, t_group);
  353. if n > 30
  354. p_group = 1 - normcdf_manual(abs(t_group));
  355. fprintf('p ≈ %.6f (one-tailed)\n', p_group);
  357. if p_group < 0.001
  358. fprintf('*** p < 0.001\n');
  359. elseif p_group < 0.01
  360. fprintf('** p < 0.01\n');
  361. elseif p_group < 0.05
  362. fprintf('* p < 0.05\n');
  363. end
  364. end
  366. % 効果量
  367. cohens_d = (mean_acc - chance_level) / std_acc;
  368. fprintf('Cohen''s d = %.3f\n', cohens_d);
  370. % 95%信頼区間
  371. se_acc = std_acc / sqrt(n);
  372. ci_lower = mean_acc - 1.96 * se_acc;
  373. ci_upper = mean_acc + 1.96 * se_acc;
  374. fprintf('95%% CI: [%.2f%%, %.2f%%]\n', ci_lower * 100, ci_upper * 100);
  376. %% 10. 結果を保存
  377. save(fullfile(output_dir, 'group_results.mat'), 'results', 'all_accuracies', 'all_balanced_accuracies', 'all_p_values');
  379. % サマリーをテキストファイルにも保存
  380. fid = fopen(fullfile(output_dir, 'summary.txt'), 'w');
  381. fprintf(fid, 'MVPA Results: LIPL × Ambiguity Classification\n');
  382. fprintf(fid, '==============================================\n\n');
  383. fprintf(fid, 'パラメータ:\n');
  384. fprintf(fid, ' ROI: Left IPL [-60, -42, 40], 6mm球\n');
  385. fprintf(fid, ' 時間窓: onset + [%s]秒の平均\n', num2str(peak_trs_offset));
  386. fprintf(fid, ' PCA次元: %d\n\n', n_pca_components);
  387. fprintf(fid, '成功: %d / %d\n', success_count, length(all_ids));
  388. fprintf(fid, '失敗: %d / %d\n\n', fail_count, length(all_ids));
  389. fprintf(fid, '平均 Accuracy: %.2f%% (SD = %.2f%%)\n', mean_acc * 100, std_acc * 100);
  390. fprintf(fid, '中央値: %.2f%%\n', median(all_accuracies) * 100);
  391. fprintf(fid, '範囲: [%.2f%%, %.2f%%]\n\n', min(all_accuracies) * 100, max(all_accuracies) * 100);
  392. fprintf(fid, 't(%d) = %.3f\n', n-1, t_group);
  393. fprintf(fid, 'Cohen''s d = %.3f\n', cohens_d);
  394. fprintf(fid, '95%% CI: [%.2f%%, %.2f%%]\n', ci_lower * 100, ci_upper * 100);
  395. fclose(fid);
  397. fprintf('\n✓ 全結果を保存: %s\n', output_dir);
  399. %% ヘルパー関数
  400. function p = normcdf_manual(x)
  401. t = 1 ./ (1 + 0.2316419 * abs(x));
  402. d = 0.3989423 * exp(-x.*x/2);
  403. p = d .* t .* (0.3193815 + t .* (-0.3565638 + t .* (1.781478 + t .* (-1.821256 + t .* 1.330274))));
  404. p(x > 0) = 1 - p(x > 0);
  405. end
  407. function p = binomial_test(k, n, p0)
  408. % 正規近似版（簡易）
  409. expected = n * p0;
  410. variance = n * p0 * (1 - p0);
  411. z = (k - expected) / sqrt(variance);
  412. p = 2 * (1 - normcdf_manual(abs(z)));
  413. p = min(p, 1);
  414. end

mvpa_LIPL_ambiguity.m, no license · at the source

Overview

Authors: Jimpei Hitsuwari1,2, Michio Nomura3
  1. Experimental Psychology Unit, Faculty of Humanities and Social Sciences, Helmut Schmidt University/University of the Federal Armed Forces Hamburg, Hamburg, Germany
  2. Japan Society for the Promotion of Science, Tokyo, Japan
  3. Graduate School of Education, Kyoto University, Kyoto, Japan
Institutions: Helmut Schmidt University (Germany); Japan Society for the Promotion of Science (Japan); Kyoto University (Japan)
Journal: Frontiers in human neuroscience, volume 20, article 1887834
Dates: received 21 May 2026; accepted 31 July 2026; published online 28 August 2026
Type: Research article · Language: English
License: CC BY
Identifiers: DOI 10.3389/fnhum.2026.1887834 · PMID 42729248 · PMCID PMC13562011 · OpenAlex W7204498382
Open access: gold, a free copy (OpenAlex)
Status: code verified
Categories: fMRI (modality), systems (subfield)
Methods: Spectral & time-frequency, Statistics, Smoothing, state filtering, decompositions, Machine learning, fMRI & imaging
Keywords: aesthetic experience, ambiguity, fMRI, haiku poetry, neuroaesthetics
Topic: Aesthetic Perception and Analysis (Cognitive Neuroscience, Neuroscience), according to OpenAlex
Funding: Japan Society for the Promotion of Science (22KJ1813, 23K22374)
Citations: not cited yet (Europe PMC); 61 references in the paper

Abstract

Although neuroaesthetics has extensively examined visual art and music, the neural mechanisms underlying poetic appreciation remain less explored. This fMRI study investigated how the brain processes ambiguity and aesthetic experiences using haiku, the world’s shortest form of poetry. Originating in Japan, haiku is characterized by extreme brevity, often omitting information to compel readers to generate mental imagery and confront inherent ambiguities. Thirty-nine participants read haiku, which were pre-categorized via an independent survey into high- and low-ambiguity conditions, and rated their beauty inside the scanner. Consistent with previous findings, the behavioral results indicated a preference for low-ambiguity haiku (a small effect, standardized β = −0.16). Neuroimaging revealed that reading low-ambiguity haiku, compared to high-ambiguity ones, elicited greater activation in the posterior cingulate cortex and the left inferior parietal lobule, including the angular gyrus (cluster-level pFWE = 0.033 and 0.005, respectively). These regions are associated with the default mode network and semantic integration, suggesting that more interpretable poems may involve stronger self-referential processing and scene construction. Higher beauty ratings were significantly associated with increased activation in the primary visual cortex (cluster-level pFWE = 0.002). Our findings imply that the aesthetic appeal of haiku depends on semantic accessibility, with reduced ambiguity associated with greater engagement of self-referential processing in the default mode network and with increased activity in the visual cortex. More broadly, using haiku as a model offers a foundation for neuroscientific investigations of ambiguity and aesthetic experience across poetry, narratives, and other art forms.

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

Repository

Its files are read in the Code ↔ Paper reader above, with 4 matches between paragraphs and lines of code.

OSF emsp4

License: none: the authors keep all their rights
State: the link answers, verified on 27 September 2026
Evidence: files inventoried
Languages: MATLAB (12), R (1)
Size: 113 files, 13 scripts
Software Heritage: not checked
Found in: “Data availability statement”
Not found: README, license file, CITATION.cff, environment file, tests, continuous integration, documentation
Tools: SPM (10 files), tidyverse (1 file)
Availability: 1 check, the latest on 27 September 2026: the link answers (HTTP 200)
  • 27 September 2026: the link answers (HTTP 200)
13 files
At the source: osf.io/emsp4/

The paper's code and data availability statement is in the Data section.

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:

  • 1 repository of the authors' code, each at its verified commit, with its license and how the link was found in the paper;
  • 13 scripts, each with its path and the digest of its content;
  • 4 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

No dataset and no data link were found in the paper.

Data availability statement

Individual-level MRI data are not publicly available due to participant privacy considerations but may be shared upon reasonable request to the corresponding author. Analysis scripts (including the nonparametric SnPM scripts) are available on OSF at https://osf.io/emsp4/. The permutation analyses used SnPM13.1.09 under SPM25 and MATLAB R2024b.

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 2, 28 September 2026

  • Funding: added Japan Society for the Promotion of Science: 22KJ1813, 23K22374

Version 1, 27 September 2026: the first record

Recorded: type, language, journal, volume, pages, dates, 2 authors, 5 keywords, 47 references.

Cite

This paper

Hitsuwari, J., & Nomura, M. (2026). Neural substrates of ambiguity and beauty in haiku poetry: an fMRI study. Frontiers in human neuroscience, 20, 1887834. https://doi.org/10.3389/fnhum.2026.1887834

BibTeX

@article{hitsuwari2026neural,
author = {Hitsuwari, Jimpei and Nomura, Michio},
title = {{Neural substrates of ambiguity and beauty in haiku poetry: an fMRI study}},
journal = {Frontiers in human neuroscience},
year = {2026},
month = aug,
volume = {20},
pages = {1887834},
publisher = {Frontiers Media SA},
issn = {1662-5161},
doi = {10.3389/fnhum.2026.1887834},
url = {https://doi.org/10.3389/fnhum.2026.1887834},
pmid = {42729248},
pmcid = {PMC13562011}
}

RIS

TY - JOUR
AU - Hitsuwari, Jimpei
AU - Nomura, Michio
TI - Neural substrates of ambiguity and beauty in haiku poetry: an fMRI study
T2 - Frontiers in human neuroscience
J2 - Front Hum Neurosci
PY - 2026
DA - 2026/08/28
VL - 20
SP - 1887834
SN - 1662-5161
PB - Frontiers Media SA
DO - 10.3389/fnhum.2026.1887834
UR - https://doi.org/10.3389/fnhum.2026.1887834
LA - en
ER -

CSL-JSON

{
"id": "10.3389/fnhum.2026.1887834",
"type": "article-journal",
"title": "Neural substrates of ambiguity and beauty in haiku poetry: an fMRI study",
"container-title": "Frontiers in human neuroscience",
"author": [
{
"family": "Hitsuwari",
"given": "Jimpei"
},
{
"family": "Nomura",
"given": "Michio"
}
],
"container-title-short": "Front Hum Neurosci",
"volume": "20",
"page": "1887834",
"DOI": "10.3389/fnhum.2026.1887834",
"PMID": "42729248",
"PMCID": "PMC13562011",
"ISSN": "1662-5161",
"publisher": "Frontiers Media SA",
"URL": "https://doi.org/10.3389/fnhum.2026.1887834",
"language": "en",
"issued": {
"date-parts": [
[
2026,
8,
28
]
]
}
}

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:10.1038/s41467-026-73153-6 [code]
Latent neural architecture organising shared aesthetic evaluations of visual artworks.
Journal: Nature communications
In common: fMRI, 6 references
[2] doi:10.1371/journal.pone.0355737
How artists experience their own art.
Journal: PloS one
In common: fMRI, 3 references
[3] doi:10.1093/cercor/bhag075 [code]
Cortical dynamics of icon perception: effects of concreteness and attractiveness.
Journal: Cerebral cortex (New York, N.Y. : 1991)
In common: 3 references
[4] doi:10.1016/j.ynirp.2026.100376 [code]
Ambiguity tolerance and resting-state functional connectivity: A preregistered conceptual replication in a Japanese sample.
Journal: Neuroimage. Reports
In common: SPM, fMRI, systems, 1 reference
[5] doi:10.1162/netn.a.549 [code]
Distributed cortical network dynamics of binocular convergent eye movements in humans.
Journal: Network neuroscience (Cambridge, Mass.)
In common: SPM, fMRI, systems, 1 reference
[6] doi:10.1093/braincomms/fcag196 [code]
Music processing in behavioural variant frontotemporal dementia and Alzheimer's disease: a functional MRI study.
Journal: Brain communications
In common: SPM, fMRI, 1 reference
[7] doi:10.1162/imag.a.1197 [code]
Blood volume-sensitive laminar fMRI with VASO in human hippocampus: Capabilities and biophysical challenges at clinical 7T scanners.
Journal: Imaging neuroscience (Cambridge, Mass.)
In common: SPM, tidyverse, fMRI, systems
[8] doi:10.7554/elife.103846 [code]
Overt visual attention modulates decision-related signals in the frontal cortex.
Journal: eLife
In common: SPM, tidyverse, 1 reference
[9] doi:10.1038/s41598-026-46149-x [code]
Dynamic neural representations of scene beauty are relatively unaffected by stimulus timing and task.
Journal: Scientific reports
In common: 2 references
[10] doi:10.1162/imag.a.1363 [code]
Social approach-avoidance conflict: Behavioural, emotional, and neural correlates along a depression-social anxiety spectrum.
Journal: Imaging neuroscience (Cambridge, Mass.)
In common: SPM, tidyverse, fMRI

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.