Participants  

English participants were 49 young adults (30 females, mean age=21.3, SD=3.6) with no history of psychiatric, neurological or other medical illness that might compromise cognitive functions. They self-identified as native English speakers, and strictly qualified as right-handed on the Edinburgh handedness inventory. All participants were paid, and gave written informed consent prior to participation, in accordance with the IRB guidelines of Cornell University. 

Chinese participants were 35 healthy, right-handed young adults (15 females, mean age=19.3, SD=1.6).  They self-identified as native Chinese speakers, and had no history of psychiatric, neurological, or other medical illness that could compromise cognitive functions. All participants were paid, and gave written informed consent prior to participation, in accordance with the IRB guidelines of Jiangsu Normal University. 

French participants were 28 healthy, right-handed adults (15 females, mean age=24.4, SD=4.6). They self-identified as native French speakers and had no history of psychiatric, neurological, or other medical illness that could compromise cognitive functions. All participants gave written informed consent prior to participation, in accordance with the Regional Committee for the Protection of Persons involved in Biomedical Research.

Experiment Procedure

After giving their informed consent, participants were familiarized with the MRI facility and assumed a supine position on the scanner. They were instructed to not move as best as they could throughout scanning as movement would make the scans unusable. Next, participants were put in the head-coil with pillows under and on the sides of their head and under the knees for comfort and to reduce movement over the scanning session. Participants were given a bulb in their right hand and told to squeeze if something was wrong or they needed a break during scanning. Once in place, participants chose an optimal stimulus volume by determining a level that was loud but comfortable. Auditory stimuli were delivered through MRI-safe, high-fidelity headphones inside the head coil (English: Confon HP-VS01, MR Confon, Magdeburg, Germany; Chinese: Ear Bud Headset, Resonance Technology, Inc, California, USA; French: Magnacoil TIM headset, Siemens, Germany). The headphones were secured against the plastic frame of the coil using foam blocks. 

The English and Chinese participants went through one scanning session, which was divided into 9 runs, and each lasted for about 10 minutes. Participants listened passively to 1 section of the audiobook in each run and completed 4 quiz questions after each run (36 questions in total). These questions were used to confirm their comprehension and were viewed by the participants via a mirror attached to the head coil and they answered through a button box. During scanning, participants were monitored by a camera over their left eye. If they appeared drowsy or seemed to move too much during the movie, the operator of the scanner gave them a warning over the intercom by producing a beep or speaking to them. During breaks between the runs, participants were told that they could relax but not move. Finally, participants were paid and sent home. The entire session lasted for around 2.5 hours. In French, due to a legal limitation, participants could not stay for longer than 1.5 hours inside the scanner; therefore, the acquisition was split into two sessions separated by a period of 1 to 2 hours out of the scanner.

Stimuli

The English audiobook is 94 minutes long, translated by David Wilkinson and read by Karen Savage. The Chinese audiobook is 99 minutes long, read by a professional female Chinese broadcaster hired by the experimenter. The French audiobook is 97 minutes long, read by Nadine Eckert-Boulet and published by the now-defunct Omilia Languages Ltd. The original French text is copyrighted by Gallimard 1946.

One of the central themes in the story is the difference between adults and children, especially the lack of imagination in the former. The narrator uses the visual cues of different drawings to emphasize this message and these drawings are present in the original text. In the English and Chinese study, to help the participants understand this point, these visual cues were incorporated during the audio presentation for the first chapter and are included in the OpenNeuro repository. In order to control for the visual stimuli and its associated neural activation, "picture events" conditions and "picture blocks" conditions are also included in the analysis to account for the visual stimuli presented to participants and its associated neural activation. The "picture events" occur at the 10 s, 35 s, and 60 s timepoints in the first section of the story while the "picture blocks" also occur at the 10 s, 35 s, and 60 s timepoints in the first section and last for 15 s, 20 s, and 15 s respectively. These conditions match the presentation and duration of the visual stimuli and are aligned with particular plot points in the story.

Acquisition

English and Chinese MRI images were acquired with a 3T MRI GE Discovery MR750 scanner with a 32-channel head coil. French MRI images were acquired with a 3T Siemens Magnetom Prisma Fit 230 scanner. Anatomical scans were acquired using a T1-weighted volumetric Magnetization Prepared Rapid Gradient-Echo (MP-RAGE) pulse sequence. Functional scans were acquired using a multi-echo planar imaging (ME-EPI) sequence with online reconstruction (TR=2000 ms; English and Chinese: TEs=12.8, 27.5, 43 ms; French: TEs=10, 25, 38 ms; FA=77◦; matrix size=72 x 72; FOV=240.0 mm x 240.0 mm; 2112x image acceleration; English and Chinese: 33 axial slices; French: 34 axial slices; voxel size=3.75 x 3.75 x 3.8 mm).

Preprocessing

MRI data files were converted from DICOM to NIfTI format and preprocessed using AFNI version 16.
Anatomical. The anatomical/structural MRI scans were deskulled using "3dSkullStrip". The resulting anatomical images were nonlinearly aligned to the Montreal Neurological Institute (MNI) N27 template brain. Resulting anatomical images were used to create grey matter masks. 
Functional. The first 4 volumes in each run were excluded from analyses to allow for T1-equilibration effects. The fMRI timeseries were then corrected for slice-timing differences ("3dTshift") and despiked ("3dDespike"). Next, volume registration was done by aligning each timepoint to the mean functional image of the centre timeseries ("3dvolreg"). Then the volume-registered and anatomically-aligned functional data were nonlinearly aligned to the MNI template brain. Multi-echo independent components analysis (ME-ICA) were used to denoise data for motion, physiology and scanner artifacts. Images were then resampled at 2 mm cubic voxels ("3dresample").

Code availability

Presentations scripts for all three languages along with code for technical validation are included in the repository.