Post by MN on Nov 14, 2024 13:43:10 GMT
Dear all,
I am trying to code an experiment somewhat based on the IAT. The main difference is that, although the task is relatively similar in terms of categorical selections, the definitions for said categories is quite subjective and can't be solely based on a single interpretation. As such, to make the test achievable (at the cost of reduced implicitness), I thought up of first having participants categorize the presented stimuli themselves, then use that categorization as definitions to justify between "right" or "wrong" responses throughout the typical IAT trials, including the combined pairs and whatnot.
The issue here is that I can't seem to figure out how to make this happen in practice. We have a set of 20 image stimuli to be used, but had they been predefined this wouldn't pose an issue and I'd go on with the usual syntax and procedures to implement table usage in tasks and parts. What I'd like to know is if it's possible to have "dynamic" tables, in which values of columns are variable rather than concrete; i.e. to be updated as the experiment proceeds. For example, it would start with an undefined or 0 value in all rows for the category column (say, 1 for a category and 2 for the other), but change when the participant is prompted to categorize the stimuli, thus defining a new (or updated) table based on their responses. So if they were to select category 1 for a presented stimulus, the corresponding cell in the table would update accordingly. Then, that table with the new values would be used throughout the rest of the experiment.
Here's an example for what I'm trying to achieve (messy ofc, consider it a pseudo-code):
table stimDefTable
stim1_1 .../img1.png #some undefined value that changes after participant response# #another one that will also change#
stim1_2 .../img2.png #same here# #same again#
stim1_3 .../img3.png #you guessed it# #same again#
...
stim2_1 .../img1.png #woo# #help plz#
stim2_2 ........ so on so forth
part stimDefTableNew
# conditional coding that awaits user input (assigned key selects categories presented to the left or right side of the stimulus, like the IAT),
# stores said input in a variable, assigns variable value to table's corresponding row/column
# OR creates a new table with these values for the same presented stimulus, whichever works better
## code for all other parts to be included in the actual IAT tasks ##
task categorizeStim
# probably open a while loop here
# code to show stimulus and categories to be chosen from (all good, i can code this myself i believe)
part stimDefTableNew
# end the loop when all stimuli has been categorized
task evaluateStim
#same thing
## code for the actual IAT tasks, feedback, blocks, etc ##Is this possible? I looked through the entire forum in this section and the PsyToolkit documentation but couldn't find an answer... Perhaps implementing a questionnaire into the experiment prior to the official tasks' start and using the responses from that survey in the tables (or other assignment method) might work? I feel like that'd still require some sort of dynamic table or array of some sort...
I have some experience with both PsyToolkit and other programming languages (like C/C++ and Python), hence why I don't mind complicated/advanced answers or implementations. It's a nice challenge as well, and would love to know if this is possible at all or if we should just try something else. ^^"
Looking forward to an answer!
Best,
MN