|
aflex23
|
|
|
Group: Forum Members
Posts: 4,
Visits: 17
|
Hello Millisecond Forum, First time poster and not very experienced with Inquisit, I am wanting to adapt the Winsconsin Card Sorting Test. I'm willing to do an experiment with Inquisit 6 (Lab version). My experiment would have one WCST as downloadable here: https://www.millisecond.com/download/library/cardsortI have then coded 4 other versions in which the only difference is the stimuli change (i.e file directory, I slightly modified the .gif files). For now, the best solution that I have found to have participants do the five tasks is that I've created my stimuli, named them exacly as they are in the original version, and run each version from separate folders (which gives me total 5 data sheet per participant and takes some time + manual randomization of order). The tasks work well but it's not very convenient. The most convenient solution for me would be to have the following parameters within one script: - all five versions - order of versions randomized from participant to participant - within each version: random order of rule (number, color, shape). Would it be possible to achieve this or is it too complicated for a beginner ? Since one of the output files implies score calculations based on trial order, I'm afraid that following my idea might break the structure of the actual code. Huge thanks in advance to people reading this and answering!
|
|
|
|
|
Dave
|
|
|
Group: Administrators
Posts: 12K,
Visits: 98K
|
+xHello Millisecond Forum, First time poster and not very experienced with Inquisit, I am wanting to adapt the Winsconsin Card Sorting Test. I'm willing to do an experiment with Inquisit 6 (Lab version). My experiment would have one WCST as downloadable here: https://www.millisecond.com/download/library/cardsortI have then coded 4 other versions in which the only difference is the stimuli change (i.e file directory, I slightly modified the .gif files). For now, the best solution that I have found to have participants do the five tasks is that I've created my stimuli, named them exacly as they are in the original version, and run each version from separate folders (which gives me total 5 data sheet per participant and takes some time + manual randomization of order). The tasks work well but it's not very convenient. The most convenient solution for me would be to have the following parameters within one script: - all five versions - order of versions randomized from participant to participant - within each version: random order of rule (number, color, shape). Would it be possible to achieve this or is it too complicated for a beginner ? Since one of the output files implies score calculations based on trial order, I'm afraid that following my idea might break the structure of the actual code. Huge thanks in advance to people reading this and answering! Certainly possible to achieve, but also certainly too complicated for a beginner. Having separate scripts is easier and more convenient. You also run less of a risk of breaking anything inadvertently. Combining separate data sheets is trivial to do in any spreadsheet or statistical analysis package. Finally, I'm not sure why separate scripts require manual randomization of order. Script order can be trivially randomized (or counterbalanced) via a <batch> script. https://www.millisecond.com/support/docs/current/html/language/attributes/selectionmode.htm// run three scripts A to C in random order
<batch example>
/ file = "script_a.iqx"
/ file = "script_b.iqx"
/ file = "script_c.iqx"
/ selectionmode = random
</batch>
|
|
|
|
|
aflex23
|
|
|
Group: Forum Members
Posts: 4,
Visits: 17
|
Thank you very much for your response, @Dave! I had no idea that script randomization with batch scripts was possible. I tried it, and it worked perfectly. This feature will save me a lot of time!
Since modifying the code to integrate randomization of the categorization rule order seems too complicated, I was wondering if there's a way to manually modify the order of the categorization rules (number, color, shape) in the scripts. For example, I would like to have script A (color, number, shape); script B (number, shape, color); and script C (shape, color, number) to control for the categorization rule order in my experiment.
However, I'm concerned that manually modifying the order may affect the calculation process for the summary .csv file within each script. Currently, I'm uncertain whether it's feasible to do so, and if it is, I'm unsure how to modify the code accordingly.
|
|
|
|
|
aflex23
|
|
|
Group: Forum Members
Posts: 4,
Visits: 17
|
+xThank you very much for your response, @Dave! I had no idea that script randomization with batch scripts was possible. I tried it, and it worked perfectly. This feature will save me a lot of time! Since modifying the code to integrate randomization of the categorization rule order seems too complicated, I was wondering if there's a way to manually modify the order of the categorization rules (number, color, shape) in the scripts. For example, I would like to have script A (color, number, shape); script B (number, shape, color); and script C (shape, color, number) to control for the categorization rule order in my experiment. However, I'm concerned that manually modifying the order may affect the calculation process for the summary .csv file within each script. Currently, I'm uncertain whether it's feasible to do so, and if it is, I'm unsure how to modify the code accordingly. edit: @Dave (the citation seems to be broken due to my exclamation mark)
|
|
|
|
|
Dave
|
|
|
Group: Administrators
Posts: 12K,
Visits: 98K
|
It's possible and if you spend some time with the original script, it should be fairly obvious how to do it.
What, exactly, is unclear to you?
|
|
|
|
|
Dave
|
|
|
Group: Administrators
Posts: 12K,
Visits: 98K
|
+xIt's possible and if you spend some time with the original script, it should be fairly obvious how to do it. What, exactly, is unclear to you? The attached files should give you the general idea: You switch out what each category is. In the original (WCST "A"), category 1 = color, category 2 = form, category 2 = number. To generate version "B", then switch categories 1 and 3: category 1 = number; category 2 = form (unchanged), category 3 = color. There's HTML in the attached ZIP that shows the exact differences between the two versions side-by-side. Proceed accordingly to generate version "C", where you'll want to switch categories 1 and 2, such that category 1 = form, category 2 = color, category 3 = number (unchanged). I believe attached version "B" to be correct, however, I have not tested it extensively. You should double-check the changes and test things properly before putting it to practical use.
|
|
|
|
|
aflex23
|
|
|
Group: Forum Members
Posts: 4,
Visits: 17
|
+x+xIt's possible and if you spend some time with the original script, it should be fairly obvious how to do it. What, exactly, is unclear to you? The attached files should give you the general idea: You switch out what each category is. In the original (WCST "A"), category 1 = color, category 2 = form, category 2 = number. To generate version "B", then switch categories 1 and 3: category 1 = number; category 2 = form (unchanged), category 3 = color. There's HTML in the attached ZIP that shows the exact differences between the two versions side-by-side. Proceed accordingly to generate version "C", where you'll want to switch categories 1 and 2, such that category 1 = form, category 2 = color, category 3 = number (unchanged). I believe attached version "B" to be correct, however, I have not tested it extensively. You should double-check the changes and test things properly before putting it to practical use. Indeed, what was unclear to me was the location of the chunks that should be modified but with your HTML I see it more clearly now. Thank you very much for your response! It is also very educational for me. I will run it and verify the outcomes in the files, then confirm that everything is functional by posting here for future users that may have the same question!
|
|
|
|