| second mtg | third mtg | fourth mtg | fifth mtg | sixth mtg | seventh mtg | eighth mtg | ninth mtg | tenth mtg | eleventh | twelfth | thirteenth | fourteenth |fifteenth |
No meeting. No information sent yet.
Present: Dr. Ryan, Wesley Hopkins
Explained plans for Comparing Problems 3. The idea will be to cross the main factor from Comparing Problems 2 , MA/ANAL vs ANAL/ANAL with the Match Features vs. Explain Steps factor of Dis 02. I gave Wes the ms about Dis 02 that was not accepted by Cog Tech along with most of the materials from Dis 02.
Present: Dr. Ryan, Laura Kise
Told her what Wes will be working on. Gave her the materials I gave to Wes. I think Wes did not get the materials for the Guided Practice, but did get the whole set of problem pairs for the unguided practice. I made sure Laura got the materials for the Guided Practice, but did not print out all the pairs for the unguided practice for her.
We agreed she would contact Wes to let him know she would be working on CP 3 with him, and that they would get together, and then get back to me.
The task will be to start altering the materials from CP2 and Dis02 to make them comparable. For example Dis02 had a pre/post design, whereas CP2 did not. They used different tests. CP2 used both MA/ANAL and ANAL/ANAL pairs. Dis02 used all ANAL/ANAL pairs, although I'm not certain that the Dis02 problems were anals in the same way that the CP2 problems were.
| back to top | next |
Present: Dr. Ryan, Andrew Malkiewicz, Nicole Yeager
Andy helped Moshe Comparing Problems 2. He will contact Wesley and Laura and work on CP3.
I trained Nicole Yeager to code Local Global. She did one subject who had been done previously (301) as a test. We discovered that whoever coded 300 on made errors on the Fals and Ident variables. Nicole will check them.
I gave the xls file with the local global data to Nicole on a disk. She will work in 294A with the disk and the raw data to do the coding.
**** On Wed. 9-7-05 Erica Kleinbaum stopped in and volunteered. I got her started. *****
Schedule conflict with Psych major's meeting. The Thursday meeting will be changed to Wednesdays at 9:00 am to avoid such conflicts.
Michkal will run PR2. Will contact me with a time we could run together to get her started.
Joe has created the first time slot for PR2.
Present: Dr. Ryan, Andrew Malkiewicz, Alyssa Rizzo, Laura Kise, Jackie Maniet
Today I emailed Steve Corcoran to make sure the following are on the folder:
Andrew Malkiewicz
Laura Kise
Alyssa Rizzo
Jackie Maniet - Definitely new.
Alyssa tried to log into the subject pool using the new email ID and was not able to. She will try again using the original user ID for the subject pool which should work. She will then create time slots.
In this meeting, I worked with Andy, Laura, and Jackie on how we will modify CP2 and DS2 to design CP3. They took copies of the training problems from both studies. They will work on instructions for the match features training. The trick will be how to train people to match features in a condition where the two problems being matched are a weighted averaging of ratios and an adding of ratios. (That is the ANAL/MA condition).
| previous | back to top | next |
Present: Dr. Ryan, Krysta Murray
From experimental class. Got her started. She will work on Local/Global. May be interested in psych and the law.
Present: Dr. Ryan, Wesley, Mary Finn, Robert Campomizzi
Got Mary's info updated in the excel file.
More discussion of how to design CP3. The major question: How do we instruct participants to do a feature matching in a condition where the two training problems are (a) an analogy to the test problems, and (b) a mere appearance to the test problems. The MA problem will not be structurally or procedurally similar to its mate in the training pair. Do we teach the participants deceptively? No. In the MA condition, we teach them whatever corresponds on a surface level and simply do not mention anything about procedural or principle structure. In the ANAL condition, we also only teach them whatever corresponds on a surface level. In that condition, as opposed to in the MA condition, the features that correspond superficially will also correspond at the level of a procedural and a principle structure, but we will not mention that. Thus, we are creating two conditions that are alike except that in the ANAL condition, the participants will have the ability to recognize the procedural structure and the principle structure similarity, but they will have to infer it, because it will not be made explicit. In the MA condition, they will not have the ability to recognize such similarity because it will not be there. However, they will still be able to recognize the procedural and/or principle structure if they can infer it from the procedural and principle difference. Again, this difference will not be made explicit, but they will be able to infer it.
Just as a side note - Here is another possibly interesting question:
Does recognizing surface feature similarities support the later recognition
of procedural structure, principle structure, or both? Does it depend upon
whether the surface feature similarities are between problems with similarities
in their structures, or between problems with differences in their structures?
No meeting. Schedule conflict.
| previous | back to top | next |
No one present.
Present: Dr. Ryan, Wesley Hopkins, Robert Campomizzi
Worked on CP3. How to instruct feature matching in a MA/ANAL condition. Wes suggested that in the ANAL problem, the second item purchased should receive 100% discount by means of a store credit. This problem could still be solved as a weighted averaging. In the ANAL/ANAL condition it could be feature matched with another ANAL in which the second item had a normal discount of less than 100%. It could be feature matched with a "find initial" problem in which the discount of the first item purchased would be the goal. But in the MA/ANAL condition, both the MA and ANAL problem could be described only in terms of the final purchase price, discount percent, and amount saved. This would be motivated by saying that in both problems, what matters is the total purchase price before any discounts or credits, and the actual amount paid after whatever discounts and credits were applied.
This method of feature matching in the MA/ANAL condition will maximize the chances of getting an effect between the MA/ANAL and the ANAL/ANAL conditions, even though, if there is such an effect, there will be a question of what exactly caused it. But if such an effect occurs, there can always be a second experiment to answer the new question.
If the effect is just between Feature Matching, vs. Explaining Steps, then a second experiment won't be necessary. It would mean that people can infer the weighted averaging principle just from Feature Matching alone, even if there is no matching of finals and initials. This would suggest that people can infer the initial/ final structure and the weighted averaging principle just from feature mathing of purchase prices, amounts, and discount percents of the final elements.
If there is an effect between the MA/ANAL and the ANAL/ANAL conditions, then the second experiment can compare ANAL/ANAL with MA/ANAL(match all elements), and with MA/ANAL(match final elements).
We still need to consider whether there will be any problem by using both fin/fin and fin/init. Maybe all fin/init.
How about an experiment comparing these two types of feature matching:
purchase prices with times, discount percents with speeds, amounts saved
with times.
vs.
initial elements with intial elements, final elements with final elements,
givens with unknowns.
Present: Dr. Ryan, Laura Kise, Mary Finn, (1st half hour), Joe Cipko, Alyssa Rizzo (2nd half hour), also new (whole time) Sarah Brandon, Danielle Witherite, and (not new, but returning and becoming more active) Kelsey Kowalski.
1st Half Hour - Brought Laura and Mary up to speed on what I did with Wes and Robert on Tues. 9-20-05. We also worked on the question of whether the problem pairs for CP3 should consist of always same procedure, or sometimes same but sometimes different procedure (as in DS2). We decided that since the MA/ANAL condition will have pairs where there is always the same surface features and always different procedures and principles, then the ANAL/ANAL condition should be its mirror image. That is, the surface features should always be different, whereas the procedures and principles should always be the same. Thus, unlike what I suggested on 9-20-05, all the pairs would be fin/fin, not all fin/init. That means we also need to think about what the test should be like. That is, like DS2 or like CP2?
2nd Half Hour - Discussed what to Joe and Alyssa will do for a poster. They will prepare a poster for the KU undergrad conference, and then also plan on presenting it (perhaps with some refinements) at either APS, or EPA (probably APS - then something from Local/Global could go to EPA). Joe and Alyssa's poster will examine the characteristics of insects that contribute to fear, disgust, or both. The finding was that, over a wide range, the characteristics that contribute to one contribute to the other. But in a limited range it is possible to separate them.
The PR2 results will be saved for possible submission to Psychological Science or elsewhere.
Also, I got Kelsey's info into my database. I got Danielle and Sarah's names, SS#'s, and email addresses into the database and will get more of their info later.
| previous | back to top | next |
No one present.
Present: Dr. Ryan, Robert Campomizzi
We considered what to do about the test, and whether or not to use any find initial problems. Regarding the test, the question is whether to use the test from CP2 or from DS2. Regarding the find initial problems, the question is whether or not to use any pairs in ANAL/ANAL where the comparison is between a SFW and a DIW (instead of all SFW/DFW).
We decided not to use any find initial problems in the training because none were used in CP2. The change from CP2 to CP3 is that we are adding more training, and we are adding the manipulation of whether the subjects are trained to match features or not. If we use any find initial problems, they will be used only in the ANAL/ANAL condition. That would result in changing two things, instead of just one. We would be both adding the new kinds of training, but we would be doing so with different problems.
We tried to decide which test to use. We have two choices. We could create an equivalent version to the 9 problem test used in CP2, so that we would have a pre and a post test. Or we could use the 4 problem pre post test from DS2. The 4 problems from DS2 were all used on CP2. If we go with the first option, we have a slightly longer experiment and more work to do, but the equivalency between CP2 and 3 is better. We could go with the second option, which would be easier and shorter. But the equivalency would not be as good, and we would get less info.
We should go with option 1 and write (or find) 9 problems equivalent to the 9 problems used in CP2. Some of the problems from DS2 can be used, because we had two equivalent tests there.
Present: Dr. Ryan, Laura Kise, Mary Finn, Kelsey Kowalski, Jackie Maniet
I brought them up to date on everything I went over with Robert on Wednesday. I gave them a packet consisting of my hand drawn diagrams of the experiments (I did a new one today showing CP2, DS2, and CP3), and the test problems from CP2 and DS2. They will now work on the 9 equivalent test problems.
The next big task will be making up multiple pairs of training problems. In DS2 I started with something like 16 pairs. In CP2 (and some of its predecessors) there are more than one pair for ANAL/ANAL and MA/ANAL, but I don't think there are 16 pairs.
Presumably, Wesley and Robert will continue working on developing the scripts. Once today's group plus Robert produce the training and test materials, Wesley can incorporate them into the scripts. Today's group can also be involved in producing the scripts.
| previous | back to top | next |
No meeting due to Psych. Dept. Meeting.
No one present.
Present: Dr. Ryan, Mary Finn, Laura Kise, Jackie Maniet, Andrew Malkiewicz
We worked further on the 9 new problems for the equivalent test for CP3. I marked Mary Finn's version of the test from CP2 on which all had written their drafts of the new problems. My notations indicated what to change. I also provided a diagram of how to select the numbers for the problems so as to avoid decimals.
Present: Dr. Ryan, Kelly Breeze
I explained Local/Global, Picture Rating, and, very briefly, CP3.
Present: Dr. Ryan, Robert Campomizzi, Wesley Hopkins
I requested Kelsy Kowalski, Kelly Breeze, and Jessica Bergmaier to be added to the folder, and Michkal Young to be deleted.
I have not yet added Danielle Witherite, or Sarah Brandon to the folder. I need to find out if they are going to attend more meetings and get involved. I sent them an email letting them know.
Wes, Robert, and I went over where we are in the development of CP3. I had written all that up in a file called "Development of CP3.doc" in the CP3 folder. They will now work on the scripts and training problems, while the others on this project work on the test problems. The "Development" document provides one training pair as a worked example for the MF-ANAL/ANAL condition, and one for the MF-ANAL/MA condition. They will take the script from DS2 for condition 1 (MF) and adapt it for those two training pairs first. Next, they will have to adapt the script for the FC-ANAL/ANAL and the FC-ANAL/MA conditions, with just those two training pairs. Then, after we have checked that out, they will begin working on more training pairs and extend the script accordingly.
Present: Dr. Ryan, Mary Finn, Laura Kise, Kelsey Kowalski, Andrew Malkiewicz, Jacki Maniet
Kelsy Kowalski, Kelly Breeze, and Jessica Bergmaier have been added to the folder, and Michkal Young has been deleted.
We checked the test problems they have done so far and did some more editing. They now have a complete set to format. They will format the 18 problems (9 for test A, 9 for equivalent test B) in three ways. They will create a set for the researchers (ours) in which all the problems are listed on as few pages as possible. Then they will create a set for the actual experiment materials that the subjects will use (theirs). In that set, each problem will be on a separate page, and the page will have a place for the subject number, the condition, and the name of the experiment. Finally, they will create the Test Key. That will be the same as "theirs", but with the math worked out.
I also gave all of them a copy of the latest version of "Development of CP3" to read. That will help them prepare to run subjects once all the materials are developed.
Just a reminder - Morgan Becker continues to run Local / Global independently. She has trained Jessica Bergmaier. I stopped in to observe a session in which Jessica ran the subjects. Morgan just watched and continued to work on training Krysta Murray. Jessica is fine to run independently like Morgan is. Krysta will soon be able to do the same. That's two people running Local / Global independently now, with a third in the wings.
Another reminder. Joe Cipko and Alyssa Rizzo continue to run the face to face version of PR2 independently. The online version has already produced results suitable to submit for publication. It will be important to backtrack and get the gender of the participants associated with their data in the face to face version (as well as PR1) to examine this interesting gender effect that popped up in the online version of PR2.
Present: Dr. Ryan, Kelly Breeze.
Kelly is interested in the Local / Global study. I put her on the Subject Pool system. She will sit in on Local/Global and finish the Human Participants Protection training.
We held this meeting at about 2:30 to 3:00. If no one else needs the 5 - 6 pm time for this meeting, I may move the meeting with Kelly up to 3:00pm.
Present: Dr. Ryan, Robert C., Wesley H., and new - Chris O'Donnell
Wesley needs more problems for the training materials for CP3. We looked at the problems from DS2. The A problems, however, are mixture problems, not sales problems. The B problems are distance problems, so they can be used. So they will have to make up more sales problems. They can use the distance problems, although they will have to be adapted to be adding ratio problems for the ANAL/MA condition.
I gave them the diagram of how to create a weighted averaging of ratios problem that will end up with an even number (no decimals) for the answer.
Present: Dr. Ryan, Mary Finn, Laura Kise, and Jackie Maniet
They are almost done with the nine new test problems for the equivalent pre/post test. They will do a little more editing and bring it back to me.
| previous | back to top | next |
Present: Dr. Ryan, Dave Zlockie, Kelly Breeze, Kate Fleming
Explained the research program for Dave and Kate.
Dave will start working on the calendar.
Kate will start working on getting the gender info for the PR studies. We had trouble figuring out how to see all the timeslots, but she will contact Joe and see if he remembers how we did it.
Present: Dr. Ryan, Wesley, Chris O'Donnell.
Wes has started writing the scripts for the MF-ANAL/ANAL condition. Next he will do the MF ANAL/MA. He had used part of the script from DS2 for the Explain Steps condition, so I steered him towards using the DS2 script from the MA condition. Wes and Chris will plan how to get together to work on it, as well as contacting Rob C. to work out times they can all meet.
No meeting due Dr. Ryan's doctor's appointment.
| previous | back to top | next |
No meeting due to Psych. Dept. meeting.
Present: Dr. Ryan, Wes Hopkins, Chris O'Donnell.
Re the attempt to get gender info on PR1:
I found out that there is no way to get the gender information from PR1.
We did not ask for a student ID number to be saved in the system. Therefore,
there is no way to re-link names with the 5 digit sign-up codes.
Re CP3:
To deal with the issue of whether the subjects should be timed (3 min. )
or untimed (as in CP 1 & 2), Wes suggested that as a pilot study we just
give the test problems, but in one condition timed, and in the other untimed.
This will also enable us to offer more opportunities to participate.
There is an issue of whether or not to state the number of steps in the training problems. In the ANAL/MA condition, one will have 3 steps, but the other 2 steps. We decided to state the number, but make no issue of it.
For the part of the training where the subjects get instructions in how to MF an ANAL and an MA:
Now let's look at the bold faced type. Even though in the first problem you were looking for a discount percent and in the second problem you were looking for an amount saved, there are certain elements in the problems that match up. The percentages of discounts in the first problem match up with the percentages of discounts in the second. When you multiply the price of the suit and the price of the shoes by their discount percents you get amounts saved, and when you multiply the price the camera times the total discount amount you get the total amount saved. So total prices and total amounts saved match up.
Present: Dr. Ryan, Laura Kise, Kelsey Kowalski
We finished up "our" version and "their" version of Test A and B. We finished the answer key for Test A. Laura will finish the answer key for Test B.
Laura and Kelsey will plan on trying to run subjects in our CP3 pilot study, not next week, but the week after. The study will cross Test A vs B with Timed vs Untimed. Timed will be 3 min. per problem. Each experimenter will run one subject at a time in the untimed condition. Several experimenters can each run their one subject in the same room at the same time. The number of subjects that are typically run in the same room at the same time in the untimed condition will determine how many can be run in a session of timed subjects.
I will create dummy room locations called "OM 294 A - Laura", OM 294 A - Kelsey", etc.
It will be imperative that the researchers check to see if either Local/Global, or PR2 is running subjects in OM 294 A before creating the time slot. The system will think "OM 294 A - Whoever" is a different room from OM 294 A. Therefore, it will not stop you from double booking the room. It will be your responsibility to make sure you don't do that.
| previous | back to top | next |
At 2:00pm.
Present: Dr. Ryan, Kelly Breeze
Kelly will work on the calendar.
Also, Kelly to start coding data in PR2.
Present: Dr. Ryan, Wes, Chris.
They have made the Worked Examples-First Problem Pair for MF-ANAL/ANAL and MF-ANAL/MA. Except that: for the MF-ANAL/MA, the section about how you actually do the matching needs to be copied from the notes for Wed. 11-2-05, to the script. Also, the electronic files of the scripts are on a disk that Wes has. They need to be put in the folder for Materials for CP3 in the Ryan Lab Group folder.
The next thing that Wes, Chris, and Bob will work on is a the Worked Examples-Second Problem Pair for MF-ANAL/ANAL and MF-ANAL/MA.
This raises another question. Previously, we worked on the question of what kinds of problem pairs we should use in the training. That is, should we use all FIN/FIN, or should we use some FIN/FIN and some FIN/INIT. We ended up deciding to use all FIN/FIN. One reason for doing this is that we want to be consistent with what was done in CP2 (as described in the notes for 9-28-05). Also, we are using the nine test problems from CP2 which contain some find INIT problems. Thus, we will be testing whether comparing FIN/FIN in an MF condition versus an FC condition at least enables the MF subjects to do better than the FC subjects on a find INIT problem after being trained only on find FIN problems.
This means, however, that when we give our CP3 subjects the Worked Examples-Second Problem Pair for MF-ANAL/ANAL and MF-ANAL/MA, the problems will be of the SAME KIND as in the Worked Examples-First Problem Pair for MF-ANAL/ANAL and MF-ANAL/MA. We decided that it would be better to give them this second training with the same kind of worked examples, rather than cutting down on the amount of training in order to avoid that redundancy. The reason for that decision is that we want a very strong training manipulation in order to try to get the effects on the test problems in CP3 that we did not get in CP2.
Once they have finished the scripts for MF (both ANAL/ANAL and ANAL/MA) all the way through the two worked examples, the two guided practice examples, and all the unguided practice pairs, then they will start on the FC. But the FC will be easier because basically they will just be eliminating the Matching Features, and just asking for a free comparison. If any other questions arise when we progress that far, we will deal with them when they arise.
Present: Dr. Ryan, Jackie Maniet, Kelsey Kowalski, Mary Finn
Continued working on Anwer Key for Test B. Needs a little more editing.
Showed them how to log on to sona-systems and create time slots. Also, went to OM294A with Mary and Jackie (Kelsey needed to leave for work) to show them the room and how to set up the files for the experiment.
I'll need to prepare the consent, debrief, and script. They will bring a calculator for their subject. (I will check on the calculator supply and add some if necessary).
By Monday, they will create time slots for later next week and early the week after.
| previous | back to top | next |
Present: Dr. Ryan, Kelly Breeze
We checked on how to code the data for PR2 Face to Face. Kelly will work on that.
We checked the calendar. Kelly updated APS and EPA. However, she needs to redo APS because she went to www.aps.org, which is the American Physical Society site, instead of www.psychologicalscience.org. She will also make sure the entries for EPA are on the correct line. Somehow, some got on the line for APS.
Present: Dr. Ryan, Wes, Chris O'Donnell
First, we looked at the Worked Examples, Second Pair, for the MF-ANAL/ANAL condition. The A problem is a sales, find final ratio, wt. ave. problem (SFW). Wes had to write a new problem, the Tom buying tires and rims problem. Next, the B problem. That one will be a distance, find final ratio, wt. ave. problem (DFW). It can be the Bill and Hillary problem, but changed from find initial to find final. Those still need to be changed.
Second, we looked at the Worked Examples, Second Pair, for the MF-ANAL/MA condition. The A problem is the same A problem as above. For the MA problem, we need a Sales, find amt. saved, adding ratios (S--A). We still need the problem and the script for that.
Present: Dr. Ryan, Danzel Novic, Laura Kise, Andrew Malkiewicz, Jackie Maniet, Mary Kay Finn, Kelsy Kowalski
I forgot to check if the finished keys are on the Ryan Lab Group folder.
I showed them the script, consent, cover sheet, and debrief for the Pilot study. They all have access to the Ryan Lab Group folder to print them out. I discussed the folders that need to be set up in OM 294A for the study.
They have time slots created and subjects signed up. We kept the meeting short and they went to OM294A to set up the necessary materials to run subjects.
I spent a little time showing Danzel around Linux and gave her a Mepis Live CD.
| previous | back to top | next |
No meeting. Thanksgiving.
No meeting. Thanksgiving.
| previous | back to top | next |
No meeting.
Present: Dr. Ryan, Wes, Chris, Robert C.
Wes tried rewriting the script for the Bill and Hillary problem for the B problem of the second pair of worked examples in MF/ANAL-ANAL. However, he tried to add Bill and Hillary's distances. He needs to re-write the problem first, and then rewrite the script. He now understands that we need to be looking for Hillary's speed, with Bill's speeds for both parts of the trip given.
Chris has made nine MA problems. They are all sales/discount, find amount saved, add to discounts. One problem, the last one about glasses, refers to insurance paying 30%. It may be necessary to write that so the subject knows to treat that as a discount.
Wes, Chris, and Bob will continue with the necessary changes to both the scripts and the subject materials (which have not been worked on at all yet).
| previous | back to top |
No meeting due to Psych Dept. Meeting.
Present: Dr. Ryan, Wes, Chris, Robert C.
Wes did the following. The script for the Worked Examples, MF-ANAL/MA condition, Second pair, problem B, which includes the explanation of how to solve the problem and how to match it to problem A. That is now finished.
Wes will now continue with the script for the Unguided Practice, MF-ANAL/ANAL and MF-ANAL/MA, both problem pairs.
Chris will work on the subject materials. We decided that on the subject materials, the problems will be presented as worked out examples in STORY FORM, as well as with the steps of arithmetic.
I need to check back on why I changed from unworked examples to worked examples in CompProb2a. I need to see if the reason for the change is equally applicable to CP3.
OK, I checked. Actually, I changed from worked to unworked, not the other way around as stated above. And it was for the choice task, not the comparison task.
In the comparison task of CP2 (which was like the training part of DS2), the problems were always worked out, but in the form of a narrative. In DS2, the Worked Examples in the training were, of course, worked out, but in a step by step manner, not in the form of a narrative.
In CP3, in the MF condition, the Worked Examples in training should be as much like DS2 as possible. But in the FC condition, they should be as much like CP2 as possible. Yet, they should be presented the same way in both conditions. Therefore, in CP3, in both the MF and the FC condition, the Worked Examples in training should be presented both with the narrative and the steps. Once we get to the guided and unguided practice, however, the problems will be unworked, because part of the subjects' task at that point in the training is to work out the problems themselves.
Robert will begin working on the script for the FC (free comparison) condition. He will go in the same order that Wes went in. He will be using the work that Wes already did, and just altering it for the FC condition.
Regarding the FC condition. We noticed that in CP2 the instructions for the free comparison referred to one training pair, called the target and comparison problems, rather than Problems A and B. In CP3, there will be multiple pairs. And we will simply call them pairs, or problems A and B, not Target and Comparison.
Furthermore, in CP2 the explanations for how to compare (which Robert will use as the model for the FC condition in CP3), spoke of using the comparison process to find similarities between the members of the pair that would be useful for trying to solve the test problems to come. In CP3, the test to come is the posttest, and the subjects will have already taken a pretest. Nevertheless, the CP3 instructions should similarly tell the subjects to look for the useful similarities. Those instruction will have to appear in the script for both the MF, as well as the FC, condition. Therefore, Wes and Robert will have to work together as Robert alters the MF script to make it the FC script. As he is doing that, Wes will also have to be integrating the "look for similarities" idea into the MF script. It will occur right at the point where the first training pair (worked examples, first pair) is introduced.