Thread Rating:
  • 0 Vote(s) - 0 Average
  • 1
  • 2
  • 3
  • 4
  • 5
catch trials
#1
Good morning!
I have a question about catch trials and if it is possible for me to implement them in my study given my study design. I have 224 experimental trials total (seven blocks of 32 trials), and I am hoping to put six short breaks after the first six blocks of trials. My trials are randomized, so if I put the blocks into my block sequence in my procedure, the breaks get randomized with the trials (which I don't want). 


From the grammar reference, it seems that catch trials would be the best way to implement these consistent breaks into randomized trials. However, I'm not sure if I will be able to use catch trials with my study design. I was wondering if you could help me determine that? 
In my study, each trials has a sequence of six stimuli: a response cue, a blank grid which acts as a pause, a "cued response", another blank grid, a "discrimination response" and finally, one more blank grid. 

The response cue, which is presented first, auto-advances after 5ms. Then the blank grid is presented, which also auto-advances after 5ms. Then the "cued response" stimulus is presented, to which participants make a keypress response. This stimulus will not advance until participants make a keypress. Participants then see another blank grid which auto-advances after 5ms to the final stimulus, the "discrimination response" stimulus,  to which participants must make another keypress before the trial ends. 

The trial templates are set up so that there's a template for the response cue + first pause (called e.g "Trial8 RC"), a template for the cued response stimulus + discrimination response stimulus (called e.g "Trial8Critical"), and a template for the two pauses that follow the cued response and then the discrimination response stimuli, respectively (called e.g "Trial8Pauses"). 

"Trial8 RC": {
    "type": "basic",
    "stimuli": ["ResponseCueLeft", "BlankGrid5"],
    "auto_advance": true
  },
 
  "Trial8Critical": {
    "type": "basic",
    "stimuli": ["GreenBottomCRS", "BlueTopCRS"],
    "responses": ["TrainingResponseTargetF", "TrainingResponseTargetJ"]
  },
 
  "Trial8Pauses": {
    "type": "basic",
    "stimuli": ["BlankGrid5", "BlankGrid2"],
    "auto_advance": true

The response cue stimulus has a stimulus level duration of 0.5 seconds and the barrier set to "true" and the trial-template is set to auto-advance. 
In the procedures, this is set up so that the block will look as follows: 

"TrialBlock5": {
      "trial_templates": ["Trial5Critical", "Trial5Pauses"],
      "pattern": {"order": "alternate", "repeat": 1},
      "cover_trials": ["Trial5 RC"].

The reason I don't think I can incorporate catch trials into my study is because I already have the "Trial[x] RC" as a cover trial. If I put the "Trial[x] RC" templates into the "trial_templates": [...], along with the Trial[x]Critical and Trial[x]Pause, then I won't be able to alternate the pauses between the two critical stimuli, because FF won't be able to evenly distribute the stimuli. I've tried using the catch trials with the cover trial (like how you can have a cover and end trial) but the stimuli were presented out of order.

Do you have any thoughts on this?

Thank you so much!

Grace

Hi again,

It seems that the font in my first paragraph in my above post is unreadably small. I will copy and paste it here:

"I have a question about catch trials. I have 224 experimental trials total (seven blocks of 32 trials), and I am hoping to put six short breaks after the first six blocks of trials. My trials are randomized, so if I put the blocks into my block sequence in my procedure, the breaks get randomized with the trials (which I don't want)."

Sorry about that!
Reply
#2
Hi Grace,

Thank you for reaching out! Based on my understanding of the issue, there are two potential solutions:

1. If you did not mean "randomized blocks", it may be easiest to create the break as its own trial template and add it to the block sequence in the Procedure section. (for example: "block_sequence": ["Block1","BreakBlock","Block2","BreakBlock", etc.])

2. If you did mean "randomized blocks", you should be able to add "end_trials": ["BreakTrial"] to each individual block. 

Hope this helps!
Victoria

I have a question about catch trials and if it is possible for me to implement them in my study given my study design. I have 224 experimental trials total (seven blocks of 32 trials), and I am hoping to put six short breaks after the first six blocks of trials. My trials are randomized, so if I put the blocks into my block sequence in my procedure, the breaks get randomized with the trials (which I don't want). 


From the grammar reference, it seems that catch trials would be the best way to implement these consistent breaks into randomized trials. However, I'm not sure if I will be able to use catch trials with my study design. I was wondering if you could help me determine that? 
In my study, each trials has a sequence of six stimuli: a response cue, a blank grid which acts as a pause, a "cued response", another blank grid, a "discrimination response" and finally, one more blank grid. 

The response cue, which is presented first, auto-advances after 5ms. Then the blank grid is presented, which also auto-advances after 5ms. Then the "cued response" stimulus is presented, to which participants make a keypress response. This stimulus will not advance until participants make a keypress. Participants then see another blank grid which auto-advances after 5ms to the final stimulus, the "discrimination response" stimulus,  to which participants must make another keypress before the trial ends. 

The trial templates are set up so that there's a template for the response cue + first pause (called e.g "Trial8 RC"), a template for the cued response stimulus + discrimination response stimulus (called e.g "Trial8Critical"), and a template for the two pauses that follow the cued response and then the discrimination response stimuli, respectively (called e.g "Trial8Pauses"). 

"Trial8 RC": {
    "type": "basic",
    "stimuli": ["ResponseCueLeft", "BlankGrid5"],
    "auto_advance": true
  },
 
  "Trial8Critical": {
    "type": "basic",
    "stimuli": ["GreenBottomCRS", "BlueTopCRS"],
    "responses": ["TrainingResponseTargetF", "TrainingResponseTargetJ"]
  },
 
  "Trial8Pauses": {
    "type": "basic",
    "stimuli": ["BlankGrid5", "BlankGrid2"],
    "auto_advance": true

The response cue stimulus has a stimulus level duration of 0.5 seconds and the barrier set to "true" and the trial-template is set to auto-advance. 
In the procedures, this is set up so that the block will look as follows: 

"TrialBlock5": {
      "trial_templates": ["Trial5Critical", "Trial5Pauses"],
      "pattern": {"order": "alternate", "repeat": 1},
      "cover_trials": ["Trial5 RC"].

The reason I don't think I can incorporate catch trials into my study is because I already have the "Trial[x] RC" as a cover trial. If I put the "Trial[x] RC" templates into the "trial_templates": [...], along with the Trial[x]Critical and Trial[x]Pause, then I won't be able to alternate the pauses between the two critical stimuli, because FF won't be able to evenly distribute the stimuli. I've tried using the catch trials with the cover trial (like how you can have a cover and end trial) but the stimuli were presented out of order.

Do you have any thoughts on this?

Thank you so much!

Grace

Hi again,

It seems that the font in my first paragraph in my above post is unreadably small. I will copy and paste it here:

"I have a question about catch trials. I have 224 experimental trials total (seven blocks of 32 trials), and I am hoping to put six short breaks after the first six blocks of trials. My trials are randomized, so if I put the blocks into my block sequence in my procedure, the breaks get randomized with the trials (which I don't want)."

Sorry about that![/quote]
Reply
#3
Hi Victoria, 

Thank you!

My sequence in procedures looks as follows: 

(there are intro and training blocks here) ...{"$randomize": ["TrialBlock1", "TrialBlock2", "TrialBlock3", "TrialBlock4", "TrialBlock5", "TrialBlock6", "TrialBlock7", "TrialBlock8", "TrialBlock9", "TrialBlock10", "TrialBlock11", "TrialBlock12", "TrialBlock13", "TrialBlock14", "TrialBlock15", "TrialBlock16", "TrialBlock17", "TrialBlock18", "TrialBlock19", "TrialBlock20", "TrialBlock21", "TrialBlock22", "TrialBlock23", "TrialBlock24", "TrialBlock25", "TrialBlock26", "TrialBlock27", "TrialBlock28", "TrialBlock29", "TrialBlock30", "TrialBlock31", "TrialBlock32" ... (there are seven more chunks of these 32 trials, for a total of 224 randomized trial blocks). 

I am hoping to have the breaks after every one of these 32 trials, no matter the order the actual trials are presented in. For example, within these 32 trials blocks here, they could be presented in the random order "TrialBlock9", "TrialBlock12", "TrialBlock18", "TrialBlock1" so on and so forth. But regardless, a break would be presented after the 32nd trial. 
I tried using "end_trials" as you suggested, but with that, the breaks are presented as an end to one trial block. Because those trial blocks are randomized, if I did an end_trial on TrialBlock32, but TrialBlock32 was presented as, say, the fifth trial block in the randomized block sequence, the break would also be presented there. I also tried making a trial template as in your first suggestion, with creating a trial template for the break in trial_templates, creating a block for it in the procedure, and then adding it to the block sequence. However, it was randomized in the same way as the end_trial option was. 

That all said, it sounds like there wouldn't be a way to implement catch trials into this study given its design?

Thank you so much!

Grace
Reply
#4
Hi Grace,

I think given the way you implemented the block sequence, the answer is probably no I'm afraid. Catch trials work within the block, but not between blocks which seems to be what you are after.

That may be the unfortunate conclusion at the moment.
Reply
#5
(11-12-2021, 08:51 PM)Ting Wrote: Hi Grace,

I think given the way you implemented the block sequence, the answer is probably no I'm afraid. Catch trials work within the block, but not between blocks which seems to be what you are after.

That may be the unfortunate conclusion at the moment.

Hi Ting, 

Thank you so much! That is really helpful to have confirmed. 

Grace
Reply


Forum Jump:


Users browsing this thread: 1 Guest(s)