As it turns out, predictability isn’t all that great. The original Libet study was done in such a way that it wasn’t possible to generate a number for this. However, in the Haynes studies, fMRI images predicted which behavior occurred with only about 60 percent accuracy, almost at the chance level. For Mele, a “60-percent accuracy rate in predicting which button a participant will press next doesn’t seem to be much of a threat to free will.”
