This dataset was collected to study the reinforcement learning process across individuals at the behavioral and neural level. The 2-step task is an established task to study the Markov decision-making process and here we used a variant of the task (Cockburn et al., 2024). We paired the task with the functional MRI to study the common and distinct decision-making processes across individuals using different strategies in this task. Please refer to the preprint for the task details.
The participant IDs of the task are coded as sub-MBMFxxxx01 or sub-MBMFxxxx02. The four digits in the middle are the participants IDs. And the '01' or '02' at the end denotes whether participants performed the task first or second, as this 2-step task was paired with another task (which studies habit) for each participant within the same MRI session.
The MRI scans were collected from 189 healthy controls, who did not report any history of substance/alcohol use disorder, anxiety disorder (Obsessive-Compulsive Disorder, Body Dysmorphic Disorder, generalized anxiety, social anxiety/social phobia), and/or depressive disorders (dysthymia, major depression). Furthermore, participants reported they did not use any medications for subclinical psychiatric disorders.
- Notes on the data:
- Participant with the following ID does not have the fieldmaps for the functional task run-1, but only has those for the functional task run-2 (for your consideration of the susceptibility distortion correction):
sub-MBMF025002
- Participants with the following ID(s) only experience state-contingent reward (different from the rest of the participants' pool, which experienced both state-contingent and stimulus-contingent rewards):
sub-MBMF021501 sub-MBMF021601 sub-MBMF021701 sub-MBMF021801 sub-MBMF021901
- Contact:
John P. O'Doherty, jdoherty@caltech.edu
Weilun Ding, wdding@caltech.edu