This kind of afterthought-based correction eventually trickles into the first-thought reasoning to some extent, because it alters the incentive structure (you learn not to say things that you’ll just end up correcting). So, it may be more useful than it sounds.
A further comment:
This kind of afterthought-based correction eventually trickles into the first-thought reasoning to some extent, because it alters the incentive structure (you learn not to say things that you’ll just end up correcting). So, it may be more useful than it sounds.