Anna4142 / PrisonerDilemmaPy

1 stars 0 forks source link

Data logging is still inaccurate #83

Open MickyHoltzman opened 4 months ago

MickyHoltzman commented 4 months ago

I hope this is not something I caused. If I did, I am very very sorry :-)

Issues I see with the latest data logger (as included in the consolidating-fixed-strategies Branch:

1) The last trial is not logged 2) When there is a return to center error it is logged with the previous trial number. This may be OK, but it logs trial #0 if it happens on the first trial. 3) Decision making error is logged with the previous trial number. when the trial is completed the decisions are not logged. it still says N/A

if I may, I think logging the trial at the beginning of the next one rather then when its completed only complicates things. I suggest logging at the end of the trial. and write only one line per trial. Instead of using the Trial validity column, use two columns to indicate errors within the trial, one will mark decision timeout (will have yes if there was one) and the other will mark return to center timeout - at the end of the previous trial (or think about it as the mouse did not go to center in time to begin a trial).