The four dropdowns near the bottom of the "Audio" tab of the VoiceAttack options window are essentially shortcuts to your Windows sound device settings. If you use them to change devices, you're changing the devices Windows itself, and most applications, will use.
The vast majority of applications will use the Windows default multimedia devices for playback and recording.
The Windows default communications devices, which can be identical to the multimedia devices, are theoretically intended mainly for something like VOIP (Voice Over Internet Protocol, basically phone calls over the internet), so a rare few applications may use those devices instead (neither VoiceAttack nor the Microsoft speech recognition system do).
If you want to use a device other than the default Windows multimedia playback device for audio generated by VoiceAttack commands, you'd want to use the "Override Default Playback Device" dropdown near the top of the "Audio" tab to change which device sound files are played back to, and the "Override Default Text-to-Speech Device" dropdown next to that to change which device synthesized text-to-speech is played back to.
Do make sure that the "Audio Output Type" option is set to "Integrated Components" (which should be the default).