AbstractRecently there has been renewed interest in understanding the different contributions of explicit and implicit learning to motor learning, such as visuomotor rotation adaptation. Visuomotor rotation adaptation also evokes proprioceptive recalibration, but the contribution of explicit and implicit learning to proprioceptive recalibration is unknown. To investigate this, we instructed one group of participants with an explicit strategy to counter the rotation and another group received no such instruction (see Werner et al., 2015). Learning in the explicitly instructed group is almost instantaneous, but settles more slowly in the implicit group. When asked not to apply the learned strategy, the implicit group is not able to do so, whereas the explicit group can. These results both show a typical difference in learning explained by the explicit and implicit instructions. We also recorded estimates of hand position before and after training in each group. Crucially, participants either moved their own hand, enabling them to use (updated) predicted sensory consequences in estimating their hand position, or the apparatus moved their hand, so that only proprioception was available. Active localization leads to larger shifts in localization than passive movements, but the shifts in localization are identical for the implicitly and explicitly instructed groups. The availability of prediction also doesn't interact with instruction. This is clear evidence that proprioceptive recalibration is an exclusively implicit process that does not rely on motor errors, or instructions on how to counter them, but rather on the discrepancy between visual and proprioceptive feedback.
Acknowledgments: Supported by DFG HA 6861/2-1 to BMtH; NSERC to DYPH