CTM real-time fMRI neuromodulation, UAMS (ds003831)

This OpenNeuro release (ds003831) comes from a National Science Foundation funded study at the University of Arkansas for Medical Sciences on how people regulate their own emotional state when guided by real-time fMRI. In a single scan session, healthy adults first viewed affective pictures and recalled the feelings they evoked, then rested, and then tried to reach target levels of valence and arousal while a real-time decoder showed their estimated affective state as a point on a screen. Each person got one run with true feedback and one run with sham feedback, in counterbalanced order. The data are shared under CC0.

Composition

The participants table lists 77 healthy adults, and 74 of them have a subject folder with images (50 women, 24 men, aged 18 to 64). Participants were right-handed, US born, free of current psychiatric disorders and psychotropic medication, and screened negative for drugs of abuse. The table also records the goal order, which feedback run was sham, the participant's guess of the sham run with a confidence rating, and flags for facial EMG, respiration and a known coding error in logged arousal feedback.

Acquisition

All imaging was done on one 3 T Philips Achieva scanner with a 32-channel head coil. The protocol had a T1-weighted MPRAGE at about 1 mm resolution and five BOLD EPI runs (TR 2 s, 3 mm voxels, 37 slices): two picture viewing and recall runs of about 9.4 minutes, a 7.5 minute resting-state run and two neurofeedback runs of about 10.3 minutes. Event files and physiological recordings come with the functional runs.

Annotations

There are no image annotations. Event timings, sham assignment and self-reports are the only labels.

Known limitations

• Three listed participants have no imaging data, one subject has no T1-weighted scan and some subjects miss individual functional runs.
• Facial EMG and respiration were not recorded for every subject.
• The sham feedback replayed a pilot subject's trace, and some logs contain erroneous arousal feedback values.
• The sample comes from one site and one scanner.