Millisecond Forums

+x

 

Dave - 1/14/2025

+x

 

Dave - 1/14/2025

+x

 

lir1995 - 1/14/2025

Hello, 

Thank you so much for helping us. We have one additional question.

In the test phase, the three faces are presented at fixed horizontal positions with the paired face always on the left, the pairmate face always in the center, and the unrelated face always on the right.

For our study, we need the horizontal positions of the three faces to be random, where the paired face can show up in any of the three horizontal positions. The vertical position can stay fixed. 

We tried to implement a counter, rather than the list that is currently being used. However, we were not successful in getting this to work

<counter positions>
/ items = (20%, 50%, 80%)
/ select = noreplace
/ selectionrate = always
</counter>

<trial testTrial>
/ ontrialbegin = [
    // pick scene to test and associated faces
    values.sceneItemnumber = list.sceneItemNumbers.nextvalue; // pick scene to test
    values.scenePairedFaceItemnumber = list.scenePairedFaceItemnumbers.nextvalue; // pick face paired to scene
    values.scenePairmateFaceItemnumber = list.scenePairmateFaceItemnumbers.nextvalue; // pick face paired to its pairmate scene
    values.sceneUnrelatedFaceItemnumber = list.sceneUnrelatedFaceItemnumbers.nextvalue; // pick a unrelated face out of the remaining (not current scene-paired or pairmate-paired)
    // randomize horizontal position of scene-paired face, pairmate-paired face, and unrelated random face
    values.scenePairedFacePositionnumber = list.testFacePositionNumbers.nextvalue;
    picture.testScenePairedFace.hposition = counter.positions.selectedvalue.item(values.scenePairedFacePositionnumber);
    values.scenePairmateFacePositionnumber = list.testFacePositionNumbers.nextvalue;
    picture.testScenePairmateFace.hposition = counter.positions.selectedvalue.item(values.scenePairmateFacePositionnumber);
    values.sceneUnrelatedFacePositionnumber = list.testFacePositionNumbers.nextvalue;
    picture.testSceneUnrelatedFace.hposition = counter.positions.selectedvalue.item(values.sceneUnrelatedFacePositionnumber);
    ]
/ stimulustimes = [0=testScene, testScenePairedFace, testScenePairmateFace, testSceneUnrelatedFace]
/ validresponse = ("1", "2", "3")
/ iscorrectresponse = [
    trial.testTrial.responsetext == values.scenePairedFacePositionNumber; // correct if position of scene-paired face identified
]
/ timeout = 5000
/ branch = [trial.pause
]
</trial>

Can you please help us with randomizing the horizontal positions of the faces during the test phase? Thanks!

You should not, under any circumstances, use <counter> elements anymore. They have been obsolete since Inquisit 4.

There is nothing counters could do that lists cannot do, and do better.

Moreover, in the code I gave you, the test stimuli are not displayed in a fixed order. Their horizontal positions are randomized in every trial.

+x

 

lir1995 - 1/14/2025

+x

 

Dave - 1/14/2025

+x

 

Dave - 1/14/2025

+x

 

lir1995 - 1/14/2025

Hello, 

Thank you so much for helping us. We have one additional question.

In the test phase, the three faces are presented at fixed horizontal positions with the paired face always on the left, the pairmate face always in the center, and the unrelated face always on the right.

For our study, we need the horizontal positions of the three faces to be random, where the paired face can show up in any of the three horizontal positions. The vertical position can stay fixed. 

We tried to implement a counter, rather than the list that is currently being used. However, we were not successful in getting this to work

<counter positions>
/ items = (20%, 50%, 80%)
/ select = noreplace
/ selectionrate = always
</counter>

<trial testTrial>
/ ontrialbegin = [
    // pick scene to test and associated faces
    values.sceneItemnumber = list.sceneItemNumbers.nextvalue; // pick scene to test
    values.scenePairedFaceItemnumber = list.scenePairedFaceItemnumbers.nextvalue; // pick face paired to scene
    values.scenePairmateFaceItemnumber = list.scenePairmateFaceItemnumbers.nextvalue; // pick face paired to its pairmate scene
    values.sceneUnrelatedFaceItemnumber = list.sceneUnrelatedFaceItemnumbers.nextvalue; // pick a unrelated face out of the remaining (not current scene-paired or pairmate-paired)
    // randomize horizontal position of scene-paired face, pairmate-paired face, and unrelated random face
    values.scenePairedFacePositionnumber = list.testFacePositionNumbers.nextvalue;
    picture.testScenePairedFace.hposition = counter.positions.selectedvalue.item(values.scenePairedFacePositionnumber);
    values.scenePairmateFacePositionnumber = list.testFacePositionNumbers.nextvalue;
    picture.testScenePairmateFace.hposition = counter.positions.selectedvalue.item(values.scenePairmateFacePositionnumber);
    values.sceneUnrelatedFacePositionnumber = list.testFacePositionNumbers.nextvalue;
    picture.testSceneUnrelatedFace.hposition = counter.positions.selectedvalue.item(values.sceneUnrelatedFacePositionnumber);
    ]
/ stimulustimes = [0=testScene, testScenePairedFace, testScenePairmateFace, testSceneUnrelatedFace]
/ validresponse = ("1", "2", "3")
/ iscorrectresponse = [
    trial.testTrial.responsetext == values.scenePairedFacePositionNumber; // correct if position of scene-paired face identified
]
/ timeout = 5000
/ branch = [trial.pause
]
</trial>

Can you please help us with randomizing the horizontal positions of the faces during the test phase? Thanks!

You should not, under any circumstances, use <counter> elements anymore. They have been obsolete since Inquisit 4.

There is nothing counters could do that lists cannot do, and do better.

Moreover, in the code I gave you, the test stimuli are not displayed in a fixed order. Their horizontal positions are randomized in every trial.

Ah, I did not realize. I won't use counters anymore!


Stimulus item 1                                            Stimulus item 2. Stimulus item 3.    Stimulus item 4
scene_files_test/Scenes/Bridge 1.jpg          Faces/9.jpg        0
scene_files_test/Scenes/Dock 1.jpg            Faces/10.jpg        0
scene_files_test/Scenes/Bridge 2.jpg          Faces/3.jpg        0
scene_files_test/Scenes/Auditorium 2.jpg    Faces/1.jpg        0
scene_files_test/Scenes/Arcade 2.jpg          Faces/2.jpg        0
scene_files_test/Scenes/Hallway 2.jpg         Faces/5.jpg        0
scene_files_test/Scenes/Auditorium 1.jpg    Faces/8.jpg        0
scene_files_test/Scenes/Dock 2.jpg              Faces/4.jpg        0
scene_files_test/Scenes/Hallway 1.jpg         Faces/6.jpg        0
scene_files_test/Scenes/Arcade 1.jpg          Faces/7.jpg        0
scene_files_test/Scenes/Hallway 1.jpg         Faces/6.jpg    Faces/5.jpg      Faces/9.jpg
scene_files_test/Scenes/Dock 2.jpg              Faces/4.jpg    Faces/10.jpg    Faces/6.jpg
scene_files_test/Scenes/Auditorium 1.jpg    Faces/8.jpg    Faces/1.jpg     Faces/3.jpg
scene_files_test/Scenes/Auditorium 2.jpg    Faces/1.jpg    Faces/8.jpg     Faces/4.jpg
scene_files_test/Scenes/Bridge 2.jpg          Faces/3.jpg    Faces/9.jpg     Faces/7.jpg
scene_files_test/Scenes/Arcade 1.jpg        Faces/7.jpg    Faces/2.jpg     Faces/10.jpg
scene_files_test/Scenes/Bridge 1.jpg         Faces/9.jpg    Faces/3.jpg     Faces/8.jpg
scene_files_test/Scenes/Dock 1.jpg          Faces/10.jpg    Faces/4.jpg    Faces/2.jpg
scene_files_test/Scenes/Arcade 2.jpg       Faces/2.jpg    Faces/7.jpg     Faces/5.jpg
scene_files_test/Scenes/Hallway 2.jpg      Faces/5.jpg    Faces/6.jpg     Faces/1.jpg


I copied the test data you attached earlier. For instance, Bridge 1 is paired with face 9. When it is presented again, face 9 is on the left and face 3 is in the middle which is the one paired with bridge 2. The pattern is the same for all of the items. I can't have this for my study because then the correct answer is always the left one. This seems to happen every time I run the code. I am not sure how to fix this with list. I apologize if I am missing something, and really appreciate the help!

+x

 

Dave - 1/14/2025

+x

 

lir1995 - 1/14/2025

+x

 

Dave - 1/14/2025

+x

 

Dave - 1/14/2025

+x

 

lir1995 - 1/14/2025

Hello, 

Thank you so much for helping us. We have one additional question.

In the test phase, the three faces are presented at fixed horizontal positions with the paired face always on the left, the pairmate face always in the center, and the unrelated face always on the right.

For our study, we need the horizontal positions of the three faces to be random, where the paired face can show up in any of the three horizontal positions. The vertical position can stay fixed. 

We tried to implement a counter, rather than the list that is currently being used. However, we were not successful in getting this to work

<counter positions>
/ items = (20%, 50%, 80%)
/ select = noreplace
/ selectionrate = always
</counter>

<trial testTrial>
/ ontrialbegin = [
    // pick scene to test and associated faces
    values.sceneItemnumber = list.sceneItemNumbers.nextvalue; // pick scene to test
    values.scenePairedFaceItemnumber = list.scenePairedFaceItemnumbers.nextvalue; // pick face paired to scene
    values.scenePairmateFaceItemnumber = list.scenePairmateFaceItemnumbers.nextvalue; // pick face paired to its pairmate scene
    values.sceneUnrelatedFaceItemnumber = list.sceneUnrelatedFaceItemnumbers.nextvalue; // pick a unrelated face out of the remaining (not current scene-paired or pairmate-paired)
    // randomize horizontal position of scene-paired face, pairmate-paired face, and unrelated random face
    values.scenePairedFacePositionnumber = list.testFacePositionNumbers.nextvalue;
    picture.testScenePairedFace.hposition = counter.positions.selectedvalue.item(values.scenePairedFacePositionnumber);
    values.scenePairmateFacePositionnumber = list.testFacePositionNumbers.nextvalue;
    picture.testScenePairmateFace.hposition = counter.positions.selectedvalue.item(values.scenePairmateFacePositionnumber);
    values.sceneUnrelatedFacePositionnumber = list.testFacePositionNumbers.nextvalue;
    picture.testSceneUnrelatedFace.hposition = counter.positions.selectedvalue.item(values.sceneUnrelatedFacePositionnumber);
    ]
/ stimulustimes = [0=testScene, testScenePairedFace, testScenePairmateFace, testSceneUnrelatedFace]
/ validresponse = ("1", "2", "3")
/ iscorrectresponse = [
    trial.testTrial.responsetext == values.scenePairedFacePositionNumber; // correct if position of scene-paired face identified
]
/ timeout = 5000
/ branch = [trial.pause
]
</trial>

Can you please help us with randomizing the horizontal positions of the faces during the test phase? Thanks!

You should not, under any circumstances, use <counter> elements anymore. They have been obsolete since Inquisit 4.

There is nothing counters could do that lists cannot do, and do better.

Moreover, in the code I gave you, the test stimuli are not displayed in a fixed order. Their horizontal positions are randomized in every trial.

Ah, I did not realize. I won't use counters anymore!


Stimulus item 1                                            Stimulus item 2. Stimulus item 3.    Stimulus item 4
scene_files_test/Scenes/Bridge 1.jpg          Faces/9.jpg        0
scene_files_test/Scenes/Dock 1.jpg            Faces/10.jpg        0
scene_files_test/Scenes/Bridge 2.jpg          Faces/3.jpg        0
scene_files_test/Scenes/Auditorium 2.jpg    Faces/1.jpg        0
scene_files_test/Scenes/Arcade 2.jpg          Faces/2.jpg        0
scene_files_test/Scenes/Hallway 2.jpg         Faces/5.jpg        0
scene_files_test/Scenes/Auditorium 1.jpg    Faces/8.jpg        0
scene_files_test/Scenes/Dock 2.jpg              Faces/4.jpg        0
scene_files_test/Scenes/Hallway 1.jpg         Faces/6.jpg        0
scene_files_test/Scenes/Arcade 1.jpg          Faces/7.jpg        0
scene_files_test/Scenes/Hallway 1.jpg         Faces/6.jpg    Faces/5.jpg      Faces/9.jpg
scene_files_test/Scenes/Dock 2.jpg              Faces/4.jpg    Faces/10.jpg    Faces/6.jpg
scene_files_test/Scenes/Auditorium 1.jpg    Faces/8.jpg    Faces/1.jpg     Faces/3.jpg
scene_files_test/Scenes/Auditorium 2.jpg    Faces/1.jpg    Faces/8.jpg     Faces/4.jpg
scene_files_test/Scenes/Bridge 2.jpg          Faces/3.jpg    Faces/9.jpg     Faces/7.jpg
scene_files_test/Scenes/Arcade 1.jpg        Faces/7.jpg    Faces/2.jpg     Faces/10.jpg
scene_files_test/Scenes/Bridge 1.jpg         Faces/9.jpg    Faces/3.jpg     Faces/8.jpg
scene_files_test/Scenes/Dock 1.jpg          Faces/10.jpg    Faces/4.jpg    Faces/2.jpg
scene_files_test/Scenes/Arcade 2.jpg       Faces/2.jpg    Faces/7.jpg     Faces/5.jpg
scene_files_test/Scenes/Hallway 2.jpg      Faces/5.jpg    Faces/6.jpg     Faces/1.jpg


I copied the test data you attached earlier. For instance, Bridge 1 is paired with face 9. When it is presented again, face 9 is on the left and face 3 is in the middle which is the one paired with bridge 2. The pattern is the same for all of the items. I can't have this for my study because then the correct answer is always the left one. This seems to happen every time I run the code. I am not sure how to fix this with list. I apologize if I am missing something, and really appreciate the help!

The column order in the data file has absolutely nothng to do with on-screen positions. It merely reflects the order in which the trial draws stimuli to screen.

There are even variables in the data file -- values.scenePairedFacePositionnumber, values.scenePairmateFacePositionnumber, and values.sceneUnrelatedFacePositionnumber -- that tell you which on.screen position every stimulus was in (1 = left, 2 = middle, 3 = right).