Millisecond Forums

I'm afraid your description is too unspecific to say anything useful or provide any example code.

- What kind(s) of problem(s) are participants supposed to solve and how? What is the response modality?
- You say there's a pool of 20 cues and participants can request a maximum of 6. You further state that "each cue [...] has a special location the screen." So, are there 6 or 20 locations? What are those locations and how do they relate to the respective cue?

Despite the various open questions and blank spots (cf. previous response), perhaps this rough sketch will be useful as a basis for you to work from:

<block myblock>
/ trials = [1=showproblem]
</block>

<values>
/ cuecount = 0
</values>

<trial showproblem>
/ inputdevice = mouse
/ stimulusframes = [1=problemstim, hintsreceived, requestcue, solveproblem]
/ validresponse = (requestcue, solveproblem)
/ branch = [if (trial.showproblem.response == "solveproblem") openended.entersolution]
/ branch = [if (trial.showproblem.response == "requestcue" && values.cuecount < 6) trial.givecue else openended.entersolution]
</trial>

<trial givecue>
/ ontrialbegin = [values.cuecount += 1; ]
/ stimulusframes = [1=cuestim]
/ validresponse = (0)
/ trialduration = 0
/ branch = [trial.showproblem]
</trial>

<openended entersolution>
/ stimulusframes = [1=clearscreen, problemstim, entersolutionstim]
/ validresponse = (anyresponse)
/ correctresponse = ("solution")
</openended>

<text problemstim>
/ items = ("[THE PROBLEM]")
/ erase = false
/ position = (50%, 10%)
</text>

<text cuestim>
/ items = cueitems
/ erase = false
/ hposition = list.cue_x.nextvalue
/ vposition = list.cue_y.nextvalue
</text>

<item cueitems>
/ 1 = "Cue A"
/ 2 = "Cue B"
/ 3 = "Cue C"
/ 4 = "Cue D"
/ 5 = "Cue E"
/ 6 = "Cue F"
/ 7 = "Cue G"
/ 8 = "Cue H"
/ 9 = "Cue I"
/ 10 = "Cue J"
/ 11 = "Cue K"
/ 12 = "Cue L"
/ 13 = "Cue M"
/ 14 = "Cue N"
/ 15 = "Cue O"
/ 16 = "Cue P"
/ 17 = "Cue Q"
/ 18 = "Cue R"
/ 19 = "Cue S"
/ 20 = "Cue T"
</item>

<text requestcue>
/ items = ("Give me a hint!")
/ txbgcolor = (grey)
/ erase = false
/ size = (20%, 10%)
/ vjustify = center
/ position = (35%, 85%)
</text>

<text solveproblem>
/ items = ("I know the solution!")
/ txbgcolor = (grey)
/ erase = false
/ size = (20%, 10%)
/ vjustify = center
/ position = (65%, 85%)
</text>

<text entersolutionstim>
/ items = ("Please enter the solution:")
/ erase = false
/ position = (50%, 40%)
</text>

<shape clearscreen>
/ color = (white)
/ shape = rectangle
/ size = (100%, 100%)
/ erase = false
</shape>

<list cue_x>
/ items = (25%, 50%, 75%, 25%, 50%, 75%)
/ selectionmode = sequence
</list>

<list cue_y>
/ items = (40%, 40%, 40%, 60%, 60%, 60%)
/ selectionmode = list.cue_x.currentindex
</list>

<text hintsreceived>
/ items = ("(Hints received: <%values.cuecount%>.)")
/ position = (50%, 15%)
/ erase = false
</text>

Thank you! I can definitely use your script as a basis for the script I am writing.

I just have one more question. Participants are supposed to solve 4 problems from a pool of 10 problems. At the moment, every time I request a cue, the problem changes. How could I keep the same problem until the participant solve it?

Thanks again,
Lucie

Set up a <list> containing the item numbers for the 10 problems (1 to 10). Start each "round" with a "start" trial that does nothing but select an item number from that <list> and stores it in a <values> entry. Use that value for selection in the "problem" stimulus element.

Here's an extension of the previous example to 2 problems. (I suppose there are different sets of cues pertaining to each problem, so that has to be worked into the script as well):

<block myblock>
/ trials = [1-2=starttrial]
</block>

<values>
/ problemnumber = 0
/ cuenumber = 0
/ cuecount = 0
</values>

<trial starttrial>
/ ontrialbegin = [values.problemnumber = list.problemitems.nextindex; values.cuecount = 0; text.cuestim.resetselection();
    list.cue_x.reset(); list.cue_y.reset();]
/ validresponse = (0)
/ branch = [trial.showproblem]
/ trialduration = 0
/ recorddata = false
</trial>

<trial showproblem>
/ inputdevice = mouse
/ stimulusframes = [1=problemstim, hintsreceived, requestcue, solveproblem]
/ validresponse = (requestcue, solveproblem)
/ branch = [if (trial.showproblem.response == "solveproblem") openended.entersolution]
/ branch = [if (trial.showproblem.response == "requestcue" && values.cuecount < 6) trial.givecue else openended.entersolution]
</trial>

<trial givecue>
/ ontrialbegin = [values.cuecount += 1; values.cuenumber = list.cueitems.nextvalue; ]
/ stimulusframes = [1=cuestim]
/ validresponse = (0)
/ trialduration = 0
/ branch = [trial.showproblem]
</trial>

<openended entersolution>
/ stimulusframes = [1=clearscreen, problemstim, entersolutionstim]
/ validresponse = (anyresponse)
/ correctresponse = ("solution")
</openended>

<text problemstim>
/ items = ("Problem A", "Problem B")
/ erase = false
/ position = (50%, 10%)
/ select = values.problemnumber
</text>

<text cuestim>
/ items = cueitems
/ erase = false
/ select = values.cuenumber
/ hposition = list.cue_x.nextvalue
/ vposition = list.cue_y.nextvalue
</text>

<item cueitems>
/ 1 = "PA Cue A"
/ 2 = "PA Cue B"
/ 3 = "PA Cue C"
/ 4 = "PA Cue D"
/ 5 = "PA Cue E"
/ 6 = "PA Cue F"
/ 7 = "PA Cue G"
/ 8 = "PA Cue H"
/ 9 = "PA Cue I"
/ 10 = "PA Cue J"
/ 11 = "PA Cue K"
/ 12 = "PA Cue L"
/ 13 = "PA Cue M"
/ 14 = "PA Cue N"
/ 15 = "PA Cue O"
/ 16 = "PA Cue P"
/ 17 = "PA Cue Q"
/ 18 = "PA Cue R"
/ 19 = "PA Cue S"
/ 20 = "PA Cue T"

/ 21 = "PB Cue A"
/ 22 = "PB Cue B"
/ 23 = "PB Cue C"
/ 24 = "PB Cue D"
/ 25 = "PB Cue E"
/ 26 = "PB Cue F"
/ 27 = "PB Cue G"
/ 28 = "PB Cue H"
/ 29 = "PB Cue I"
/ 30 = "PB Cue J"
/ 31 = "PB Cue K"
/ 32 = "PB Cue L"
/ 33 = "PB Cue M"
/ 34 = "PB Cue N"
/ 35 = "PB Cue O"
/ 36 = "PB Cue P"
/ 37 = "PB Cue Q"
/ 38 = "PB Cue R"
/ 39 = "PB Cue S"
/ 40 = "PB Cue T"
</item>

<text requestcue>
/ items = ("Give me a hint!")
/ txbgcolor = (grey)
/ erase = false
/ size = (20%, 10%)
/ vjustify = center
/ position = (35%, 85%)
</text>

<text solveproblem>
/ items = ("I know the solution!")
/ txbgcolor = (grey)
/ erase = false
/ size = (20%, 10%)
/ vjustify = center
/ position = (65%, 85%)
</text>

<text entersolutionstim>
/ items = ("Please enter the solution:")
/ position = (50%, 40%)
</text>

<shape clearscreen>
/ color = (white)
/ shape = rectangle
/ size = (100%, 100%)
/ erase = false
</shape>

<list cue_x>
/ items = (25%, 50%, 75%, 25%, 50%, 75%)
/ selectionmode = sequence
</list>

<list cue_y>
/ items = (40%, 40%, 40%, 60%, 60%, 60%)
/ selectionmode = list.cue_x.currentindex
</list>

<text hintsreceived>
/ items = ("(Hints received: <%values.cuecount%>.)")
/ position = (50%, 15%)
/ erase = false
</text>

<list problemitems>
/ poolsize = 2
</list>

//list of lists of cues pertaining to the different problems
<list cueitems>
/ items = (list.pa_cueitems.nextvalue, list.pb_cueitems.nextvalue)
/ selectionmode = values.problemnumber
</list>

//cue items pertaining to problem A are item numbers 1 to 20
<list pa_cueitems>
/ items = (1,2,3,4,5,6,7,8,9,10,11,12,13,14,15,16,17,18,19,20)
</list>

//cue items pertaining to problem B are item numbers 21 to 40
<list pb_cueitems>
/ items = (21,22,23,24,25,26,27,28,29,30,31,32,33,34,35,36,37,38,39,40)
</list>

Thanks to your numerous advices, I am nearly done programing this experiment! Nevertheless, I am still facing different issues I haven't been able to solve by myself yet.

#1 : I need to define a correct answer with each problem BUT I also need to associate a text element with a picture element and a value. 
When I do that I don't get the right answers anymore : 

<values>
/ problemnumber = 0
</values>

<list answers1>
/ items = ("no", "no","yes", "yes")
/ selectionmode =text.problemstim_frame.currentindex
</list>

<text problemstim_frame>

/ items =problemitem_frame

/ fontstyle =("Arial", 15pt, true)

/ txcolor = (255, 0,0)

/ position = (50, 30)

/ select =picture.problemstim_stimuli.currentitemnumber

/ erase = false

</text>

 

<picture problemstim_stimuli>

/ items =problemitem_stimuli

/ position = (50, 30)

/ select =values.problemnumber

/ erase = false

</picture>

 

If I do that instead, it works fine but it isn't associated with a value anymore : 

<picture Question_0click>
/ items = question_0click
/ position = (50, 30)
/ select = text.question_opposite_0click.currentindex
</picture>

<text question_opposite_0click>
/ items = question_opposite_0click
/ fontstyle = ("Arial", 15pt, true)
/ txcolor = (255, 0, 0)
/ position = (50, 30)
/ select = noreplace
</text>

What do you think?


#2 : The second problem I can't solve is about the timer. I only want to reset the timer when a <trial showproblem>  is starting (not when I ask for a cue).

<block myblock>
/ trials = [1=showproblem]
</block>

<values>
/ cuecount = 0
/ tasktimeout = 60000
/ tasktimeoutleft = 0
</values>

<trial showproblem>
/ ontrialbegin = [values.tasktimeoutleft = values.tasktimeout- trial.showproblem.elapsedtime]
/ inputdevice = mouse
/ stimulusframes = [1=problemstim, hintsreceived, requestcue, solveproblem, timer]
/ validresponse = (requestcue, solveproblem)
/ timeout = values.tasktimeoutleft
/ branch = [if (trial.showproblem.response == "solveproblem") openended.entersolution]
/ branch = [if (trial.showproblem.response == "requestcue" && values.cuecount < 6) trial.givecue else openended.entersolution]
</trial>

<trial givecue>
/ ontrialbegin = [values.cuecount += 1; ]
/ stimulusframes = [1=cuestim]
/ validresponse = (0)
/ trialduration = 0
/ branch = [trial.showproblem]
</trial>

<clock Timer>
/ mode = timer
/ resetrate = trial
/ erase = false
/ txcolor = (0, 255, 0)
/ txbgcolor = black
/ timeout = values.tasktimeout
/ position = (90%, 90%)
/ format = "mm:ss"
</clock>

#3 : My last problem concerns a likert scale I want to include. 
Every 5 problems, I want to know if the participant is confident in the final answer ("yes" , "no") he just provided, before giving him a feedback on his performance ("congratulations", "you loose"). 

<values>
/ instructionnumber = 0
/ problemnumber = 0
/ cuenumber = 0
/ cuecount = 0
/ points_available = 100
/ problem_number = 1
/ numberoftries = 0
/ score = 0
</values>

<trial starttrial>
/ ontrialbegin = [values.instructionnumber = list.instructionitems.nextindex ; values.problemnumber = list.problemitems.nextindex;
values.cuecount = 0; text.cuestim_frame.resetselection(); picture.cuestim_stimuli.resetselection() ; list.cue_x.reset(); list.cue_y.reset();
values.points_available = 100; values.numberoftries = 0; values.problem_number = list.problemitems.nextindex ]
/ validresponse = (0)
/ branch = [trial.showproblem]
/ trialduration = 0
/ recorddata = false
</trial>


<trial showproblem>
/ ontrialbegin = [values.numberoftries += 1]
/ ontrialbegin = [if (values.numberoftries > 1) values.points_available -= list.subtractpoints.nextindex; ]
/ ontrialend = [if (trial.showproblem.correct) values.score += values.points_available; ]
/ inputdevice = mouse
/ stimulusframes = [1= instructionframe, instruction1_stimuli, instruction2_stimuli,instruction1_frame, instruction2_frame, problemstim_stimuli,
problemstim_frame, Is, ?, requestcue, yes, no, points, hintsreceived]
/ validresponse = (requestcue, yes, no)
/ iscorrectresponse = [trial.showproblem.response == list.answers1.nextvalue]
/ branch = [if (trial.showproblem.correct) trial.congratulation]
/ branch = [if ( trial.showproblem.response == "yes" || trial.showproblem.response =="no" && values.problem_number = 5) likert.valence]
/ branch = [if (trial.showproblem.response == "requestcue" && values.cuecount < 6) trial.givecue else trial.error]
</trial>

<item confidencescale>
/1 = "How confident are you that your answer is correct?"
</item>

<text confidencescale>
/ items = confidencescale
/ position = (50, 10)
</text>

<likert valence>
/ stimulusframes = [1 = confidencescale]
/ anchors = [1 = "Not at all confident"; 4= "Moderately confident"; 7= "Very confident"]
/ numpoints = 7
/ anchorwidth=150px
/ inputdevice = mouse
/ branch = [trial.congratulations]
</likert>

<trial congratulations>
/ stimulusframes = [1=clearscreen, congratulations, next]
/ inputdevice = mouse
/ validresponse = ("next")
</trial>

<trial error>
/ stimulusframes = [1=clearscreen, error, next]
/ inputdevice = mouse
/ validresponse = ("next")
</trial>

Thanks again for your help!
Lucie

Re. #1: You simply use the <values> entry you populate at the start of every round for selection in all relevant elements.

<list answers1>
/ items = ("no", "no","yes", "yes")
/ selectionmode =values.problemnumber
</list>

<text problemstim_frame>

/ items =problemitem_frame

/ fontstyle =("Arial", 15pt, true)

/ txcolor = (255, 0,0)

/ position = (50, 30)

/ select =values.problemnumber

/ erase = false

</text>

Re. #2: I don't understand what you want to do with the timer. To simply reset it, you do

<trial showproblem>
/ ontrialbegin = [clock.timer.resettime(); ]
...
</trial>

Re. #3: This /branch does not appear to make sense:

/ branch = [if ( trial.showproblem.response == "yes" || trial.showproblem.response =="no" && values.problem_number = 5) likert.valence]

(1) values.problem_number just replicates what you already have with values.problemnumber:

<trial starttrial>
/ ontrialbegin = [values.instructionnumber = list.instructionitems.nextindex ; values.problemnumber = list.problemitems.nextindex;
values.cuecount = 0; text.cuestim_frame.resetselection(); picture.cuestim_stimuli.resetselection() ; list.cue_x.reset(); list.cue_y.reset();
values.points_available = 100; values.numberoftries = 0; values.problem_number = list.problemitems.nextindex ]
...
</trial>

It *does not* reflect a *count* of the problems administered so far; as discussed previously, it represents the *item number* of the particular problem selected for that "round" of the game.

If you want a count, you need to implement one, i.e., you need to do

<values>
/ problemcount = 0
...
</values>

<trial starttrial>
/ ontrialbegin = [...; values.problemcount += 1]
...
</trial>

Then you can /branch to your likert based on whether that count is a multiple of 5:

/ branch = [if ( trial.showproblem.response != "requestcue" && mod(values.problemcount,5)==0 ) likert.valence]

I've been looking for a solution for ages and when I finally see it, it always looks so simple, thanks!

#2 : On the contrary, I want the timer to keep running until the participant gives his answer to the problem.
At the moment, the timer resets every time I ask for a cue  (<trial.givecue>).

If you want the timer to be displayed & ticking down across all trials for a given problem, you should display it via the <block>'s /bgstim like so:

<expt>
/ blocks = [1-2 = problemblock]
</expt>

<block problemblock>
/ bgstim = (timer)
/ trials = [1=starttrial]
</block>

<clock timer>
/ mode = timer
/ timeout = values.timeout
/ erase = false
/ position = (80%, 5%)
</clock>

<values>
/ timeout = 20000
/ problemnumber = 0
/ cuenumber = 0
/ cuecount = 0
</values>

<trial starttrial>
/ ontrialbegin = [values.problemnumber = list.problemitems.nextindex; values.cuecount = 0; text.cuestim.resetselection();
    list.cue_x.reset(); list.cue_y.reset();]
/ validresponse = (0)
/ branch = [trial.showproblem]
/ trialduration = 0
/ recorddata = false
</trial>

<trial showproblem>
/ inputdevice = mouse
/ stimulusframes = [1=problemstim, hintsreceived, requestcue, solveproblem]
/ validresponse = (requestcue, solveproblem)
/ branch = [if (trial.showproblem.response == "solveproblem") openended.entersolution]
/ branch = [if (trial.showproblem.response == "requestcue" && values.cuecount < 6) trial.givecue else openended.entersolution]
</trial>

<trial givecue>
/ ontrialbegin = [values.cuecount += 1; values.cuenumber = list.cueitems.nextvalue; ]
/ stimulusframes = [1=cuestim]
/ validresponse = (0)
/ trialduration = 0
/ branch = [trial.showproblem]
</trial>

<openended entersolution>
/ stimulusframes = [1=clearscreen, problemstim, entersolutionstim]
/ validresponse = (anyresponse)
/ correctresponse = ("solution")
</openended>

<text problemstim>
/ items = ("Problem A", "Problem B")
/ erase = false
/ position = (50%, 10%)
/ select = values.problemnumber
</text>

<text cuestim>
/ items = cueitems
/ erase = false
/ select = values.cuenumber
/ hposition = list.cue_x.nextvalue
/ vposition = list.cue_y.nextvalue
</text>

<item cueitems>
/ 1 = "PA Cue A"
/ 2 = "PA Cue B"
/ 3 = "PA Cue C"
/ 4 = "PA Cue D"
/ 5 = "PA Cue E"
/ 6 = "PA Cue F"
/ 7 = "PA Cue G"
/ 8 = "PA Cue H"
/ 9 = "PA Cue I"
/ 10 = "PA Cue J"
/ 11 = "PA Cue K"
/ 12 = "PA Cue L"
/ 13 = "PA Cue M"
/ 14 = "PA Cue N"
/ 15 = "PA Cue O"
/ 16 = "PA Cue P"
/ 17 = "PA Cue Q"
/ 18 = "PA Cue R"
/ 19 = "PA Cue S"
/ 20 = "PA Cue T"

/ 21 = "PB Cue A"
/ 22 = "PB Cue B"
/ 23 = "PB Cue C"
/ 24 = "PB Cue D"
/ 25 = "PB Cue E"
/ 26 = "PB Cue F"
/ 27 = "PB Cue G"
/ 28 = "PB Cue H"
/ 29 = "PB Cue I"
/ 30 = "PB Cue J"
/ 31 = "PB Cue K"
/ 32 = "PB Cue L"
/ 33 = "PB Cue M"
/ 34 = "PB Cue N"
/ 35 = "PB Cue O"
/ 36 = "PB Cue P"
/ 37 = "PB Cue Q"
/ 38 = "PB Cue R"
/ 39 = "PB Cue S"
/ 40 = "PB Cue T"
</item>

<text requestcue>
/ items = ("Give me a hint!")
/ txbgcolor = (grey)
/ erase = false
/ size = (20%, 10%)
/ vjustify = center
/ position = (35%, 85%)
</text>

<text solveproblem>
/ items = ("I know the solution!")
/ txbgcolor = (grey)
/ erase = false
/ size = (20%, 10%)
/ vjustify = center
/ position = (65%, 85%)
</text>

<text entersolutionstim>
/ items = ("Please enter the solution:")
/ position = (50%, 40%)
</text>

<shape clearscreen>
/ color = (white)
/ shape = rectangle
/ size = (100%, 80%)
/ erase = false
</shape>

<list cue_x>
/ items = (25%, 50%, 75%, 25%, 50%, 75%)
/ selectionmode = sequence
</list>

<list cue_y>
/ items = (40%, 40%, 40%, 60%, 60%, 60%)
/ selectionmode = list.cue_x.currentindex
</list>

<text hintsreceived>
/ items = ("(Hints received: <%values.cuecount%>.)")
/ position = (50%, 15%)
/ erase = false
</text>

<list problemitems>
/ poolsize = 2
</list>

//list of lists of cues pertaining to the different problems
<list cueitems>
/ items = (list.pa_cueitems.nextvalue, list.pb_cueitems.nextvalue)
/ selectionmode = values.problemnumber
</list>

//cue items pertaining to problem A are item numbers 1 to 20
<list pa_cueitems>
/ items = (1,2,3,4,5,6,7,8,9,10,11,12,13,14,15,16,17,18,19,20)
</list>

//cue items pertaining to problem B are item numbers 21 to 40
<list pb_cueitems>
/ items = (21,22,23,24,25,26,27,28,29,30,31,32,33,34,35,36,37,38,39,40)
</list>

It is working great! Thanks a lot!