Somatotopic organization of brainstem analgesic circuitry
Data files
Aug 11, 2025 version files 234.67 GB
-
adu8846_Pain_Ratings_and_Demography.xlsx
15.22 KB
-
Defaced_Arm_Placebo_Nonresponders.7z
45.61 GB
-
Defaced_Arm_Placebo_Responders.7z
41.01 GB
-
Defaced_Face_Placebo_Nonresponders.7z
39.11 GB
-
Defaced_Face_Placebo_Responders.7z
37.35 GB
-
Defaced_Leg_Placebo_Nonresponders.7z
27.68 GB
-
Defaced_Leg_Placebo_Responders.7z
43.91 GB
-
DisciplineSpecificMetadata.json
9.59 KB
-
README.md
5.36 KB
Abstract
Rationale: By conducting a placebo analgesia conditioning paradigm using a falsely labelled and described cream, 93 pain-free participants believed this cream could reduce the intensity of painful heat applied to either their face, arm, or leg. In reality, this “lidocaine” cream was a placebo, and the temperature of a thermode surreptitiously lowered whenever placed onto this cream relative to a control Vaseline cream site. During collection of images of brainstem activity using a 7-Tesla Magnetic Resonance Imaging scanner, both the placebo and control cream sites received identical temperatures, with a decrease in perceived pain intensity during the placebo cream site stimulation being a placebo analgesia response. We analysed activity changes in the PAG and an area that the PAG contacts to inhibit pain, the rostral ventromedial medulla (RVM), during analgesia responses on the face, arm, and leg. Further, using the hidden application of a third cream to a separate body site, we explored whether analgesia was restricted to the area of stimulation or whether it was more widespread.
Results: Our findings show that analgesia evoked on the face, arm, and leg changes activity in a somatotopically organized fashion, i.e. face analgesia is towards the rostral (top) of the PAG and RVM, arm and leg analgesia represented more caudally (lower). These somatotopically organized analgesic responses occurred on opposite sides of the PAG but at the same locations as pain activity changes, and the analgesia was restricted to the site where conditioning was applied.
Dataset DOI: 10.5061/dryad.np5hqc064
These data accompany the submission "Somatotopic Organization of Brainstem Analgesic Circuitry" - including behavioural pain and expectancy ratings, and raw 7-Tesla brain imaging datasets for conditioned placebo analgesia evoked on the face, arm, or leg. This readme file explains the content of each folder, and any further questions regarding the data can be sent directly to the first or corresponding author.
Description of the data and file structure
Human 7-Tesla functional MRI data of conditioned placebo analgesia evoked on the face, arm, or leg.
Accompanying Excel provides average pain intensity data during each fMRI sequence, as well as demographic data including age and sex.
Functional sequences consisted of a block design with 134 dynamic scans included in each 4D nifti file for responder and nonresponder groups of each skin site. Each sequence consisted of eight blocks of pain stimuli, separated by inter-trial intervals.
Model timings: 6-38 = Baseline, 39-42 = Stimulus 1, 52-55 = Stimulus 2, 64-67 = Stimulus 3, 77-80 = Stimulus 4, 89-92 = Stimulus 5, 102-105 = Stimulus 6, 115-118 = Stimulus 7, 127-130 = Stimulus 8.
A canonical HRF curve was applied to all data prior to the generation of 1st level GLMs, with a TR = 2.5 seconds
As part of the data upload, all participants' fMRI series went through a defacing procedure, whereby a manual mask was derived for each participant encompassing identifiable facial structures, inverted, and applied as an inclusion mask to each individual 4D series, thereby masking out all facial features.
Files and variables
File: adu8846_Pain_Ratings_and_Demography.xlsx
Description: Pain intensity ratings and demography for each participant in the responder or nonresponder group for placebo analgesia evoked on the right side of the face, arm, or leg. During each functional sequence, the position of a thermal probe was moved between a portion of these skin sites where either a control or inert placebo "lidocaine" cream had been applied.
File: Defaced_Leg_Placebo_Responders.7z
Description: Leg Responder group, organized into separate "control" and "placebo" 4D Nifti files for the two independent fMRI sequences.
File: Defaced_Leg_Placebo_Nonresponders.7z
Description: Leg Nonresponder group, organized into separate "control" and "placebo" 4D Nifti files for the two independent fMRI sequences.
File: Defaced_Face_Placebo_Nonresponders.7z
Description: Face Nonresponder group, organized into separate "control" and "placebo" 4D Nifti files for the two independent fMRI sequences.
File: Defaced_Arm_Placebo_Nonresponders.7z
Description: Arm Nonresponder group, organized into separate "control" and "placebo" 4D Nifti files for the two independent fMRI sequences.
File: Defaced_Face_Placebo_Responders.7z
Description: Face Responder group, organized into separate "control" and "placebo" 4D Nifti files for the two independent fMRI sequences.
File: Defaced_Arm_Placebo_Responders.7z
Description: Arm Responder group, organized into separate "control" and "placebo" 4D Nifti files for the two independent fMRI sequences.
File: DisciplineSpecificMetadata.json
Description: Information regarding ethical approval and point of contact for data queries, MRI sequence acquisition, and experimental design structure.
Variables
- Ongoing pain intensity
Code/software
Respiratory and Physiological noise modelling and removal: https://www.sciencedirect.com/science/article/pii/S1053811912000845
Synthetic Distortion Correction: https://github.com/MASILab/SynBOLD-DisCo
Brainstem isolation and normalization: https://www.diedrichsenlab.org/imaging/suit.htm
Image processing pipeline:
- Slice timing correction
- Motion correction
- Physiological and respiratory noise modelled and removed
- Linear detrending
- Motion parameter noise removed
- 1mm isotropic reslice
- T1 coregistration
- Synthetic distortion correction
- SUIT brainstem isolation
- SUIT normalization
- 0.5mm isotropic reslice
- Spatial smoothing 1mm FWHM
Steps 1, 2, 4, 5, 6, 7, 11, and 12 can be performed in any standard installation of Matlab with the statistical parametric mapping (SPM12) toolbox installed
Step 3 requires the DRIFTER toolbox linked above
Step 8 requires the SynBOLD DisCO pipeline linked above
Steps 9,10 require the SUIT toolbox linked above
Access information
Data was derived from the following sources:
- These data were collected at a research imaging facility using a 7-Tesla MRI scanner by the first and corresponding author
Human subjects data
Consent was attained for future use of de-identified functional imaging data, in this instance uploaded to the public domain.
Image de-identification was performed through DICOM to Nifti conversion, stripping the header information from individual participant data.
Participants have been coded as "participant 1, 2, ..." in both nifti imaging and behavioural datasets.