OSCR

The language network responds robustly to sentences across tasks.

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] § Results › Task demands lead to higher responses in the language network but also engage the multiple demand network. ↔ effect_estimation/visualize_data.ipynb, lines 62–86 · score 0.53 · hard memory probe, easy memory probe, button press, sentiment, V5, V3
  2. [2] § Results › Task demands lead to higher responses in the language network but also engage the multiple demand network. ↔ spatial_correlation/visualize_data.ipynb, lines 14–96 · score 0.53 · hard memory probe, easy memory probe, button press, sentiment, V5, V3
  3. [3] § Methods › Behavioral performance in the scanner ↔ effect_estimation/visualize_data.ipynb, lines 204–262 · score 0.51 · hard memory probe, easy memory probe, button press, sentiment, nonword, sentence
  4. [4] § Methods › fMRI data preprocessing ↔ spatial_correlation/MATLAB_Scripts/spcorr_mega_all.m, lines 152–180 · score 0.50 · MATLAB scripts, SPM, voxels, regressors

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

Jupyter notebook · 417 lines · 15 KB · MIT · 2 matches

  1. # %%
  2. import pandas as pd
  3. import pathlib
  4. import numpy as np
  5. import matplotlib.pyplot as plt
  6. import seaborn as sns
  7. cwd = pathlib.Path('/Users/rgao76/Documents/DiffTasks/effect_estimation_2025')
  9. # %%
  10. def brighten_hex_color(hex_color, alpha):
  11. # Convert hex to RGB
  12. r = int(hex_color[1:3], 16)
  13. g = int(hex_color[3:5], 16)
  14. b = int(hex_color[5:7], 16)
  16. # Brighten each channel
  17. r_new = int(r + (255 - r) * alpha)
  18. g_new = int(g + (255 - g) * alpha)
  19. b_new = int(b + (255 - b) * alpha)
  21. # Clamp to [0, 255] and convert back to hex
  22. return '#{:02x}{:02x}{:02x}'.format(min(r_new, 255), min(g_new, 255), min(b_new, 255))
  24. # %%
  25. data = pd.read_csv(cwd / 'Data' / 'all_data.csv')
  26. data = data[~data['ROI'].str.contains('AngG')]
  27. data['Version'] = data['Version'].fillna('')
  29. def process_data(df, hemisphere=None, system=None, localizer=None,
  30. include_localizer=False, include_ROI=False):
  31. mask = (df['System'] == system)
  32. if hemisphere:
  33. mask &= (df['Hemisphere'] == hemisphere)
  34. if localizer:
  35. mask &= (df['Localizer'] == localizer)
  37. df_filtered = df[mask].copy()
  38. df_filtered['Task_Effect'] = (df_filtered['Version'] + ' ' + df_filtered['Effect']).str.strip()
  40. group_cols = ['System', 'Subject', 'Task_Effect']
  41. if include_localizer:
  42. group_cols.append('Localizer')
  43. if include_ROI:
  44. group_cols.append('ROI')
  46. return df_filtered[group_cols + ['EffectSize']].groupby(group_cols).mean().reset_index()
  48. data_language_lh_localizers = process_data(data, hemisphere='LH', system='Language', include_localizer=True)
  49. data_language_rh_localizers = process_data(data, hemisphere='RH', system='Language', include_localizer=True)
  50. data_language_lh_ROI = process_data(data, hemisphere='LH', system='Language', include_ROI=True)
  51. data_language_lh = process_data(data, hemisphere='LH', system='Language', localizer='V1')
  52. data_language_rh = process_data(data, hemisphere='RH', system='Language', localizer='V1')
  53. data_md_lh = process_data(data, hemisphere='LH', system='MD')
  54. data_md_rh = process_data(data, hemisphere='RH', system='MD')
  56. # %%
  57. data_language_lh
  59. # %%
  60. beh_data
  62. # %%
  63. hue_order = ['V1 S', 'V1 N', 'V2 S', 'V2 N', 'V3 S', 'V3 N', 'V6 S', 'V6 N', 'V4 S', 'V4 N', 'V5 S', 'V5 N']
  64. all_colors_expanded = ["#606060", brighten_hex_color("#606060", 0.4),
  65. "#0000ff", brighten_hex_color("#0000ff", 0.7),
  66. "#00aaff", brighten_hex_color("#00aaff", 0.7),
  67. "#606060", brighten_hex_color("#0000ff", 0.7),
  68. "#ff007f", brighten_hex_color("#0000ff", 0.7),
  69. "#ff6e00", brighten_hex_color("#0000ff", 0.7),
  70. ]
  71. localizer_renaming = {
  72. 'V1': 'V1 (button press)',
  73. 'V2': 'V2 (hard memory probe)',
  74. 'V3': 'V3 (easy memory probe)',
  75. 'V4': 'V5 (comprehension q)',
  76. 'V5': 'V6 (sentiment q)',
  77. 'V6': 'V4 (button press)',
  78. }
  79. order = ['V2 (hard memory probe)', 'V3 (easy memory probe)',
  80. 'V4 (button press)', 'V5 (comprehension q)', 'V6 (sentiment q)']
  82. header_sz = 10
  83. axis_label_sz = 6
  84. stripplot_sz = 1.0
  85. annot_height = 1
  86. jitter = 0.2
  88. # %%
  89. data_language_lh_withSet = data_language_lh.copy()
  90. beh_data = pd.read_csv(cwd / '..' / 'behavioral_analysis_2025' / 'qc' / 'Data' / 'processed_data_by_trial.csv')
  91. beh_data = beh_data[beh_data['Version'] == 1]
  92. data_language_lh_withSet = data_language_lh_withSet.merge(beh_data[['Subject', 'Set']].drop_duplicates(), on='Subject', how='left')
  94. plt.figure(figsize=(7, 2), tight_layout=True, dpi=300)
  95. ax = sns.barplot(
  96. x='Set', y='EffectSize', hue='Task_Effect',
  97. data=data_language_lh_withSet,
  98. legend=False, errorbar='se',
  99. palette=all_colors_expanded, hue_order=hue_order,
  100. err_kws={'linewidth': 1.5})
  101. sns.stripplot(
  102. x='Set', y='EffectSize', hue='Task_Effect',
  103. data=data_language_lh_withSet,
  104. dodge=True, jitter=jitter, legend=False,
  105. size=stripplot_sz, palette='dark:black',
  106. hue_order=hue_order)
  107. plt.xticks(fontsize=8)
  108. plt.yticks(fontsize=8)
  109. plt.ylabel("Effect size", fontsize=8)
  110. plt.xlabel("", fontsize=axis_label_sz)
  111. ax.set_xticklabels([f'Set {i}' for i in range(1, 6)], fontsize=8)
  112. plt.savefig(cwd / 'Figures' / 'effects_V1_bySet.png', dpi=300, format='png',
  113. transparent=True)
  115. # print the number of unique subjects in each set
  116. for set_num in range(1, 6):
  117. num_subjects = data_language_lh_withSet[data_language_lh_withSet['Set'] == set_num]['Subject'].nunique()
  118. print(f'Set {set_num}: {num_subjects} unique subjects')
  120. # %%
  121. # Localizers
  122. plt.figure(figsize=(7, 2), tight_layout=True, dpi=300)
  123. data_language_lh_localizers['Localizer'] = data_language_lh_localizers['Localizer'].replace(localizer_renaming)
  124. ax = sns.barplot(
  125. x='Localizer', y='EffectSize', hue='Task_Effect',
  126. data=data_language_lh_localizers,
  127. legend=False, errorbar='se',
  128. palette=all_colors_expanded, hue_order=hue_order, order=order,
  129. err_kws={'linewidth': 1.5})
  130. sns.stripplot(
  131. x='Localizer', y='EffectSize', hue='Task_Effect',
  132. data=data_language_lh_localizers,
  133. dodge=True, jitter=jitter, legend=False,
  134. size=stripplot_sz, palette='dark:black',
  135. hue_order=hue_order, order=order)
  136. plt.xticks(fontsize=6)
  137. plt.yticks(fontsize=6)
  138. plt.ylabel("Effect size", fontsize=axis_label_sz)
  139. plt.xlabel("", fontsize=axis_label_sz)
  140. plt.savefig(cwd / 'Figures' / 'effects_localizers.png', dpi=300, format='png',
  141. transparent=True)
  144. # %%
  145. # V1
  146. plt.figure(figsize=(2, 2), tight_layout=True, dpi=300)
  147. data_language_lh['Localizer'] = 'V1'
  149. ax = sns.barplot(
  150. x='Localizer', y='EffectSize', hue='Task_Effect',
  151. data=data_language_lh, legend=False, errorbar='se',
  152. palette=all_colors_expanded, hue_order=hue_order,
  153. err_kws={'linewidth': 1.5})
  154. sns.stripplot(
  155. x='Localizer', y='EffectSize', hue='Task_Effect',
  156. data=data_language_lh,
  157. dodge=True, jitter=jitter, legend=False,
  158. size=stripplot_sz, palette='dark:black',
  159. hue_order=hue_order)
  161. plt.xticks(fontsize=6)
  162. plt.yticks(fontsize=6)
  164. plt.ylabel("Effect size", fontsize=axis_label_sz)
  165. plt.xlabel("Localizer:\nV1 (button press)", fontsize=axis_label_sz)
  166. plt.xticks([])
  167. plt.savefig(cwd / 'Figures' / 'effects_V1.png', dpi=300, format='png',
  168. transparent=True)
  170. # %%
  171. # ROI
  172. for roi in data_language_lh_ROI['ROI'].unique():
  173. fig = plt.figure(figsize=(2, 1.5), dpi=300)
  175. data_subset = data_language_lh_ROI[data_language_lh_ROI['ROI'] == roi]
  176. data_subset_mean = data_subset[['Subject', 'Task_Effect', 'EffectSize']].groupby(
  177. ['Subject', 'Task_Effect']
  178. ).mean().reset_index()
  179. data_subset_mean['Localizer'] = 'V1'
  181. ax = sns.barplot(
  182. x='Localizer', y='EffectSize', hue='Task_Effect',
  183. data=data_subset_mean, legend=False, errorbar='se',
  184. palette=all_colors_expanded, hue_order=hue_order,
  185. err_kws={'linewidth': 1.5})
  186. sns.stripplot(
  187. x='Localizer', y='EffectSize', hue='Task_Effect',
  188. data=data_subset_mean, dodge=True, jitter=jitter, legend=False,
  189. size=stripplot_sz, palette='dark:black',
  190. hue_order=hue_order)
  192. # change xticks size
  193. plt.xticks(fontsize=6)
  194. plt.yticks(fontsize=6)
  196. plt.ylabel("Effect size", fontsize=axis_label_sz)
  197. plt.xlabel(None, fontsize=axis_label_sz)
  198. plt.xticks([])
  199. plt.title(roi.replace('LH_', ''), fontsize=axis_label_sz)
  200. fig.subplots_adjust(left=0.25)
  201. plt.savefig(cwd / 'Figures' / f'effects_{roi}.png', dpi=300, format='png',
  202. transparent=True)
  204. # %%
  205. from matplotlib.patches import Rectangle
  207. colors_s = ['#606060', '#0000ff', '#00aaff', '#ff007f', '#ff6e00']
  208. labels = ['Button press',
  209. 'Hard memory probe',
  210. 'Easy memory probe',
  211. 'Comprehension q',
  212. 'Sentiment q']
  213. colors_col2 = colors_s[:3] + [None, None] # Repeat first 3, leave last 2 blank
  215. # Parameters
  216. box_width = 1.2
  217. box_height = 0.4
  218. margin = 0.3
  219. label_width = 4
  220. gap_between_columns = 0.8
  222. total_rows = len(colors_s)
  223. total_height = total_rows * (box_height + margin)
  224. total_width = label_width + 2 * box_width + gap_between_columns
  226. fig, ax = plt.subplots(figsize=(total_width, total_height))
  228. # Draw label column and first color column
  229. for i, (label, color) in enumerate(zip(labels, colors_s)):
  230. y_pos = total_height - (i + 1) * (box_height + margin)
  232. # Label
  233. ax.text(label_width - 4, y_pos + box_height / 2, label,
  234. ha='left', va='center', fontsize=24)
  236. # First column rectangle
  237. rect1 = Rectangle((label_width, y_pos), box_width, box_height,
  238. color=color, edgecolor='black')
  239. ax.add_patch(rect1)
  241. # Draw second column
  242. for i, color in enumerate(colors_col2):
  243. y_pos = total_height - (i + 1) * (box_height + margin)
  244. if color is not None:
  245. color = brighten_hex_color(color, 0.4) if color == '#606060' else brighten_hex_color(color, 0.7)
  246. rect2 = Rectangle((label_width + box_width + gap_between_columns, y_pos),
  247. box_width, box_height, color=color, edgecolor='black')
  248. ax.add_patch(rect2)
  250. # Set limits and remove axes
  251. ax.set_xlim(0, total_width)
  252. ax.set_ylim(0, total_height)
  253. ax.axis('off')
  255. # Add column headers
  256. header_y = total_height # Slightly above the top row
  257. ax.text(label_width + box_width / 2, header_y, "Sentence", ha='center', va='bottom', fontsize=24)
  258. ax.text(label_width + box_width + gap_between_columns + box_width / 2, header_y, "Nonwords", ha='center', va='bottom', fontsize=24)
  260. plt.tight_layout()
  261. plt.savefig(cwd / 'Figures' / 'color_legend.png', dpi=300, format='png',
  262. transparent=True)
  264. # %%
  265. # Left to right: Langauge LH, Language RH, MD LH, MD RH
  266. def plot_bar_with_strip(ax, data, palette, hue_order_labels_local, hue_order_plot,
  267. is_grouped=False, title=None):
  268. sns.barplot(x='System', y='EffectSize', hue='Task_Effect', data=data,
  269. ax=ax, legend=False, errorbar='se',
  270. palette=palette, hue_order=hue_order_plot)
  271. sns.stripplot(x='System', y='EffectSize', hue='Task_Effect', data=data,
  272. dodge=True, jitter=jitter, legend=False, size=stripplot_sz,
  273. ax=ax, palette='dark:black',
  274. hue_order=hue_order_plot)
  276. bar_centers = [bar.get_x() + bar.get_width() / 2 for bar in ax.patches]
  277. ax.set_xticks([])
  278. ax.set_xticks(bar_centers)
  279. ax.set_xticklabels(hue_order_labels_local, rotation=90)
  280. ax.set_ylabel("")
  281. ax.set_xlabel("")
  282. ax.tick_params(axis='y', labelsize=8)
  283. ax.tick_params(axis='x', labelsize=6)
  284. ax.set_xlim(ax.patches[0].get_x(), ax.patches[-1].get_x() + ax.patches[-1].get_width())
  286. if title:
  287. ax.set_title(title, fontsize=header_sz-2, pad=5)
  290. # Adjust hue_order labels
  291. hue_order_labels = [label.replace('V6', 'V1b').replace('V5', 'V6').replace('V4', 'V5').replace('V1b', 'V4') for label in hue_order]
  293. # Setup figure
  294. fig, axs = plt.subplots(1, 4, figsize=(6, 2.5), constrained_layout=True, dpi=300, sharey=True,
  295. width_ratios=[1, 1, 1, 1])
  297. # Plot Language LH
  298. plot_bar_with_strip(
  299. axs[0], data_language_lh,
  300. all_colors_expanded,
  301. hue_order_labels,
  302. hue_order,
  303. title='Left Hemisphere'
  304. )
  306. # Plot Language RH
  307. plot_bar_with_strip(
  308. axs[1], data_language_rh,
  309. all_colors_expanded,
  310. hue_order_labels,
  311. hue_order,
  312. title='Right Hemisphere'
  313. )
  315. # Plot MD LH
  316. plot_bar_with_strip(
  317. axs[2], data_md_lh,
  318. all_colors_expanded + ['white', '#20CC00', '#28FF00'],
  319. hue_order_labels + ['H', 'E'],
  320. hue_order + ['placeholder', 'H', 'E'],
  321. title='Left Hemisphere'
  322. )
  324. # Plot MD RH
  325. plot_bar_with_strip(
  326. axs[3], data_md_rh,
  327. all_colors_expanded + ['white', '#20CC00', '#28FF00'],
  328. hue_order_labels + ['H', 'E'],
  329. hue_order + ['placeholder', 'H', 'E'],
  330. title='Right Hemisphere'
  331. )
  333. # Final layout and save
  334. fig.supylabel("Effect size", fontsize=axis_label_sz, x=0.)
  335. fig.savefig(cwd / 'Figures' / 'effects_networks.png', dpi=300, format='png',
  336. bbox_inches='tight', transparent=True)
  339. # %%
  340. # Color legend for MD
  341. import matplotlib.pyplot as plt
  342. from matplotlib.patches import Rectangle
  344. fig, ax = plt.subplots(figsize=(5, 1), dpi=300, constrained_layout=True)
  346. colors = ['#20CC00', '#28FF00']
  347. labels = ['Hard', 'Easy']
  349. for i, (color, label) in enumerate(zip(colors, labels)):
  350. x_pos = i * 3 # spacing between blocks
  351. rect = Rectangle((x_pos, 0), 1.2, 0.4, color=color, edgecolor='black')
  352. ax.add_patch(rect)
  353. ax.text(x_pos + 1.5, 0.2, label, fontsize=24, va='center')
  355. ax.set_xlim(-0.2, 4.2)
  356. ax.set_ylim(-0.2, 0.8)
  357. ax.axis('off')
  359. plt.savefig(cwd / 'Figures' / 'color_legend_MD.png', dpi=300, format='png',
  360. transparent=True)
  363. # %%
  364. # print the number of subjects for each localizer:
  365. for localizer in data_language_lh_localizers['Localizer'].unique():
  366. n_subjects = data_language_lh_localizers[
  367. data_language_lh_localizers['Localizer'] == localizer
  368. ]['Subject'].nunique()
  369. print(f"{localizer}: {n_subjects} subjects")
  371. # %%
  372. # first, only keep tasks_effect where there is 'S' or 'N' in the string, and recode them as 'S' and 'N'
  373. data_language_lh_copy = data_language_lh_localizers.copy()
  374. data_language_lh_copy_s = data_language_lh_copy[
  375. data_language_lh_copy['Task_Effect'].str.contains('S')
  376. ]
  377. data_language_lh_copy_s['Task_Effect'] = 'S'
  378. data_language_lh_copy_n = data_language_lh_copy[
  379. data_language_lh_copy['Task_Effect'].str.contains('N')
  380. ]
  381. data_language_lh_copy_n['Task_Effect'] = 'N'
  382. data_language_lh_copy_s['EffectSize'].mean() / data_language_lh_copy_n['EffectSize'].mean(), data_language_lh_copy_s['EffectSize'].mean(), data_language_lh_copy_n['EffectSize'].mean()
  384. # %%
  385. data_language_lh_copy_s
  387. # %%
  388. # for each Localizer, S/N
  389. for localizer in data_language_lh_localizers['Localizer'].unique():
  390. data_subset = data_language_lh_localizers[
  391. data_language_lh_localizers['Localizer'] == localizer
  392. ]
  393. data_s = data_subset[
  394. data_subset['Task_Effect'].str.contains('S')
  395. ]
  396. data_s['Task_Effect'] = 'S'
  397. data_n = data_subset[
  398. data_subset['Task_Effect'].str.contains('N')
  399. ]
  400. data_n['Task_Effect'] = 'N'
  401. ratio = data_s['EffectSize'].mean() / data_n['EffectSize'].mean()
  402. mean_s = data_s['EffectSize'].mean()
  403. mean_n = data_n['EffectSize'].mean()
  404. print(f"{localizer}: S/N ratio = {ratio:.2f}, Mean S = {mean_s:.3f}, Mean N = {mean_n:.3f}")
  406. # %%
  407. # first, only keep tasks_effect where there is 'S' or 'N' in the string, and recode them as 'S' and 'N'
  408. data_language_rh_copy = data_language_rh_localizers.copy()
  409. data_language_rh_copy_s = data_language_rh_copy[
  410. data_language_rh_copy['Task_Effect'].str.contains('S')
  411. ]
  412. data_language_rh_copy_s['Task_Effect'] = 'S'
  413. data_language_rh_copy_n = data_language_rh_copy[
  414. data_language_rh_copy['Task_Effect'].str.contains('N')
  415. ]
  416. data_language_rh_copy_n['Task_Effect'] = 'N'
  417. data_language_rh_copy_s['EffectSize'].mean() / data_language_rh_copy_n['EffectSize'].mean(), data_language_rh_copy_s['EffectSize'].mean(), data_language_rh_copy_n['EffectSize'].mean()

visualize_data.ipynb at commit 152807b, under MIT · at the source

Overview

Authors: Ruimin Gao1, Chandler Cheung2, Matthew Siegelman3, Alvincé L A Pongos4,5, Hope H Kean6,7, Alyx Tanner8, Evelina Fedorenko6,7, Anna A Ivanova1
ORCID iDs: Ruimin Gao, Alvincé L A Pongos, Anna A Ivanova
  1. School of Psychological and Brain Sciences, Georgia Institute of Technology, Atlanta, GA, United States
  2. Department of Computer and Information Science, University of Pennsylvania, Philadelphia, PA, United States
  3. Department of Psychology, Columbia University, New York, NY, United States
  4. University of California, Berkeley, CA, United States
  5. University of California, San Francisco, CA, United States
  6. Department of Brain and Cognitive Sciences, Massachusetts Institute of Technology, Cambridge, MA, United States
  7. McGovern Institute for Brain Research, Massachusetts Institute of Technology, Cambridge, MA, United States
  8. Department of Psychology, New York University, New York, NY, United States
Institutions: Georgia Institute of Technology (United States); University of Pennsylvania (United States); Columbia University (United States); University of California, San Francisco (United States); University of California, Berkeley (United States); McGovern Institute for Brain Research (United States); Massachusetts Institute of Technology (United States); New York University (United States)
Journal: Imaging neuroscience (Cambridge, Mass.), volume 4, article IMAG.a.1283
Dates: received 7 December 2025; accepted 20 May 2026; published online 25 June 2026
Type: Research article · Language: English
License: CC BY
Identifiers: DOI 10.1162/imag.a.1283 · PMID 42368745 · PMCID PMC13308801 · OpenAlex W7163410756
Open access: diamond, a free copy (OpenAlex)
Status: code verified
Categories: fMRI (modality), human (organism)
Methods: Connectivity, Statistics, Machine learning, Preprocessing, fMRI & imaging, Spectral & time-frequency
Keywords: language network, Multiple Demand network, fMRI, sentence comprehension, task demands
Topic: Neurobiology of Language and Bilingualism (Cognitive Neuroscience, Neuroscience), according to OpenAlex
Funding: McGovern Institute for Brain Research, Massachusetts Institute of Technology; School of Psychology, Georgia Institute of Technology (Startup Fund); Simons Center for the Social Brain, Massachusetts Institute of Technology; Poitras Center for Psychiatric Disorders Research, Massachusetts Institute of Technology; Quest for Intelligence, Massachusetts Institute of Technology
Citations: cited by 4 papers (Europe PMC); 95 references in the paper

Abstract

A network of left frontal and temporal brain areas supports language comprehension and production, implementing computations related to word retrieval and combinatorial linguistic processing. Here, we ask: to what extent are responses to language in this language network stable across task contexts, and how does this stability compare to task sensitivity in the domain-general multiple demand (MD) network? Participants (n = 52) read sentences and nonword lists under six task conditions, including passive reading, reading with a memory probe after each stimulus, and reading and answering questions that require deep semantic engagement. The sentences > nonwords contrast isolated the same set of language-responsive voxels across all tasks; the locations of those voxels were participant-specific, highlighting the value of individual-specific functional localization. We, therefore, conclude that language localization is robust to task variation. We then examined the magnitudes and fine-grained activation patterns in these language-responsive voxels (the language network) and in the domain-general MD network, to test whether task demands modulate linguistic computations and/or recruit a distinct brain system. The language network responded robustly to sentences across all tasks, with somewhat higher responses to semantically engaging tasks. In contrast, the MD network responded to both sentences and nonwords in the presence of a task, which warrants caution when using language paradigms that include task demands, as such paradigms engage two independent networks. A multivariate analysis further revealed that stimulus information is more easily decodable in the language network, whereas task information is more decodable in the MD network. These results suggest that the language and MD networks perform complementary functions during task-driven language comprehension, with the language network primarily extracting information from linguistic input and the MD network determining the appropriate response to the task.

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.

RuiminGao/DiffTasks

License: MIT
State: the link answers, verified on 27 September 2026
Evidence: files inventoried
Commit: 152807bcfc2383f8767ec322b2704aed410ca741, 14 July 2026
Languages: MATLAB (17), Jupyter (12), Python (7), R (5), Shell (1)
Size: 188 files, 42 scripts
Software Heritage: not archived
Found in: “Data and Code Availability”
Holds: README, license file, 17 notebooks
Not found: CITATION.cff, environment file, tests, continuous integration, documentation
Tools: pandas (17 files), NumPy (13 files), Psychtoolbox (9 files), Matplotlib (8 files), seaborn (7 files), emmeans (5 files), lmerTest (5 files), multcomp (5 files), tidyverse (5 files), SciPy (4 files), CONN (3 files), Nilearn (3 files), scikit-learn (2 files), SPM (2 files), broom (1 file), h5py (1 file), statannotations (1 file), statsmodels (1 file)
Availability: 1 check, the latest on 27 September 2026: the link answers
  • 27 September 2026: the link answers
44 files
At the source: github.com/RuiminGao/DiffTasks.git

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;
  • 42 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

Datasets cited

  • osf:zyfmu, at OSF; found in the text, “Design, materials, and procedure”

Data and Code Availability

The experiment scripts, data, codes are available at https://github.com/RuiminGao/DiffTasks.git. The language and the MD parcels are available at https://www.evlab.mit.edu/resources-all/download-parcels.

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, pages, dates, 8 authors, 5 keywords, 5 funders, 94 references.

Cite

This paper

Gao, R., Cheung, C., Siegelman, M., Pongos, A. L. A., Kean, H. H., Tanner, A., Fedorenko, E., & Ivanova, A. A. (2026). The language network responds robustly to sentences across tasks. Imaging neuroscience (Cambridge, Mass.), 4, IMAG.a.1283. https://doi.org/10.1162/imag.a.1283

BibTeX

@article{gao2026language,
author = {Gao, Ruimin and Cheung, Chandler and Siegelman, Matthew and Pongos, Alvincé L A and Kean, Hope H and Tanner, Alyx and Fedorenko, Evelina and Ivanova, Anna A},
title = {{The language network responds robustly to sentences across tasks}},
journal = {Imaging neuroscience (Cambridge, Mass.)},
year = {2026},
month = jun,
volume = {4},
pages = {IMAG.a.1283},
publisher = {MIT Press},
issn = {2837-6056},
doi = {10.1162/imag.a.1283},
url = {https://doi.org/10.1162/imag.a.1283},
pmid = {42368745},
pmcid = {PMC13308801}
}

RIS

TY - JOUR
AU - Gao, Ruimin
AU - Cheung, Chandler
AU - Siegelman, Matthew
AU - Pongos, Alvincé L A
AU - Kean, Hope H
AU - Tanner, Alyx
AU - Fedorenko, Evelina
AU - Ivanova, Anna A
TI - The language network responds robustly to sentences across tasks
T2 - Imaging neuroscience (Cambridge, Mass.)
J2 - Imaging Neurosci (Camb)
PY - 2026
DA - 2026/06/25
VL - 4
SP - IMAG.a.1283
SN - 2837-6056
PB - MIT Press
DO - 10.1162/imag.a.1283
UR - https://doi.org/10.1162/imag.a.1283
LA - en
ER -

CSL-JSON

{
"id": "10.1162/imag.a.1283",
"type": "article-journal",
"title": "The language network responds robustly to sentences across tasks",
"container-title": "Imaging neuroscience (Cambridge, Mass.)",
"author": [
{
"family": "Gao",
"given": "Ruimin"
},
{
"family": "Cheung",
"given": "Chandler"
},
{
"family": "Siegelman",
"given": "Matthew"
},
{
"family": "Pongos",
"given": "Alvincé L A"
},
{
"family": "Kean",
"given": "Hope H"
},
{
"family": "Tanner",
"given": "Alyx"
},
{
"family": "Fedorenko",
"given": "Evelina"
},
{
"family": "Ivanova",
"given": "Anna A"
}
],
"container-title-short": "Imaging Neurosci (Camb)",
"volume": "4",
"page": "IMAG.a.1283",
"DOI": "10.1162/imag.a.1283",
"PMID": "42368745",
"PMCID": "PMC13308801",
"ISSN": "2837-6056",
"publisher": "MIT Press",
"URL": "https://doi.org/10.1162/imag.a.1283",
"language": "en",
"issued": {
"date-parts": [
[
2026,
6,
25
]
]
}
}

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:10.1162/imag.a.1246 [code]
A 3.5-minute-long reading-based fMRI localizer for the language network.
Journal: Imaging neuroscience (Cambridge, Mass.)
In common: Psychtoolbox, lmerTest, h5py, 7 other tools, fMRI, 51 references
[2] doi:10.1038/s41467-026-76598-x
Preserved topography, lateralization, selectivity, and functional connectivity of the language network in older brains.
Journal: Nature communications
In common: fMRI, 30 references
[3] doi:10.1523/jneurosci.0638-25.2026 [code]
The Extended Language Network: Language-Responsive Brain Areas Whose Contributions to Language Remain To Be Discovered.
Journal: The Journal of neuroscience : the official journal of the Society for Neuroscience
In common: fMRI, 27 references
[4] doi:10.1038/s41467-026-75745-8 [code]
A language network in the individualized functional connectomes of 1199 human brains doing arbitrary tasks.
Journal: Nature communications
In common: Nilearn, h5py, seaborn, 5 other tools, 22 references
[5] doi:10.1038/s41467-026-72916-5 [code]
Precision fMRI reveals that the language network exhibits adult-like left-hemispheric lateralization by 4 years of age.
Journal: Nature communications
In common: emmeans, lmerTest, SPM, 1 other tool, fMRI, 21 references
[6] doi:10.1016/j.neuron.2026.04.011 [code]
Precision fMRI reveals densely interdigitated network patches with conserved motifs in the lateral prefrontal cortex.
Journal: Neuron
In common: Nilearn, SPM, scikit-learn, 3 other tools, fMRI, 11 references
[7] doi:10.21203/rs.3.rs-9326213/v1 [code]
Multi-task fMRI outperforms resting-state fMRI for revealing task-invariant organization of the human brain
Journal: Research Square (preprint)
In common: statannotations, Nilearn, SPM, 7 other tools, fMRI, 3 references
[8] doi:10.64898/2026.03.09.710558 [code]
Multi-task fMRI outperforms resting-state fMRI for revealing task-invariant organization of the human brain
Journal: bioRxiv (preprint)
In common: statannotations, Nilearn, SPM, 7 other tools, fMRI, 3 references
[9] doi:10.1093/nc/niag029 [code]
A data-driven approach to identifying and evaluating connectivity-based neural correlates of conscious visual perception.
Journal: Neuroscience of consciousness
In common: broom, emmeans, lmerTest, 10 other tools, 1 reference
[10] doi:10.1038/s41467-026-73895-3 [code]
Sign language narrative reveals universal and modality-specific features of cortical timescale hierarchy.
Journal: Nature communications
In common: Nilearn, statsmodels, seaborn, 4 other tools, fMRI, 6 references

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.