, 2004). Figure 1 (bottom) summarises the experimental procedure. We applied 1-Hz rTMS at 110% of RMT (Kantak et al., 2010a) with a Magstim 70 mm figure-of-eight coil attached to a Magstim Rapid2 stimulator. For the Control–dPM and Probe–dPM groups, a total of 600 pulses (10 min) at 110% of RMT were applied over dPM of the contralateral hemisphere.

The Probe–M1 group received 600 pulses at 110% of RMT over contralateral primary motor cortex (M1). We used a structural MRI-based stereotaxic frameless neuronavigation system (Brainsight; Rogue Research) to precisely localise the TMS coil over dPM. Individual structural brain scans were acquired in 15 out of 20 dPM participants prior to the experiment. For those without individual brain

scans, we applied rTMS 2.5 cm anterior to the hot spot of FDI as a previously identified anatomical location of dPM (Gerschlager et al., 2001; Selleckchem PI3K inhibitor Rizzo et al., 2004). Primary task performance was quantified as movement time (time to finish the four key presses). We averaged movement time across 12-trial blocks. We removed trials with inaccurate responses (premature start, incorrect order or incorrect number of key presses). Approximately 12% of practice trials (~ 17 out of 144) and 8% of retention trials (~ 1 out of 12) were discarded from the movement time analysis. The error rates were similar across groups. Secondary task (audio–vocal Inhibitor Library supplier reaction time task) performance was measured as RT, the time interval

between the onset of the stimulus Mirabegron and the vocal response. RT was measured under the single-task condition for every participant before task practice began. Dual-task cost was computed by subtracting RT measured under the single-task condition (performing audio–vocal RT task alone) from RT measured under probe condition (performing both audio–vocal RT and finger sequence tasks). To evaluate how well participants had learned the finger sequence, we computed ‘forgetting’, i.e. the difference in movement time between immediate and delayed retention tests (Fig. 1, bottom; R2 – R1). A positive value in forgetting indicates an increase in movement time from immediate to delayed retention, suggesting participants demonstrate some degree of forgetting of the learned skill. In contrast, a negative value in forgetting suggests a decrease in movement time from immediate to delayed retention, indicating an off-line gain in skill. Thus, a smaller value in forgetting indicates superior learning of the motor skill than does a larger value. Forgetting data was analysed with one-way anova. To test our hypotheses, prior planned post hoc comparisons were done to examine the difference between (i) Control–NoTMS vs. Probe–NoTMS and (ii) Probe–dPM vs. Probe–NoTMS vs. Probe–M1.