作者: Felix C. Huang , James L. Patton , Ferdinando A. Mussa-Ivaldi
DOI: 10.1109/ICORR.2009.5209528
关键词: Viscosity 、 Dynamics (mechanics) 、 Inertia 、 Radius 、 Biomechanics 、 Mathematics 、 Control theory 、 Feedforward neural network 、 Feed forward 、 Impedance control
摘要: We investigated how learning of inertial load manipulation is influenced by movement amplification with negative viscosity. Using a force-feedback device, subjects trained on anisotropic loads (5 orientations) free movements in one three conditions (inertia only, viscosity or combined), prior to common evaluation (prescribed circular pattern inertia only). Training Combined-Load resulted lower error (6.89±3.25%) compared Inertia-Only (8.40±4.32%) and Viscosity-Only (8.17±4.13%) according radial deviation analysis (% trial mean radius). groups exhibited similar unexpected no-load trials (8.38±4.31% versus 8.91±4.70% radius), which suggests comparable low-impedance strategies. These findings are remarkable since viscosity, only available during training, evidently enhanced when combined inertia. Modeling that feedforward after-effect cannot predict such performance gains. Instead, results from training consistent greater compensation along small increase impedance control. The capability the nervous system generalize an intriguing new method for enhancing sensorimotor adaptation.