Counterbalancing and [correct_response] in logfile

kirbyj

Hi everyone,

did what I hope was a thorough forum search before posting this, but this behavior seems very strange to me and I can't figure out why it should be so...

I'm trying to implement a very simple counterbalanced response rule. Take as an example the lexical-decision task example experiment, as referenced here, which also says:

Now, in your block_loop, instead of setting correct_response to a fixed value, you set it to a variable: [verb_response] or [noun_response]. 

OK; in the lexical-decision task experiment, that means [word_reponse] or [nonword_response] .

What I don't understand is why correct_response doesn't get interpreted when it is logged: you end up with a logfile that looks like this (irrelevant variable columns removed):

I know that I can get whether or not the trial was correct from the correct column. But what I don't understand is why nonword_response and word_response are logged as the values that they take on (depending on the value of subject_parity , as set in the inline script), but the value of correct_response is not. Is this the expected behavior? In other words, in the above logfile, I would expect the values in the correct_response column to be m where it currently shows [word_response] and z where it shows [nonword_response] , and for these to be reversed when subject_parity is odd .

James

eduard

Hi James,

This is indeed to intended behaviour. One reason this happens could be that you have accidentally put quotation marks around [word_response]. However, without seeing the experiment, it is hard to give the precise reason. So, share it if you like to find out.

Eduard