The long test is painfully long.
Quote:
You have now reached the end of the MERT and your test scores have been computed. Below you see two columns of accuracy scores, indicating the percentage of expression items in different modalities that were correctly judged. The left column contains your scores, the right column contains the corresponding values for the MERT validation sample (72 psychology students at the University of Geneva.)
YOUR accuracy scores (in %)
Still Face only--------57
Voice only-------------43
Video Face only-----67
Audio-Video----------57
Comparison Group Mean accuracy scores (in %)
Still Face only--------56
Voice only-------------49
Video Face only-----69
Audio-Video----------70
Interesting that controls do better with more information, and I don't....probably because I can only pay attention to one part of the nonverbals at a time no matter how many I'm presented with.
It would be a more challenging test if they didn't just recycle the same limited clips in bits and pieces -- if you can recognize the voice + string of nonsense words and associate them with the visuals they go with (or vice versa), then it's easier than if you had to figure it out from only one source of nonverbal information...plus you get multiple tries at the exact same one and thereby increase your chances of getting the right answer on later questions.
_________________
"Coming back to where you started is not the same as never leaving." -- Terry Pratchett, A Hat Full of Sky
Love transcends all.