These are the motor imagery datasets
biceps_mi folder includes recordings from five muscles (biceps, triceps, FDI, APB, and ADM) (~18k MEPs). The biceps and triceps have shorter MEP latencies (incorrect marked from DELMEP - see below).
All MEPs were recorded from a single general hotspot.
The interstimulus interval was fixed at 7 seconds (which, in hindsight, was a methodological mistake 😭).
There are six conditions:
- B1 - baseline before task presentation;
- KI - kinesthetic motor imagery;
- V1 - first-person visual motor imagery;
- V3 - third-person visual motor imagery;
- BM - baseline with a metronome (because I paced the MI peak, similar to the real movement training task, using the metronome beat: https://www.nature.com/articles/s41598-026-42691-w);
- B2 - resting baseline after all other conditions.
I also have a second dataset (tdcs_mi folder) from my second experiment. The tasks are basically the same, but each of the 15+ participants completed three tDCS sessions over the primary visual cortex (anodal, cathodal, and sham).
In this study, the interstimulus interval was randomized and approximately evenly distributed between 5, 6, and 7 seconds, if I remember correctly. In total, there are roughly 45000+ MEPs.
This dataset includes only the FDI, APB, and ADM muscles.
I've also run DELMEP ( https://www.nature.com/articles/s41598-023-34801-9 ) on these data. And I have in each top folder .xsls results table (with p2p, rms pre etc and latency there).
!!!However, please note that the latency of biceps and triceps in the dataset are wrong (possibly bcs of the DELMEP limitations)!!!