Skip to main content
Dryad

Pupil size reflects activation of subcortical ascending arousal system nuclei during rest

Cite this dataset

Lloyd, Beth; de Voogd, Lycia; Mäki-Marttunen, Verónica; Nieuwenhuis, Sander (2023). Pupil size reflects activation of subcortical ascending arousal system nuclei during rest [Dataset]. Dryad. https://doi.org/10.5061/dryad.7m0cfxpzn

Abstract

Neuromodulatory nuclei that are part of the ascending arousal system (AAS) play a crucial role in regulating cortical state and optimizing task performance. Pupil diameter, under constant luminance conditions, is increasingly used as an index of activity of these AAS nuclei. Indeed, task-based functional imaging studies in humans have begun to provide evidence of stimulus-driven pupil-AAS coupling. However, whether there is such a tight pupil-AAS coupling during rest is not clear. To address this question, we examined simultaneously acquired resting-state fMRI and pupil-size data from 74 participants, focusing on six AAS nuclei: the locus coeruleus, ventral tegmental area, substantia nigra, dorsal and median raphe nuclei, and cholinergic basal forebrain. Activation in all six AAS nuclei was optimally correlated with pupil size at 0- to 2-second lags, suggesting that spontaneous pupil changes were almost immediately followed by corresponding BOLD-signal changes in the AAS. These results suggest that spontaneous changes in pupil size that occur during states of rest can be used as a noninvasive general index of activity in AAS nuclei. Importantly, the nature of pupil-AAS coupling during rest appears to be vastly different from the relatively slow canonical hemodynamic response function that has been used to characterize task-related pupil-AAS coupling.

Methods

This dataset contains resting-state fMRI data and simultaneous pupil recordings for 72 individuals across two sessions of 5 minutes.

MRI data

MRI data were acquired using a Siemens MAGNETOM Prisma 3T MR scanner. All images (functional and structural) have been defaced for anonymity purposes.

  • functional EPI images (150) per session (ses-day1, ses-day2), 5 minutes resting-state (no task).
    • T2*-weighted BOLD images were recorded using a customized multi-echo EPI sequence with ascending slice acquisition (58 axial slices; TR = 2 s; TE = 14.4, 39.1 ms; partial Fourier = 6/8; GRAPPA acceleration factor = 2; multiband acceleration factor = 2; flip angle = 65°; slice matrix size 104 x 104 mm; slice thickness = 2.0 mm; FOV: 208 x 208 mm; slice gap = 0; bandwidth: 2090 Hz/px; echo spacing: 0.56 ms). 
    • the only preprocessing carried out on the functional EPI images is that these images have been realigned to the first functional image (using the realignment parameters, see: nuisance regressors 26–32), and each functional image is the voxel-wise weighted sums of both echoes, which were calculated using in-house scripts, based on local contrast-to-noise ratio.
    • the first five scans have been removed.
      • directory:
        • data\sub-001\ses-day1\func\sub-001_ses-day1_task-rest_acq-normal_run-01_bold
        • data\sub-001\ses-day2\func\sub-001_ses-day2_task-rest_acq-normal_run-01_bold
  • structural T1-weighted MRI scan.
    •  The structural T1-weighted image (0.9 mm isotropic) was acquired using a T1-weighted 3D MP-RAGE (TR = 2.3 s; TE = 2.32 ms; flip angle = 8°, FOV = 256 x 256 x 230 mm).
      • directory: data\sub-001\ses-day2\anat\sub-001_ses-day2_acq-highres_run-02_T1w
  • structural fast-spin echo (FSE) scan (for locus coeruleus localisation [LC])
    •  A neuromelanin-sensitive structural scan was acquired for delineation of the LC (11 axial slices, TR = 750 ms, TE = 10 ms, flip-angle = 120°, bandwidth = 220 Hz/Px, slice thickness = 2.5 mm, slice gap = 3.5 mm; in-plane resolution = 0.429 x 0.429 mm).
      • directory: data\sub-001\ses-day2\anat\sub-001_ses-day2_acq-fse_run-03_T1w
  • nuisance regressors:
    • nuisance regressors 1–26: heart rate pulse + breathing regressors
      • Raw pulse was preprocessed using PulseCor (https://github.com/lindvoo/PulseCor) implemented in Python for artifact correction and peak detection. Fifth-order Fourier models of the cardiac and respiratory phase-related modulation of the BOLD signal were specified (Van Buuren et al., 2009), yielding 10 nuisance regressors for cardiac noise and 10 for respiratory noise. Additional regressors were calculated for heart rate frequency, heart rate variability, (raw) abdominal circumference, respiratory frequency, respiratory amplitude, and respiration volume per unit time (Birn et al., 2006), yielding a total of 26 RETROICOR regressors (https://github.com/can-lab/RETROICORplus).
    • nuisance regressors 26–32: realignment parameters
      • six movement parameter regressors (3 translations, 3 rotations) derived from rigid-body motion correction, high-pass filtering (1/128Hz cut-off) and AR(1) serial autocorrelation corrections.
    •  directory:
      • data\sub-001\ses-day1\func\sub-001_ses-day1_task-rest_acq-normal_run-01_bold\log
      • data\sub-001\ses-day2\func\sub-001_ses-day2_task-rest_acq-normal_run-01_bold\log

Pupil data:

    • raw pupil .ascii file for the right eye collected at 250 Hz with EyeLink 1000 Plus (SR Research, Osgoode, ON, Canada)
    • The eye-tracker was placed at the end of the scanner bore, such that the participant’s right eye could be tracked via the head coil mirror. Before the start of each resting-state session, we began with a calibration of the eye-tracker using the standard five-point EyeLink calibration procedure.
    • The pupil data is raw; however, moments when the eye-tracker received no pupil signal (i.e., during eye blinks) were marked (-1.00) automatically during acquisition by the manufacturer's blink detection algorithm.
      • directory:
        • data\sub-001\logfiles-ses-day1\raw_pup
        • data\sub-001\logfiles-ses-day2\raw_pup

Demographic data:

    • subject data (linked to subject number) containing age and gender
      • file: data\group_data\sub_demographics.csv

 

Additional files included in version 2:

JSON files:

    • json files for: 
      • T1: data\JSON\sub-MRIFCWML001_ses-day2_acq-highres_run-02_T1w.json
      • EPI echo 1 + echo 2: data\JSON\func_echo1_seq_json_file.json, data\JSON\func_echo2_seq_json_file.json
      • FSE scan: data\JSON\FSE_seq_json_file.json

 

Changes from version 1 to version 2:

  • Pupil data in version 1 had underegone preprocessing steps (interpolated over blinks and downsampled to 50 Hz) 
  • Pupil data in version 2 is the raw pupil data (no interpolation and sampled at 250 Hz). 

Funding

Dutch Research Council, Award: VI.C.181.032

Templeton World Charity Foundation, Award: 0366