Recently I spoke about overcorrection, a linguistic process through which the desire to avoid some mistakes in one's own language or a second language leads one to make others.
There's a correlative to that, but I don't know the name for it, or even whether it's an identified phenomenon. Lets look again at Korean. In Korean, the same 'p' sound as in English exists, as well as a sound reasonably close to English 'j' and the 'l' sound. There are no sounds close to 'f' or 'z' or 'r', and these sounds thus become associated with English. Sounds present in Korean come to be considered Korean and thus people tend to avoid them in speaking English even when they are correct.
That's the root of overcorrection. The correlative of this is that English sounds like 'f' and 'z' and 'r' get associated with one another. People expect a word to contain all 'English' sounds or all Korean sounds and never both. Thus in words with 'p' and 'z' or 'f' and 'j' together, you find people choosing either the two 'Korean' sounds, 'p' and 'j', or the two 'English' sounds, 'f' and 'z'. So you find people pronouncing the word 'puzzle' as either 'fuzzle' or 'pudgel'. typically if someone can pronounce the 'English' sounds, they will tend to use all of them together, thus 'fuzzle'.
This problem also occurs with other pairs of Korean ad English sounds, for example English 'r' and Korean 'l', English 'v' and Korean 'b', and English 'th' and Korean 's'. Here are some examples of this phenomenon:
'fly' > 'fry'
rubber > 'ruvver'
'Jeff' > 'Zeff'
'zipper' > 'ziffer'
'really'> 'reary' or 'leally'
'beef' > 'veef'
There are plenty more that you can hear often among people of a certain level of English competence. I am guilty of this one as well. When I was first learning Korean I had a hard time with the 'double consonants' (ㅃ, ㄸ, ㅆ,ㄲ, ㅉ). When I finally managed to start making the sounds, I was only able to do so if I made every sound in the word a double consonant. Thus 가짜 became 까짜, 바쁘다 became 빠쁘다 and 가깝다 became 까깝다.