Summary: weekly mtg 20170308 (Byron, Steve, Matt, me) - mobeets/nullSpaceControl GitHub Wiki

SSS: mention in text binning by cursor vel

range per dimension the same in the marginal hists?

add manuscript commentary any time something looks weird:

  • e.g., min firing having spikiness in dim 1
    • and why is it only spiky in dim1 and not dim2
    • maybe flip sign of dim1 so that min firing is to the left
  • and uniform hists not looking uniform

re-do SSS analysis by making sure there's a similar amount of points in each output-potent bin?

make SSS Fig. 6C with black as intuitive data

within each SSS bin, see how similar the speeds are in intuitive vs. perturbation

  • also just look at this overall

SSS: need to match intent, not actual movement

  • so, say, cursor-target angle
  • and also also movements only in the center of the workspace

SSS-new-analysis: the joint

  • same analysis, only no conditioning
  • variance in SSS over all movements in intuitive vs. perturbation
  • might want to try to match targets and speeds

Steve: still concerned about uneven sampling based on spending more time in one direction during perturbation, say

discussion: need to discuss unevenness in movement directions, speeds, etc., across mappings