OK had a bit of time to look on the internet, I could not find bugger all proper tests, lots of anecdotes on forums, you would think there was more data
thus 2 possible results
A/ the written conclusion is wrong
B/ they stuck the wrong graph in the paper
vs the graph, which seems to show the expert listeners not being able to tell the differenceConclusion
Trained listeners can not discriminate between CD quality and mp3 compression (256-320 kb/s) while expert listeners could.
thus 2 possible results
A/ the written conclusion is wrong
B/ they stuck the wrong graph in the paper