The problem was in the incompleteness of AI2's Duck Typing
(If it walks like a duck, talks like a duck, it must be a duck.)
Text and logical 'true' don't compare as equal.
Also, the green logical '=' block is more list friendly than the math
'=' block, according to its tool tips.
So I replaced the text answers with logic answers, and the
list comparison block too.
While I was in there, I also
renamed your procedure appropriately,
and cleaned up your limit response in the Next
button to avoid a subscript error.
See attached.
ABG