This showed up on my Facebook feed so I took the bait. However, after reading the actual study here:
you'll see that their methods were essentially BS to give the results they wanted.
Musical instrument sounds have distinct timbral and emotional characteristics that can change when audio processing is applied. This paper investigates the effects of MP3 compression on the emotional characteristics of eight sustained instrument sounds using listening tests. The experimental paradigm involved a pairwise comparison of compressed and uncompressed samples at several bit rates over ten emotional categories. The results showed that MP3 compression strengthened neutral and negative emotional characteristics such as Mysterious, Shy, Scary, and Sad, and weakened positive emotional characteristics such as Happy, Heroic, Romantic, Comic, and Calm. Angry was relatively unaffected by MP3 compression, probably because the background “growl” artifacts added by MP3 compression decreased positive emotional characteristics and increased negative characteristics such as Mysterious and Scary. Compression effected some instruments more and others less; trumpet was the most effected and the horn the least.
However, they only tested up to 112K bitrate. See table 2: http://www.aes.org/e-lib/download.cfm/18523.pdf?ID=18523
Full disclosure, I don't care for the MP3 format mainly due to how it can't do true gapless playback. You'll only notice this if continuous music is split into tracks. While I listen to flac format, I'd be hard pressed to tell the difference between a 256 to 320 Kbs MP3 and lossless in most cases. When you get into 198 and lower MP3 territory, I can start hearing discrepancies in the upper range such as cymbals. Low volume sections of music are also handled poorly in lower rate MP3. The worst offender I've heard was The Door's Riders on the Storm. At the beginning where you should be able to hear a bit of low tape hiss that's present on the CD or Vinyl, you hear the lower bitrate MP3 try to suppress the hiss and the result is a glassy-sounding, weak organ. A 256+ MP3 won't do this if the encoding was done well.
In my opinion, this study didn't go far enough in their testing. The bitrates they used should've been the controls as they knew going into this that instruments would sound bad. 128, 156,198, 256, and 320 should've also been included. I can almost guarantee that at some point the emotional response to the instruments change. Also, lossless encoded at lower bitrates should give the same result. Poor encoding is poor encoding and it's not inherent to any certain format.