I actually still think of katakana and hiragana primarily as random strokes.  I don't really associate them with the kanji they came from, with the exception of の's relationship to 乃--and that's mostly because a lot of place names use the kanji 乃 with the same meaning as the possessive particle の.  The only other minor exception might be チ, because it looks so close to the original 千.

I intellectually am aware of the roots of them, and if you asked me off the top of my head to produce the original kanji I could probably do it for about half of them... but it just doesn't transfer over into an everyday mental association.

yudantaiteki wrote:

Unless you're reading older manuscripts that use hentai-gana and alternate cursive forms of the standard kana, I don't know if it's incredibly useful as far as modern Japanese goes.

Don't get me wrong, I'm not arguing that this list should be anything for anyone to study vigorously. I just think it's crazy seeing how katakana were derived from kanji, especially the obscure ones :p