Possible Stairway using 3 stimuli

Back

Possible Stairway using 3 stimuli

Author	Message
Dave	Dave posted 8 Years Ago ANSWER Post Details Share Post
Group: Administrators Posts: 13K, Visits: 109K	+x capr - Tuesday, July 25, 2017 +x Dave - Thursday, July 6, 2017 +x capr - Thursday, July 6, 2017 This is probably going to be a long shot but I thought I would give this a try. I would like to create an odd one out task, which would require to the participant to be able to identify which of three stimuli is different from the other two. For this I would require for three jpegs be shown on one screen in which two of them are the same and one is different (in which the location of the different image is randomized). As I am trying to detect a threshold, i would need these images to change depending on the response being correct or incorrect. Another issue is that i want to set it up so that i can make my stimuli at the full range of what the experiment is (make every item from 1-100) and then have inquisit randomly select one of these items, copy it for another face and then take the required difference in range for the difference in the staircase and make the third image there. For example, have a screen in which three seperate jpegs come up all next to each other in a row with two at the highest point in the range (100) and one at the lowest point (0), if the response would be right then that difference would go shorter, then requiring two stimuli of a random value (e.g. 75) and the third of a smaller range of the stimuli (e.g. 50 difference, meaning stimulus 25) all whilst the odd image out is being randomised between the locations. I know that it is not the most easy of tasks but the amount of coding and stimulus making i would have to produce if I cannot get three seperate jpegs would be a lot. I know this is a long shot and I am vary new to inquisit and coding so any instructions or advice would be majorly appreciated. Also if there suggestions of other ways to do something similar, I am more than happy to hear them out. I have looked at the staircase method that is in the examples but was not sure how to manipulate that to what I need. Thanks > I would like to create an odd one out task, which would require to the participant to be able to identify which of three stimuli is different from the other two. > For this I would require for three jpegs be shown on one screen in which two of them are the same and one is different (in which the location of the > different image is randomized). This is reasonably straightforward to do. You need three <picture> elements -- <picture same1>, <picture same2>, and <picture different> -- with same1 and same2 set to select the same item number, and <picture different> selecting a different one. This can be done using values and lists. You then simply randomize the picture elements' on-screen positions also using a <list> containing the three candidate positions. Basic example using <text> stimuli (works the same for <picture>s): <block myblock> / trials = [1-10 = oddoneout] </block> <trial oddoneout> / ontrialbegin = [ values.sameitem=list.sameitemnumbers.nextindex; values.differentitem=list.differentitemnumbers.nextindex; ] / ontrialbegin = [ values.same1_x=list.xpositions.nextvalue; values.same2_x=list.xpositions.nextvalue; values.different_x=list.xpositions.nextvalue; ] / stimulusframes = [1=same1, same2, different] / inputdevice = mouse / validresponse = (same1, same2, different) / correctresponse = (different) </trial> <text same1> / items = allitems / select = values.sameitem / hposition = values.same1_x </text> <text same2> / items = allitems / select = values.sameitem / hposition = values.same2_x </text> <text different> / items = allitems / select = values.differentitem / hposition = values.different_x </text> <item allitems> / 1 = "A" / 2 = "B" / 3 = "C" / 4 = "D" / 5 = "E" / 6 = "F" / 7 = "G" / 8 = "H" / 9 = "I" / 10 = "J" </item> <list sameitemnumbers> / poolsize = 10 </list> <list differentitemnumbers> / poolsize = 10 / not = (list.sameitemnumbers.currentindex) </list> <list xpositions> / items = (25%, 50%, 75%) / selectionrate = always </list> <values> / sameitem = 1 / differentitem = 1 / same1_x = 0% / same2_x = 0% / different_x = 0% </values> As for the staircase / adjustment procedure, there is too little detail for me to offer concrete syntax or code. As the above should make clear, however, what it ultimately comes down to is mapping the adjustment to _item numbers_. I.e., based on response correctness, constrain item selection to a specific range of item numbers / calculate the applicable item number for the "different" stimulus using some (more or less) simple math. Thank You for your help as this worked perfectly for the presentation of Stimuli. However by myself I have not been able to figure out how to get the staircase portion of it to work. I will try my best to explain to see if you can help me out as I have used the above code as a template of what I have so far. So my stimuli, (The pictures in items list) would be be organised in order from 0-100, so that the item 0 and item 100 are the most different and then as they come closer together they get closer together. In real terms the items are just one stimuli that I have manipulated to range from 0% manipulation to 100% Manipulation, and I am trying to discover how much manipulation a person can actually identify. As well as this I am aware that the easiest way to do this would be to have the manipulation start at either 0 or 100, and then let it change from there. The amount of change i would like it to be halved (for example if they identify correct where the same stimuli is 100% and the different is 0%, the same stimuli would stay at 100% and the different would move to 50%, if they identified that they would move on to 25%, then 13% (rounded up to whole number)). After 3 incorrect responses it would end the trail For this experiment I would also like there to be two staiways in which one would decrease the difference in stimuli (as above) and one would increase the difference (Start at 100% same, 99% different; then go onto 98% different; then 96%; then 92%; etc..), .Also for this increasing staircase it would cancel after 3 correct responses. I was not sure if it would be an easy thing to be able to complete and once again any help would be amazing. "have the manipulation start at either 0 or 100, and then let it change from there. The amount of change i would like it to be halved (for example if they identify correct where the same stimuli is 100% and the different is 0%, the same stimuli would stay at 100% and the different would move to 50%" I'm not sure what the problem is: If you arrange your item as numbers 1 to 101, it seems to me that you should be to calculate the proper item number for the next trial quite straightforwardly if the response in the current trial is correct. I must be missing something. What exactly are you not sure how to do? I'm also not clear on what you plan to do in case of an incorrect response. Are stimuli supposed to remain at the same "level" or change in some way?
	Reply Like 59
capr	capr posted 8 Years Ago ANSWER Post Details Share Post
Group: Forum Members Posts: 7, Visits: 22	+x Dave - Thursday, July 6, 2017 +x capr - Thursday, July 6, 2017 This is probably going to be a long shot but I thought I would give this a try. I would like to create an odd one out task, which would require to the participant to be able to identify which of three stimuli is different from the other two. For this I would require for three jpegs be shown on one screen in which two of them are the same and one is different (in which the location of the different image is randomized). As I am trying to detect a threshold, i would need these images to change depending on the response being correct or incorrect. Another issue is that i want to set it up so that i can make my stimuli at the full range of what the experiment is (make every item from 1-100) and then have inquisit randomly select one of these items, copy it for another face and then take the required difference in range for the difference in the staircase and make the third image there. For example, have a screen in which three seperate jpegs come up all next to each other in a row with two at the highest point in the range (100) and one at the lowest point (0), if the response would be right then that difference would go shorter, then requiring two stimuli of a random value (e.g. 75) and the third of a smaller range of the stimuli (e.g. 50 difference, meaning stimulus 25) all whilst the odd image out is being randomised between the locations. I know that it is not the most easy of tasks but the amount of coding and stimulus making i would have to produce if I cannot get three seperate jpegs would be a lot. I know this is a long shot and I am vary new to inquisit and coding so any instructions or advice would be majorly appreciated. Also if there suggestions of other ways to do something similar, I am more than happy to hear them out. I have looked at the staircase method that is in the examples but was not sure how to manipulate that to what I need. Thanks > I would like to create an odd one out task, which would require to the participant to be able to identify which of three stimuli is different from the other two. > For this I would require for three jpegs be shown on one screen in which two of them are the same and one is different (in which the location of the > different image is randomized). This is reasonably straightforward to do. You need three <picture> elements -- <picture same1>, <picture same2>, and <picture different> -- with same1 and same2 set to select the same item number, and <picture different> selecting a different one. This can be done using values and lists. You then simply randomize the picture elements' on-screen positions also using a <list> containing the three candidate positions. Basic example using <text> stimuli (works the same for <picture>s): <block myblock> / trials = [1-10 = oddoneout] </block> <trial oddoneout> / ontrialbegin = [ values.sameitem=list.sameitemnumbers.nextindex; values.differentitem=list.differentitemnumbers.nextindex; ] / ontrialbegin = [ values.same1_x=list.xpositions.nextvalue; values.same2_x=list.xpositions.nextvalue; values.different_x=list.xpositions.nextvalue; ] / stimulusframes = [1=same1, same2, different] / inputdevice = mouse / validresponse = (same1, same2, different) / correctresponse = (different) </trial> <text same1> / items = allitems / select = values.sameitem / hposition = values.same1_x </text> <text same2> / items = allitems / select = values.sameitem / hposition = values.same2_x </text> <text different> / items = allitems / select = values.differentitem / hposition = values.different_x </text> <item allitems> / 1 = "A" / 2 = "B" / 3 = "C" / 4 = "D" / 5 = "E" / 6 = "F" / 7 = "G" / 8 = "H" / 9 = "I" / 10 = "J" </item> <list sameitemnumbers> / poolsize = 10 </list> <list differentitemnumbers> / poolsize = 10 / not = (list.sameitemnumbers.currentindex) </list> <list xpositions> / items = (25%, 50%, 75%) / selectionrate = always </list> <values> / sameitem = 1 / differentitem = 1 / same1_x = 0% / same2_x = 0% / different_x = 0% </values> As for the staircase / adjustment procedure, there is too little detail for me to offer concrete syntax or code. As the above should make clear, however, what it ultimately comes down to is mapping the adjustment to _item numbers_. I.e., based on response correctness, constrain item selection to a specific range of item numbers / calculate the applicable item number for the "different" stimulus using some (more or less) simple math. Thank You for your help as this worked perfectly for the presentation of Stimuli. However by myself I have not been able to figure out how to get the staircase portion of it to work. I will try my best to explain to see if you can help me out as I have used the above code as a template of what I have so far. So my stimuli, (The pictures in items list) would be be organised in order from 0-100, so that the item 0 and item 100 are the most different and then as they come closer together they get closer together. In real terms the items are just one stimuli that I have manipulated to range from 0% manipulation to 100% Manipulation, and I am trying to discover how much manipulation a person can actually identify. As well as this I am aware that the easiest way to do this would be to have the manipulation start at either 0 or 100, and then let it change from there. The amount of change i would like it to be halved (for example if they identify correct where the same stimuli is 100% and the different is 0%, the same stimuli would stay at 100% and the different would move to 50%, if they identified that they would move on to 25%, then 13% (rounded up to whole number)). After 3 incorrect responses it would end the trail For this experiment I would also like there to be two staiways in which one would decrease the difference in stimuli (as above) and one would increase the difference (Start at 100% same, 99% different; then go onto 98% different; then 96%; then 92%; etc..), .Also for this increasing staircase it would cancel after 3 correct responses. I was not sure if it would be an easy thing to be able to complete and once again any help would be amazing.
	Reply Like 51
Dave	Dave posted 8 Years Ago ANSWER Post Details Share Post
Group: Administrators Posts: 13K, Visits: 109K	+x capr - Thursday, July 6, 2017 This is probably going to be a long shot but I thought I would give this a try. I would like to create an odd one out task, which would require to the participant to be able to identify which of three stimuli is different from the other two. For this I would require for three jpegs be shown on one screen in which two of them are the same and one is different (in which the location of the different image is randomized). As I am trying to detect a threshold, i would need these images to change depending on the response being correct or incorrect. Another issue is that i want to set it up so that i can make my stimuli at the full range of what the experiment is (make every item from 1-100) and then have inquisit randomly select one of these items, copy it for another face and then take the required difference in range for the difference in the staircase and make the third image there. For example, have a screen in which three seperate jpegs come up all next to each other in a row with two at the highest point in the range (100) and one at the lowest point (0), if the response would be right then that difference would go shorter, then requiring two stimuli of a random value (e.g. 75) and the third of a smaller range of the stimuli (e.g. 50 difference, meaning stimulus 25) all whilst the odd image out is being randomised between the locations. I know that it is not the most easy of tasks but the amount of coding and stimulus making i would have to produce if I cannot get three seperate jpegs would be a lot. I know this is a long shot and I am vary new to inquisit and coding so any instructions or advice would be majorly appreciated. Also if there suggestions of other ways to do something similar, I am more than happy to hear them out. I have looked at the staircase method that is in the examples but was not sure how to manipulate that to what I need. Thanks > I would like to create an odd one out task, which would require to the participant to be able to identify which of three stimuli is different from the other two. > For this I would require for three jpegs be shown on one screen in which two of them are the same and one is different (in which the location of the > different image is randomized). This is reasonably straightforward to do. You need three <picture> elements -- <picture same1>, <picture same2>, and <picture different> -- with same1 and same2 set to select the same item number, and <picture different> selecting a different one. This can be done using values and lists. You then simply randomize the picture elements' on-screen positions also using a <list> containing the three candidate positions. Basic example using <text> stimuli (works the same for <picture>s): <block myblock> / trials = [1-10 = oddoneout] </block> <trial oddoneout> / ontrialbegin = [ values.sameitem=list.sameitemnumbers.nextindex; values.differentitem=list.differentitemnumbers.nextindex; ] / ontrialbegin = [ values.same1_x=list.xpositions.nextvalue; values.same2_x=list.xpositions.nextvalue; values.different_x=list.xpositions.nextvalue; ] / stimulusframes = [1=same1, same2, different] / inputdevice = mouse / validresponse = (same1, same2, different) / correctresponse = (different) </trial> <text same1> / items = allitems / select = values.sameitem / hposition = values.same1_x </text> <text same2> / items = allitems / select = values.sameitem / hposition = values.same2_x </text> <text different> / items = allitems / select = values.differentitem / hposition = values.different_x </text> <item allitems> / 1 = "A" / 2 = "B" / 3 = "C" / 4 = "D" / 5 = "E" / 6 = "F" / 7 = "G" / 8 = "H" / 9 = "I" / 10 = "J" </item> <list sameitemnumbers> / poolsize = 10 </list> <list differentitemnumbers> / poolsize = 10 / not = (list.sameitemnumbers.currentindex) </list> <list xpositions> / items = (25%, 50%, 75%) / selectionrate = always </list> <values> / sameitem = 1 / differentitem = 1 / same1_x = 0% / same2_x = 0% / different_x = 0% </values> As for the staircase / adjustment procedure, there is too little detail for me to offer concrete syntax or code. As the above should make clear, however, what it ultimately comes down to is mapping the adjustment to _item numbers_. I.e., based on response correctness, constrain item selection to a specific range of item numbers / calculate the applicable item number for the "different" stimulus using some (more or less) simple math.
	Reply Like 52
capr	capr posted 8 Years Ago HOT Topic Details Share Topic
Group: Forum Members Posts: 7, Visits: 22	This is probably going to be a long shot but I thought I would give this a try. I would like to create an odd one out task, which would require to the participant to be able to identify which of three stimuli is different from the other two. For this I would require for three jpegs be shown on one screen in which two of them are the same and one is different (in which the location of the different image is randomized). As I am trying to detect a threshold, i would need these images to change depending on the response being correct or incorrect. Another issue is that i want to set it up so that i can make my stimuli at the full range of what the experiment is (make every item from 1-100) and then have inquisit randomly select one of these items, copy it for another face and then take the required difference in range for the difference in the staircase and make the third image there. For example, have a screen in which three seperate jpegs come up all next to each other in a row with two at the highest point in the range (100) and one at the lowest point (0), if the response would be right then that difference would go shorter, then requiring two stimuli of a random value (e.g. 75) and the third of a smaller range of the stimuli (e.g. 50 difference, meaning stimulus 25) all whilst the odd image out is being randomised between the locations. I know that it is not the most easy of tasks but the amount of coding and stimulus making i would have to produce if I cannot get three seperate jpegs would be a lot. I know this is a long shot and I am vary new to inquisit and coding so any instructions or advice would be majorly appreciated. Also if there suggestions of other ways to do something similar, I am more than happy to hear them out. I have looked at the staircase method that is in the examples but was not sure how to manipulate that to what I need. Thanks
	Reply Like 45