+x

 

aliciay1 - 6/3/2020

Hi, 

I am wondering if it is possible to randomise the selection from two linked lists when presenting in trials with set conditions on the randomisation.

For example I am trying to randomise selection from the trial type list and digit sequence list each time it is run. However, there must be an 80/20 ratio of trial types presented and no two same digits can be presented in a row. I have tried the below code, however I am unsure if the reset function is correct or in the right place, secondly the creation of a new digit list is not that effective. How could this be resolved? 

Any suggestions on how to improve this would be greatly appreciated.

Thank you. 



Note: list of practice trialtypes
<list practice_trialtype>
/items = (trial.practice_go, trial.practice_go, trial.practice_go, trial.practice_go, trial.practice_go, trial.practice_go, trial.practice_go, trial.practice_go,
            trial.practice_nogo, trial.practice_nogo)
/replace = true
/poolsize = 24
</list>

Note: list of digits used for practice
<list practice_digitsequence>
/items = (1,2,3,4,5,6,7,8,9)
/poolsize = 24
</list>


<trial practice_go>
/pretrialpause = values.iti
/ontrialbegin = [
    values.trialtype = "Go";
    values.digit = list.practice_digitsequence.nextvalue;
    values.digitLastChosen = list.practice_digitsequence.currentvalue;
    values.digit_new = list.practice_digitsequence.removeitem(values.digitLastChosen);
    values.digit_new = list.practice_digitsequence.nextvalue;
    list.practice_digitsequence.resetselection(values.digit);
    values.fontsize = list.practice_fontsizes.nextvalue;
    trial.practice_go.insertstimulustime(text.mask, values.digitpresentationtime);
    trial.practice_go.insertstimulustime(shape.background, (values.digitpresentationtime + values.maskpresentationtime));
]
/ ontrialend = [
        trial.practice_go.resetstimulusframes();
]
/stimulustimes = [0 = background, digit]
/responsetime = 0
/responseinterrupt = frames
/monkeyresponse = (57)
/iscorrectresponse = [trial.practice_go.response == 0]
/trialduration = values.digitpresentationtime + values.maskpresentationtime

/ontrialend = [
    if (trial.practice_go.response == 0)
        values.RT = trial.practice_go.latency
    else {
        values.RT = "";
        values.latencytype = 0;
    };
    
    if (values.RT != "" && values.RT < values.anticipatoryresponsetime) {
        values.latencytype = 1;
    };

    if (values.RT != "" && values.RT >= values.anticipatoryresponsetime && values.RT < values.validresponsetime) {
        values.latencytype = 2;
    };
    
    if (values.RT != "" && values.RT >= values.validresponsetime) {
        values.latencytype = 3;
    };
    
    if (values.RT == "") {
        values.responsetype = "Omission";
    };

    if (values.RT != "" && values.RT < values.anticipatoryresponsetime) {
        values.responsetype = "Go Anticipatory";
    };
    
    if (values.RT >= values.anticipatoryresponsetime && values.RT < values.validresponsetime)
        values.responsetype = "Go Ambiguous";
        
    if (values.RT >= values.validresponsetime) {
        values.responsetype = "Go Success";
    };
]
/recorddata = true
/branch = [if (trial.practice_go.error) trial.feedbackerror]
/branch = [if (trial.practice_go.correct) trial.feedbackcorrect]
</trial>


<trial practice_nogo>
/pretrialpause = values.iti
/ontrialbegin = [
    values.trialtype = "NoGo";
   values.digitnogo = list.practice_digitsequence.nextvalue;
    values.digitLastChosen = list.practice_digitsequence.currentvalue;
    values.digitnogo_new = list.practice_digitsequence.removeitem(values.digitLastChosen)
    values.digitnogo_new = list.practice_digitsequence.nextvalue;  
    list.practice_digitsequence.resetselection(values.digitnogo);
  values.fontsize = list.practice_fontsizes.nextvalue;
    if (mod(values.digitnogo, 2) == 0) {
        values.even = true;
        values.correctresponsekey = values.responsekey2;
    }
    else {
        values.even = false;
        values.correctresponsekey = values.responsekey1;
    }
    trial.practice_nogo.insertstimulustime(text.mask, values.digitpresentationtime);
    trial.practice_nogo.insertstimulustime(shape.background, (values.digitpresentationtime + values.maskpresentationtime));
]
/ontrialend = [
    trial.practice_nogo.resetstimulusframes();
]

/stimulustimes = [0 = background, digitnogo]
/responsetime = 0
/responseinterrupt = frames
/monkeyresponse = (57,0)
/iscorrectresponse = [trial.practice_nogo.response == values.correctresponsekey]
/trialduration = values.digitpresentationtime + values.maskpresentationtime

/ontrialend = [

    if (trial.practice_nogo.response == values.correctresponsekey)
        values.RT = trial.practice_nogo.latency
    else {
        values.RT = "";
        values.latencytype = 0;
    };
    
    if (values.RT != "" && values.RT < values.anticipatoryresponsetime) {
        values.latencytype = 1;
    };
    
    if (values.RT != "" && values.RT >= values.anticipatoryresponsetime && values.RT < values.validresponsetime) {
        values.latencytype = 2;
    };
    
    if (values.RT != "" && values.RT >= values.validresponsetime) {
        values.latencytype = 3;
    };
    
    if (values.RT == "") {
        values.responsetype = "NoGo Success";
    } else {
        values.responsetype = "NoGo Failure";
    };
]

/recorddata = true
/branch = [if (trial.practice_nogo.error) trial.feedbackerror]
/branch = [if (trial.practice_nogo.correct) trial.feedbackcorrect]
</trial>

+x

 

Dave - 6/3/2020

+x

 

aliciay1 - 6/3/2020

Hi, 

I am wondering if it is possible to randomise the selection from two linked lists when presenting in trials with set conditions on the randomisation.

For example I am trying to randomise selection from the trial type list and digit sequence list each time it is run. However, there must be an 80/20 ratio of trial types presented and no two same digits can be presented in a row. I have tried the below code, however I am unsure if the reset function is correct or in the right place, secondly the creation of a new digit list is not that effective. How could this be resolved? 

Any suggestions on how to improve this would be greatly appreciated.

Thank you. 



Note: list of practice trialtypes
<list practice_trialtype>
/items = (trial.practice_go, trial.practice_go, trial.practice_go, trial.practice_go, trial.practice_go, trial.practice_go, trial.practice_go, trial.practice_go,
            trial.practice_nogo, trial.practice_nogo)
/replace = true
/poolsize = 24
</list>

Note: list of digits used for practice
<list practice_digitsequence>
/items = (1,2,3,4,5,6,7,8,9)
/poolsize = 24
</list>


<trial practice_go>
/pretrialpause = values.iti
/ontrialbegin = [
    values.trialtype = "Go";
    values.digit = list.practice_digitsequence.nextvalue;
    values.digitLastChosen = list.practice_digitsequence.currentvalue;
    values.digit_new = list.practice_digitsequence.removeitem(values.digitLastChosen);
    values.digit_new = list.practice_digitsequence.nextvalue;
    list.practice_digitsequence.resetselection(values.digit);
    values.fontsize = list.practice_fontsizes.nextvalue;
    trial.practice_go.insertstimulustime(text.mask, values.digitpresentationtime);
    trial.practice_go.insertstimulustime(shape.background, (values.digitpresentationtime + values.maskpresentationtime));
]
/ ontrialend = [
        trial.practice_go.resetstimulusframes();
]
/stimulustimes = [0 = background, digit]
/responsetime = 0
/responseinterrupt = frames
/monkeyresponse = (57)
/iscorrectresponse = [trial.practice_go.response == 0]
/trialduration = values.digitpresentationtime + values.maskpresentationtime

/ontrialend = [
    if (trial.practice_go.response == 0)
        values.RT = trial.practice_go.latency
    else {
        values.RT = "";
        values.latencytype = 0;
    };
    
    if (values.RT != "" && values.RT < values.anticipatoryresponsetime) {
        values.latencytype = 1;
    };

    if (values.RT != "" && values.RT >= values.anticipatoryresponsetime && values.RT < values.validresponsetime) {
        values.latencytype = 2;
    };
    
    if (values.RT != "" && values.RT >= values.validresponsetime) {
        values.latencytype = 3;
    };
    
    if (values.RT == "") {
        values.responsetype = "Omission";
    };

    if (values.RT != "" && values.RT < values.anticipatoryresponsetime) {
        values.responsetype = "Go Anticipatory";
    };
    
    if (values.RT >= values.anticipatoryresponsetime && values.RT < values.validresponsetime)
        values.responsetype = "Go Ambiguous";
        
    if (values.RT >= values.validresponsetime) {
        values.responsetype = "Go Success";
    };
]
/recorddata = true
/branch = [if (trial.practice_go.error) trial.feedbackerror]
/branch = [if (trial.practice_go.correct) trial.feedbackcorrect]
</trial>


<trial practice_nogo>
/pretrialpause = values.iti
/ontrialbegin = [
    values.trialtype = "NoGo";
   values.digitnogo = list.practice_digitsequence.nextvalue;
    values.digitLastChosen = list.practice_digitsequence.currentvalue;
    values.digitnogo_new = list.practice_digitsequence.removeitem(values.digitLastChosen)
    values.digitnogo_new = list.practice_digitsequence.nextvalue;  
    list.practice_digitsequence.resetselection(values.digitnogo);
  values.fontsize = list.practice_fontsizes.nextvalue;
    if (mod(values.digitnogo, 2) == 0) {
        values.even = true;
        values.correctresponsekey = values.responsekey2;
    }
    else {
        values.even = false;
        values.correctresponsekey = values.responsekey1;
    }
    trial.practice_nogo.insertstimulustime(text.mask, values.digitpresentationtime);
    trial.practice_nogo.insertstimulustime(shape.background, (values.digitpresentationtime + values.maskpresentationtime));
]
/ontrialend = [
    trial.practice_nogo.resetstimulusframes();
]

/stimulustimes = [0 = background, digitnogo]
/responsetime = 0
/responseinterrupt = frames
/monkeyresponse = (57,0)
/iscorrectresponse = [trial.practice_nogo.response == values.correctresponsekey]
/trialduration = values.digitpresentationtime + values.maskpresentationtime

/ontrialend = [

    if (trial.practice_nogo.response == values.correctresponsekey)
        values.RT = trial.practice_nogo.latency
    else {
        values.RT = "";
        values.latencytype = 0;
    };
    
    if (values.RT != "" && values.RT < values.anticipatoryresponsetime) {
        values.latencytype = 1;
    };
    
    if (values.RT != "" && values.RT >= values.anticipatoryresponsetime && values.RT < values.validresponsetime) {
        values.latencytype = 2;
    };
    
    if (values.RT != "" && values.RT >= values.validresponsetime) {
        values.latencytype = 3;
    };
    
    if (values.RT == "") {
        values.responsetype = "NoGo Success";
    } else {
        values.responsetype = "NoGo Failure";
    };
]

/recorddata = true
/branch = [if (trial.practice_nogo.error) trial.feedbackerror]
/branch = [if (trial.practice_nogo.correct) trial.feedbackcorrect]
</trial>

It's entirely unclear what the "go" and what the "no go" stimuli are here. That said, if all you care about is having no two identical digits in a row, /maxrunsize is your friend:

https://www.millisecond.com/support/docs/v5/html/language/attributes/maxrunsize.htm

+x

 

Dave - 6/3/2020

If color is the sole distinguisher, then /maxrunsize=1 in the digit <list> should do the trick.

+x

 

aliciay1 - 6/3/2020

+x

 

Dave - 6/3/2020

If color is the sole distinguisher, then /maxrunsize=1 in the digit <list> should do the trick.

OK great, thanks very much! That is very helpful. 

I have a secondary issue, as one form of the task is a 1-back task, to correctly score resposnses it is required that the randomly selected digit sequence is saved to determine whether the response correct. Is there a way to track and save the selection of digits using maxrunsize, or do you have any suggestions on how I could correctly score performance? 

I currently have a separate digit list with the correct sequence of answers to the 1-back digit sequence which is fixed but I am unsure how to apply this when digit sequence is randomised.