Is there a way for Drag & Drop answers to populate next screen?

Question

I'm working on a project in Captivate 7 where the user is filling out a table (1 row, 3 columns, 9 possible correct answer combinations) using drag and drop words. The user will ultimately build a table with 3 rows and 3 columns, doing 1 row at a time per screen.  I would like the user's correct answers to the rows they have already completed correctly to appear on each successive screen (but not for the entire rest of the project, just the following few screens). I have tried to get this working every which way and I can't seem to make it happen.
The closest I have gotten is to use the "accept" panel to set each correct answer to assign its own value to a variable placed in a text caption on the following slide. However, that seems to trigger an auto-submit (I have the global "auto submit correct answers" turned off) so that the first item the user places correctly in the table advances them to the next slide even though there are still 2 more cells of the table to complete.

Answer 1

Here's a shot of the intiial screen:
As you can see, there are multiple correct combinations of answers. (Users will be children, so it does need to be very user-friendly with the ability to undo/reset answers.) In the word bank are 3 correct variable items and 6 correct values (I have it set so that only correct combinations of variables and values will be accepted as correct, so "lemonade" will never be a correct value for "age of seller", for example).
Here's the next screen which I have made fill from selections from previous via an "on accept" advanced action (show a text caption unique to that answer and location):
The problem with this advanced action solution is that if the user makes a selection in the first screen, then gets it partially wrong or for some other reason resets the words back to the bank, those text captions on the next screen still have been shown and will stick around, thus:
Every solution I've attempted has had some fatal flaw in the end. What I really need to happen is a conditional action that basically says "If the contents of drop target X = drag source Y, then set VarA [on screen 2] to $text$." For example, "If the contents of drop target [Variables Cell] = Text Caption timeofday, then set $Var1$ to "Time of Day".
I am about to throw in the towel here and just have students type in words to text boxes (these are mainly disadvantaged students I work with, so their typing skills are not great) just to utilize being able to make a user-generated variable out of text box entries.