Skip to main content
. Author manuscript; available in PMC: 2013 Jun 11.
Published in final edited form as: Neural Netw. 2012 Feb 13;0:60–69. doi: 10.1016/j.neunet.2012.02.002

Fig. 6.

Fig. 6

Effect of temporarily suppressing output from strongly connected cells. The network was the same as that considered in Fig. 4, with 800 strongly-connected cells in M1 (cells 1–800 in B and C), and 200 weakly connected cells in SMA (cells 801–1000 in B and C). A: force production versus movement attempt without suppression (solid) and with suppression of the M1 cells (dashed) for 10,000 movement attempts (dashed). The solid horizontal line at the top shows the maximum force the network is capable of achieving. Temporarily suppressing the output of the M1 cells allowed the network to learn to produce more force. B: cell activations after 10,000 movement attempts, before suppression. The simulated SMA cells (cells 801–1000) were not well optimized because of their weaker influence on the motoneuronal pools C: cell activations after 10,000 more movements with M1 suppressed. The simulated SMA cells (cells 801–1000) were better optimized after movement practice without the M1 cells present. For these plots σ = 0.02.