I can see how it may seem counter-intuitive, but there's a good reason for setting it up that way: the articulation data is actually embedded within the note itself. That means the articulation setting on the plug-in GUI isn't a playback mode for already recorded notes (they already have their own articulation that was determined when you recorded them), it's a setting that determines the articulation of the notes you're recording.
This allows for various notes of different articulations at the same time. For example let's say you just recorded some sustained notes and want to do a second pass, recording a few staccato notes on top of the sustained notes, you can do that because the articulation setting in the plug-in doesn't affect the notes you've already recorded, only the new ones you'll be recording now.
that does make sense!