Millisecond Forums » Millisecond Forums » Inquisit 6 » Define correct answers in

Define correct answers in

L123 — Mon, 01 Aug 2022 13:09:11 GMT

Hey,

I have an experiment where I want to have users decide between 1, 2 or 3 (keyboard input). If they decide for a specific number (either 1 or 3), they should get a message of being correct or incorrect. If they decide for number 2, they should see another trial. This works fine so far, but now I want to specify the correct (and incorrect) numbers using a list. Unfortunately, it doesn't work and the correct number is not assigned in my data either.

Here is the code: (copied the relevant parts only):

<item item_correct_answers>
/1 = "1"
/2 = "3"
/3 = "1"
/4 = "3"
</item>

<list correct_answers>
/ items = [item.item_correct_answers]
/ select = values.currentscenario
// select = 2
</list>

<item item_incorrect_answers>
/1 = "3"
/2 = "1"
/3 = "3"
/4 = "1"
</item>

<list incorrect_answers>
/ items = [item.item_incorrect_answers]
/ select = values.currentscenario
</list>

<values>
/ currentscenario = 0
</values>

<trial task_70>
/ stimulusframes = [1 = fixation; 100 = override_fixation, instruction, stimulus_70, orangeChoice, blueChoice, automationChoice, press_1, press_2, press_3]
/ inputdevice = keyboard
/ validresponse = ("1", "2", "3")
/ correctresponse = (list.correct_answers.currentvalue)
/ responsemessage = (list.correct_answers.currentvalue, feedback_correct, 1500)
/ responsetrial = ("2", task_70_advice)
/ responsemessage = (list.incorrect_answers.currentvalue, feedback_incorrect, 1500)
/ ontrialbegin = [
values.currentscenario = values.currentscenario + 1;
]
</trial>

<block task>
/ trials = [1-4 = sequence(task_70)]
</block>

<expt>
/ blocks = [
1 = task;
]
</expt>

Thanks in advance!

RE: Define correct answers in

L123 — Mon, 01 Aug 2022 13:09:11 GMT

Works perfectly! Thank you so much!

RE: Define correct answers in

Dave — Mon, 01 Aug 2022 12:36:19 GMT

+x[quote]

[b]L123 - 8/1/2022[/b]
Thank you!
Below you can find a small demo that runs. What I want to do is iterate over a list with different answers. Participants have to choose between A or B. In the first trial, A is correct, in the second trial, B is correct (and so on - that is subject to change in the future). They can also request advice. Now the advice also iterates over a list of items. However, I cannot set the correct and incorrect responses (with corresponding response message) based on the current trial / list. The iteration of advice works for each trial, but setting the correct answers and response message and incorrect answers and response message for the current trial does not work.

// define correct answers (would be great to use this item element for easier customization)
<item item_correct_answers>
/ 1 = "1"
/ 2 = "3"
/ 3 = "1"
/ 4 = "3"
</item>

// define the correct answers, as of now in a list element
<list correct_answers>
/ items = (1, 3, 1, 3)
/ select = values.currentscenario
// select = 2
</list>

// define incorrect answers (would be great to use this item element for easier customization)
<item item_incorrect_answers>
/ 1 = "3"
/ 2 = "1"
/ 3 = "3"
/ 4 = "1"
</item>

// define the incorrect answers, as of now in a list element
<list incorrect_answers>
/ items = [item.item_incorrect_answers]
/ select = values.currentscenario
// select = 2
</list>

// define texts
<text fixation>
/items = ("+")
/ fontstyle = ("Arial", 5%, false, false, false, false, 5, 1)
/ size = (30%, 30%)
/ txbgcolor = white
/ txcolor = black
/ vjustify = center
/ hjustify = center
/ position = (50%, 50%)
</text>

<text feedback_correct>
/items = ("Correct!")
/ fontstyle = ("Arial", 5%, false, false, false, false, 5, 1)
/ size = (30%, 30%)
/ txbgcolor = white
/ txcolor = black
/ vjustify = center
/ hjustify = center
/ position = (50%, 50%)
</text>

<text feedback_incorrect>
/ items = ("Incorrect!")
/ fontstyle = ("Arial", 5%, false, false, false, false, 5, 1)
/ size = (30%, 30%)
/ txbgcolor = white
/ txcolor = black
/ vjustify = center
/ hjustify = center
/ position = (50%, 50%)
</text>

// Ask participants to press either 1 or 3 on their keyboard, if they want help, they should press 2
<text instruction>
/ items = ("A (Press 1) or B (Press 3)? If you want support, press 2")
/ fontstyle = ("Arial", 4%)
/ position = (50%, 60%)
/ txcolor = black
/ txbgcolor = white
</text>

// Override the fixation with a white screen
<shape override_fixation>
/ shape = rectangle
/ size = (30%, 30%)
/ color = (255,255,255)
/ vjustify = center
/ hjustify = center
/ position = (50%, 50%)
</shape>

// specify advice
<item advice_answer_text>
/1 = "Advice: 1"
/2 = "Advice: 3"
/3 = "Advice: 1"
/4 = "Advice: 3"
</item>

// provide advice
<text advice_answer>
/ items = advice_answer_text
/ select = values.currentscenario
/ fontstyle = ("Arial", 5%)
/ position = (50%, 75%)
/ size = (32%, 15%)
/ vjustify = center
/ hjustify = center
/ txcolor = white
/ txbgcolor = grey
</text>

// define data to be recorded
<data>
/ columns = (build, computer.platform, date, time, subject, group, session,
blockcode, blocknum, trialcode, trialnum, list.correct_answers.currentvalue, response, latency, correct)
</data>

// define values
<values>
/ currentscenario = 0
</values>

// In this practice trial, I can define the correct answers. 1 is correct, 1 gives the "correct" Feedback and 3 gives the "incorrect" feedback. However, the correct response might change in other trials (to be B, so 3, for example)
<trial practice>
/ stimulusframes = [1 = fixation; 100 = override_fixation, instruction]
/ inputdevice = keyboard
/ validresponse = ("1", "2", "3")
/ correctresponse = ("1")
/ responsemessage = ("1", feedback_correct, 0)
/ responsemessage = ("3", feedback_incorrect, 0)
/ posttrialpause = 1500
</trial>

// this code works, but it does not give the "feedback_correct" or "feedback_incorrect" response and also does not code the correctresponse in the data section
<trial task_70>
/ stimulusframes = [1 = fixation; 100 = override_fixation, instruction]
/ inputdevice = keyboard
/ validresponse = ("1", "2", "3")
/ correctresponse = (list.correct_answers.currentvalue)
/ responsemessage = (list.correct_answers.currentvalue, feedback_correct, 1500)
/ responsetrial = ("2", task_70_advice)
/ responsemessage = (list.incorrect_answers.currentvalue, feedback_incorrect, 1500)
/ ontrialbegin = [
values.currentscenario = values.currentscenario + 1;
]
</trial>

// if participants wanted to get advice (pressed 2 in the other trial), they can get advice here and choose again. However, it does not correctly code correct/incorrect either
<trial task_70_advice>
/ stimulusframes = [1 = override_fixation, instruction, advice_answer]
/ inputdevice = keyboard
/ validresponse = ("1", "3")
/ responsemessage = (list.correct_answers.currentvalue, feedback_correct, 1500)
/ responsemessage = (list.incorrect_answers.currentvalue, feedback_incorrect, 1500)
</trial>

// define blocks
<block practice_task>
/ trials = [1 = practice]
</block>

<block task>
/ trials = [1-4 = sequence(task_70)]
</block>

// define experiment
<expt>
/ blocks = [
1 = practice_task;
2 = task;
]
</expt>
[/quote]

I don't see any need to use lists here at all. The simplest way to get this working properly based on what you have is this:

[code]// define correct answers (would be great to use this item element for easier customization)
<item item_correct_answers>
/ 1 = "1"
/ 2 = "3"
/ 3 = "1"
/ 4 = "3"
</item>

// define incorrect answers (would be great to use this item element for easier customization)
<item item_incorrect_answers>
/ 1 = "3"
/ 2 = "1"
/ 3 = "3"
/ 4 = "1"
</item>

// define texts
<text fixation>
/items = ("+")
/ fontstyle = ("Arial", 5%, false, false, false, false, 5, 1)
/ size = (30%, 30%)
/ txbgcolor = white
/ txcolor = black
/ vjustify = center
/ hjustify = center
/ position = (50%, 50%)
</text>

<text feedback_correct>
/items = ("Correct!")
/ fontstyle = ("Arial", 5%, false, false, false, false, 5, 1)
/ size = (30%, 30%)
/ txbgcolor = white
/ txcolor = black
/ vjustify = center
/ hjustify = center
/ position = (50%, 50%)
</text>

<text feedback_incorrect>
/ items = ("Incorrect!")
/ fontstyle = ("Arial", 5%, false, false, false, false, 5, 1)
/ size = (30%, 30%)
/ txbgcolor = white
/ txcolor = black
/ vjustify = center
/ hjustify = center
/ position = (50%, 50%)
</text>

// Ask participants to press either 1 or 3 on their keyboard, if they want help, they should press 2
<text instruction>
/ items = ("A (Press 1) or B (Press 3)? If you want support, press 2")
/ fontstyle = ("Arial", 4%)
/ position = (50%, 60%)
/ txcolor = black
/ txbgcolor = white
</text>

// Override the fixation with a white screen
<shape override_fixation>
/ shape = rectangle
/ size = (30%, 30%)
/ color = (255,255,255)
/ vjustify = center
/ hjustify = center
/ position = (50%, 50%)
</shape>

// specify advice
<item advice_answer_text>
/1 = "Advice: 1"
/2 = "Advice: 3"
/3 = "Advice: 1"
/4 = "Advice: 3"
</item>

// provide advice
<text advice_answer>
/ items = advice_answer_text
/ select = values.currentscenario
/ fontstyle = ("Arial", 5%)
/ position = (50%, 75%)
/ size = (32%, 15%)
/ vjustify = center
/ hjustify = center
/ txcolor = white
/ txbgcolor = grey
</text>

// define data to be recorded
<data>
/ columns = (build, computer.platform, date, time, subject, group, session,
blockcode, blocknum, trialcode, trialnum, values.incorrectanswer, values.correctanswer, responsetext, latency, correct)
</data>

// define values
<values>
/ currentscenario = 0
/ correctanswer = " "
/ incorrectanswer = " "
</values>

// In this practice trial, I can define the correct answers. 1 is correct, 1 gives the "correct" Feedback and 3 gives the "incorrect" feedback. However, the correct response might change in other trials (to be B, so 3, for example)
<trial practice>
/ stimulusframes = [1 = fixation; 100 = override_fixation, instruction]
/ inputdevice = keyboard
/ validresponse = ("1", "2", "3")
/ correctresponse = ("1")
/ responsemessage = ("1", feedback_correct, 0)
/ responsemessage = ("3", feedback_incorrect, 0)
/ posttrialpause = 1500
</trial>

// this code works, but it does not give the "feedback_correct" or "feedback_incorrect" response and also does not code the correctresponse in the data section
<trial task_70>
/ stimulusframes = [1 = fixation; 100 = override_fixation, instruction]
/ inputdevice = keyboard
/ validresponse = ("1", "2", "3")
/ correctresponse = (values.correctanswer) // correct answer is pressing the correct key as stored in values.correctanswer
/ responsemessage = (values.correctanswer, feedback_correct, 1500) // give correct feedback
/ responsetrial = ("2", task_70_advice)
/ responsemessage = (values.incorrectanswer, feedback_incorrect, 1500) // give error feedback
/ ontrialbegin = [
values.currentscenario = values.currentscenario + 1;
// store correct key for the current scenario in a variable
values.correctanswer = item.item_correct_answers.item(values.currentscenario);
// store incorrect key for the current scenario in a variable
values.incorrectanswer = item.item_incorrect_answers.item(values.currentscenario);
]
</trial>

// if participants wanted to get advice (pressed 2 in the other trial), they can get advice here and choose again. However, it does not correctly code correct/incorrect either
<trial task_70_advice>
/ stimulusframes = [1 = override_fixation, instruction, advice_answer]
/ inputdevice = keyboard
/ validresponse = ("1", "3")
/ responsemessage = (values.correctanswer, feedback_correct, 1500) // give correct feedback
/ responsemessage = (values.incorrectanswer, feedback_incorrect, 1500) // give error feedback
</trial>

// define blocks
<block practice_task>
/ trials = [1 = practice]
</block>

<block task>
/ trials = [1-4 = sequence(task_70)]
</block>

// define experiment
<expt>
/ blocks = [
1 = practice_task;
2 = task;
]
</expt>[/code]

RE: Define correct answers in

L123 — Mon, 01 Aug 2022 12:10:59 GMT

Thank you!
Below you can find a small demo that runs. What I want to do is iterate over a list with different answers. Participants have to choose between A or B. In the first trial, A is correct, in the second trial, B is correct (and so on - that is subject to change in the future). They can also request advice. Now the advice also iterates over a list of items. However, I cannot set the correct and incorrect responses (with corresponding response message) based on the current trial / list. The iteration of advice works for each trial, but setting the correct answers and response message and incorrect answers and response message for the current trial does not work.

// define correct answers (would be great to use this item element for easier customization)
<item item_correct_answers>
/ 1 = "1"
/ 2 = "3"
/ 3 = "1"
/ 4 = "3"
</item>

// define the correct answers, as of now in a list element
<list correct_answers>
/ items = (1, 3, 1, 3)
/ select = values.currentscenario
// select = 2
</list>

// define incorrect answers (would be great to use this item element for easier customization)
<item item_incorrect_answers>
/ 1 = "3"
/ 2 = "1"
/ 3 = "3"
/ 4 = "1"
</item>

// define the incorrect answers, as of now in a list element
<list incorrect_answers>
/ items = [item.item_incorrect_answers]
/ select = values.currentscenario
// select = 2
</list>

// define texts
<text fixation>
/items = ("+")
/ fontstyle = ("Arial", 5%, false, false, false, false, 5, 1)
/ size = (30%, 30%)
/ txbgcolor = white
/ txcolor = black
/ vjustify = center
/ hjustify = center
/ position = (50%, 50%)
</text>

<text feedback_correct>
/items = ("Correct!")
/ fontstyle = ("Arial", 5%, false, false, false, false, 5, 1)
/ size = (30%, 30%)
/ txbgcolor = white
/ txcolor = black
/ vjustify = center
/ hjustify = center
/ position = (50%, 50%)
</text>

<text feedback_incorrect>
/ items = ("Incorrect!")
/ fontstyle = ("Arial", 5%, false, false, false, false, 5, 1)
/ size = (30%, 30%)
/ txbgcolor = white
/ txcolor = black
/ vjustify = center
/ hjustify = center
/ position = (50%, 50%)
</text>

// Ask participants to press either 1 or 3 on their keyboard, if they want help, they should press 2
<text instruction>
/ items = ("A (Press 1) or B (Press 3)? If you want support, press 2")
/ fontstyle = ("Arial", 4%)
/ position = (50%, 60%)
/ txcolor = black
/ txbgcolor = white
</text>

// Override the fixation with a white screen
<shape override_fixation>
/ shape = rectangle
/ size = (30%, 30%)
/ color = (255,255,255)
/ vjustify = center
/ hjustify = center
/ position = (50%, 50%)
</shape>

// specify advice
<item advice_answer_text>
/1 = "Advice: 1"
/2 = "Advice: 3"
/3 = "Advice: 1"
/4 = "Advice: 3"
</item>

// provide advice
<text advice_answer>
/ items = advice_answer_text
/ select = values.currentscenario
/ fontstyle = ("Arial", 5%)
/ position = (50%, 75%)
/ size = (32%, 15%)
/ vjustify = center
/ hjustify = center
/ txcolor = white
/ txbgcolor = grey
</text>

// define data to be recorded
<data>
/ columns = (build, computer.platform, date, time, subject, group, session,
blockcode, blocknum, trialcode, trialnum, list.correct_answers.currentvalue, response, latency, correct)
</data>

// define values
<values>
/ currentscenario = 0
</values>

// In this practice trial, I can define the correct answers. 1 is correct, 1 gives the "correct" Feedback and 3 gives the "incorrect" feedback. However, the correct response might change in other trials (to be B, so 3, for example)
<trial practice>
/ stimulusframes = [1 = fixation; 100 = override_fixation, instruction]
/ inputdevice = keyboard
/ validresponse = ("1", "2", "3")
/ correctresponse = ("1")
/ responsemessage = ("1", feedback_correct, 0)
/ responsemessage = ("3", feedback_incorrect, 0)
/ posttrialpause = 1500
</trial>

// this code works, but it does not give the "feedback_correct" or "feedback_incorrect" response and also does not code the correctresponse in the data section
<trial task_70>
/ stimulusframes = [1 = fixation; 100 = override_fixation, instruction]
/ inputdevice = keyboard
/ validresponse = ("1", "2", "3")
/ correctresponse = (list.correct_answers.currentvalue)
/ responsemessage = (list.correct_answers.currentvalue, feedback_correct, 1500)
/ responsetrial = ("2", task_70_advice)
/ responsemessage = (list.incorrect_answers.currentvalue, feedback_incorrect, 1500)
/ ontrialbegin = [
values.currentscenario = values.currentscenario + 1;
]
</trial>

// if participants wanted to get advice (pressed 2 in the other trial), they can get advice here and choose again. However, it does not correctly code correct/incorrect either
<trial task_70_advice>
/ stimulusframes = [1 = override_fixation, instruction, advice_answer]
/ inputdevice = keyboard
/ validresponse = ("1", "3")
/ responsemessage = (list.correct_answers.currentvalue, feedback_correct, 1500)
/ responsemessage = (list.incorrect_answers.currentvalue, feedback_incorrect, 1500)
</trial>

// define blocks
<block practice_task>
/ trials = [1 = practice]
</block>

<block task>
/ trials = [1-4 = sequence(task_70)]
</block>

// define experiment
<expt>
/ blocks = [
1 = practice_task;
2 = task;
]
</expt>

RE: Define correct answers in

Dave — Mon, 01 Aug 2022 11:53:20 GMT

+x[quote]

[b]L123 - 8/1/2022[/b]
Can I get it working with a textbox then?

I also tried adding the correct answers directly within the list element:

<list correct_answers>
/ items = ("1" "3" "1" "3")
/ select = values.currentscenario
</list>

It still does not work though.
[/quote]

There are multple other issues with your code. It's not even clear what exactly it's supposed to do, since you only posted some non-runnable excerpts. Do the following:

- Create a small, self-contained demo that mirrors the actual script's setup.
- Post that code here and explain clearly what you intend it to do.

Then I can show you how to do it right

RE: Define correct answers in

L123 — Mon, 01 Aug 2022 11:50:23 GMT

Can I get it working with a textbox then?

I also tried adding the correct answers directly within the list element:

<list correct_answers>
/ items = ("1" "3" "1" "3")
/ select = values.currentscenario
</list>

It still does not work though.

RE: Define correct answers in

Dave — Mon, 01 Aug 2022 11:33:26 GMT

+x[quote]

[b]L123 - 8/1/2022[/b]
Hey,

I have an experiment where I want to have users decide between 1, 2 or 3 (keyboard input). If they decide for a specific number (either 1 or 3), they should get a message of being correct or incorrect. If they decide for number 2, they should see another trial. This works fine so far, but now I want to specify the correct (and incorrect) numbers using a list. Unfortunately, it doesn't work and the correct number is not assigned in my data either.

Here is the code: (copied the relevant parts only):

<item item_correct_answers>
/1 = "1"
/2 = "3"
/3 = "1"
/4 = "3"
</item>

<list correct_answers>
/ items = [item.item_correct_answers]
/ select = values.currentscenario
// select = 2
</list>

<item item_incorrect_answers>
/1 = "3"
/2 = "1"
/3 = "3"
/4 = "1"
</item>

<list incorrect_answers>
/ items = [item.item_incorrect_answers]
/ select = values.currentscenario
</list>

<values>
/ currentscenario = 0
</values>

<trial task_70>
/ stimulusframes = [1 = fixation; 100 = override_fixation, instruction, stimulus_70, orangeChoice, blueChoice, automationChoice, press_1, press_2, press_3]
/ inputdevice = keyboard
/ validresponse = ("1", "2", "3")
/ correctresponse = (list.correct_answers.currentvalue)
/ responsemessage = (list.correct_answers.currentvalue, feedback_correct, 1500)
/ responsetrial = ("2", task_70_advice)
/ responsemessage = (list.incorrect_answers.currentvalue, feedback_incorrect, 1500)
/ ontrialbegin = [
values.currentscenario = values.currentscenario + 1;
]
</trial>

<block task>
/ trials = [1-4 = sequence(task_70)]
</block>

<expt>
/ blocks = [
1 = task;
]
</expt>

Thanks in advance!
[/quote]

<list correct_answers>
/ items = [item.item_correct_answers]
/ select = values.currentscenario
// select = 2
</list>

This is not how you define items in a list. Lists don't use <item> elements.