Differences

This shows you the differences between two versions of the page.

--- monkeypsych:monkeypsych_todo [2014/02/18 11:57] – created igorkagan
+++ monkeypsych:monkeypsych_todo [2017/10/24 15:58] – [20140217] lukas.schneider
@@ Line 8: / Line 8: @@
 a.       Use something in between blue and green?
-)      Cue colors for memory reaches are the same for both hands (might be an issue when we want to slightly approach the target colors, for random hands…)
+)      Randomization parameters in the custom_condition_file are ignored, because randomizations happen before the file is run. Especially with randomized reach hands this leads to huge problems, since the colors don’t match the task any more…
-a.        Specify cue colors for all effectors seperately?
+a.      Don't define fractions in the custom conditions file, but rather assign the exact condition directly!
-)      Manual inputs with F1-F5 don’t overwrite the custom conditions from the custom_condition_file, since one is called in ITI, and the other one in INI_trial.
+)      Issues with trial.reach_hand…
-f.e.: reward, target size, …
-a.       Call manual inputs in INI_trial, after custom_conditions as well?
+a.      If not, trial.reach_hand is first overwritten by trial.task.reach_hand (required hand), then again (!) with the hand that released the sensor (chosen hand), which potentially causes problems...
+\\ Not sure if this is still true
-b.      Keep manual inputs and apply them in the next loop ?
+b.       So far, we have some sessions for Magnus (dec 2013), where trial.task.reach_hand and trial.reach_hand are not identical, though the trials were successful and  both sensors were used (rest_hand=[1 1])
-)      Randomization parameters in the custom_condition_file are ignored, because randomizations happen before the file is run. Especially with randomized reach hands this leads to huge problems, since the colors don’t match the task any more…
-a.      Maybe create One get_monkey_dev file for each monkey seperately where we basically combine the get_monkey_dev and custom_conditions files, and call this file twice?
-b.      Else we have to be careful in what parameters we specify in the custom_conditions… (f.e.: microstim_fraction works, but choice_fraction not)
-)      Issues with trial.reach_hand…
-a.       Were there recent changes in the code regarding this parameter?
-b.      If not, trial.reach_hand is first overwritten by trial.task.reach_hand (required hand), then again (!) with the hand that released the sensor (chosen hand), which potentially causes problems...
-c.       So far, we have some sessions for Magnus (dec 2013), where trial.task.reach_hand and trial.reach_hand are not identical, though the trials were successful and  both sensors were used (rest_hand=[1 1])
 We yet could not figure out how this happens, but for recent data it seems to be not the case any more, which is why we wanted to know if there was a recent change