I'm struggling to explain myself properly with this.
I guess all I was trying to say is that if I've developed the film for too long then information that should've fallen at a mid-tone value may have been shifted up in density value as the curve became steeper/gamma or contrast index increased - In the end I have lost mid-tone detail. The lower gradient that should have been produced by correct development contains the ability to capture a wider spread of exposure values in the mid-tones.
When I develop Delta 400 too far I get the same effect - I always stop short of the recommended times now.
My roll of film has a variety of under and over exposed frames - I bracketed quite a few by a stop either way. All frames require a grade 0 or grade 1/2 filter to print the mid-tones (either dark when under-exposed or light when over-exposed), and after reading the attachment about variable contrast paper in john_s's reply I'd really like to get back up to grade 2 or thereabouts.
I work a lot with response curves digitally (both in photography for exposure and in audio for dynamic compression) and don't disagree with anything I've read in the replies so far - but due to unclear communication seem to have dug a hole for myself!!
Thanks for your patience.