Timescales of motor memory formation in dual-adaptation
Fig 5
Comparison between the 3 participants best-fit by the dual-rate model (left side) and the 7 participants best-fit by the triple-rate model (right side).
A. Mean of force compensation for the dual-rate best-fit group of participants. B. Force compensation for the triple-rate best-fit group of participants. C-F. Simulation of the dual and triple-rate models (weighted switch) from the fitted parameters of the dual-rate best-fit group (C and D respectively) and the triple-rate best-fit group (E and F respectively). The models were fit to the mean participant data rather than individual participants. Here, the de-adaptation phase contained 80 (C, D) and 208 (E, F) trials corresponding to the average trials number for the dual-rate best-fit and triple-rate best-fit groups respectively. The total output for each contextual cue thick red and blue lines) is composed of the summation of each process: fast (solid lines), slow (dotted lines) and ultraslow (dashed) process. G. BIC differences between dual-rate and triple-rate models for both groups. The red dashed line shows a BIC difference of 6 from the best-fit model.