search

mobeets / nullSpaceControl

1 stars 0 forks source link

does cloud have to stay the same? #224

Closed mobeets closed 8 years ago

mobeets commented 8 years ago

say the cloud is the same between intuitive and perturbation. is this a requirement for any sampling-based hypothesis?

maybe, by showing results on splitIntuitive ("Exp. 1"), this assumption is no longer relevant, since you'd assume the cloud is the same between the first and second half of intuitive session. and then the fact that the error is only slightly higher (though see #223) is evidence that maybe the cloud is still the same in perturbation?

mobeets commented 8 years ago

fairest comparison seems to be:

  1. fitting 2nd half of perturbation using all of intuitive
  2. fitting 2nd half of perturbation using 1st half of perturbation

with the exception of 20120525, cloud doesn't appear that much worse in (1) compared to (2).

screen shot 2016-07-07 at 5 41 06 pm screen shot 2016-07-07 at 5 41 09 pm screen shot 2016-07-07 at 5 41 13 pm screen shot 2016-07-07 at 5 41 16 pm screen shot 2016-07-07 at 5 41 19 pm