Combining reward and M1 transcranial direct current stimulation enhances the retention of newly learnt sensorimotor mappings

Abstract

Background Reward-based feedback given during motor learning has been shown to improve the retention of the behaviour being acquired. Interestingly, applying transcranial direct current stimulation (tDCS) during learning over the primary motor cortex (M1), an area associated with motor retention, also results in enhanced retention of the newly formed motor memories. However, it remains unknown whether combining these distinct interventions result in an additive benefit of motor retention. Methods We investigated whether combining both interventions while participants learned to account for a visuomotor transformation results in enhanced motor retention (total n = 56; each group n = 14). To determine whether these interventions share common physiological mechanisms underpinning learning, we assessed motor cortical excitability and inhibition (i.e. SICI) on a hand muscle before and after all participants learned the visuomotor rotation using their entire arm and hand. Results We found that both the Reward-Stim (i.e. reward + tDCS) and Reward-Sham (i.e. reward-only) groups had increased retention at the beginning of the retention phase, indicating an immediate effect of reward on behaviour. However, each intervention on their own did not enhance retention when compared to sham, but rather, only the combination of both reward and tDCS demonstrated prolonged retention. We also found that only the Reward-Stim group had a significant reduction in SICI after exposure to the perturbation. Conclusions We show that combining both interventions are additive in providing stronger retention of motor adaptation. These results indicate that the reliability and validity of using tDCS within a clinical context may depend on the type of feedback individuals receive when learning a new motor pattern.

Modulation of neural activity in frontopolar cortex drives reward-based motor learning

“…In tDCS studies, there is evidence that changes in motor cortical excitability induced by M1 or SMA tDCS can have a positive effect on different aspects of motor learning, including explicit or implicit sequence learning and visuomotoradaptation tasks [18,15,19,20]. Reward-based motor learning, however, had not been the focus of tDCS investigations until recently [28]. This study demonstrated that combining reward signals with M1-tDCS could improve motor learning, as reflected in the retention of motor skills.…”

“…The outcome was, however, not entirely unexpected, as standard M1-tDCS protocols generally assess effects on motor retention or consolidation [43,19,44,15] and have so far not reported modulations of motor learning processes that depend on reward signals. One notable exception is a recent anodal tDCS study that used smaller electrodes anterior and posterior to M1 [28]. This work demonstrated a benefit of the combination of reward signals and M1-tDCS on motor retention during a motor adaptation paradigm.…”

“…This study did not, however, identify a faster learning rate under M1-tDCS, which was the effect we observed in the rFPC-tDCS protocol. The authors in [28] argued that reward signals combined with tDCS may influence motor retention through long-term potentiation mechanisms. It is therefore likely that an effect of M1-tDCS on reward-based motor learning primarily affects retention of motor skills and not initial learning, as our task was designed to assess.…”

“…Nineteen right-handed participants (10 females, mean = 27.7 yrs, std = 3.3 yrs, range [21][22][23][24][25][26][27][28][29][30][31][32][33] with no musical training were recruited. Laterality quotient was assessed by the Oldfield handedness inventory ( [49]; mean = 90, SEM = 3.2; values available in 17/19 participants).…”

“…In this scenario, the optimal strategy to maximize the mean total reward would be to keep track of the action-reward associations during initial exploration, and to swiftly switch to exploitation of the performance inferred as most rewarding. As control tDCS condition we modulated contralateral motor cortex activity using contralateral (left) M1 tDCS -shown to positively influence different components of motor skill learning with a less established modulation of reward processing [19,20,28]. Each active tDCS condition was contrasted to sham stimulation.…”