vbot
The vBot robotic manipulandum is able to record movement of the handle and also exert forces on the participant’s hand. When used within its custom 2D virtual reality environment, it can simulate simple virtual worlds and objects with different movement dynamics. Using this equipment, it is easy to assess movement performance and learning of novel tasks.
Here a participant is performing a 2-part movement task. This consists of a lead-in movement followed immediately by the main movement while grasping the handle of the robotic device. We recently showed that lead-in movement (related to backstroke in sports like tennis) that immediately precedes a forward stroke plays an important role in learning. These results could have important implications for both skill learning and movement rehabilitation, since it demonstrates that any immediately preceding movement, such as in a golf stroke or tennis swing, needs to be consistent to achieve fast learning. 

Publications

Howard IS, Ford C, Cangelosi A & Franklin DW (2017)
Activelead-in variability affects motor memory formation and slows motor learning
ScientificReports 7, Article number: 7806(2017) doi:10.1038/s41598-017-05697-z

http://www.ianhoward.de/publications/%20HowForCanFra_SR_2017.pdf

First paper describing 2-part movement tasks given:
HowardIS, Ingram JN, Franklin DW & Wolpert DM (2012),
Gonein 0.6 seconds: The encoding of motor memories depends on recent sensorimotorstates, Journal of Neuroscience 32(37):12756-12768.

http://www.ianhoward.de/publications/HowIngFraWol_JNS_2012.pdf

Details of the robotic manipulandum given:
HowardIS, Ingram JN & Wolpert DM (2009),
Amodular planar robotic manipulandum with end-point torque control, Journal ofNeuroscience Methods 181, 199-211.

http://www.ianhoward.de/publications/HowardJNM2009.pdf