Desimone also points out that fMRI data is limited because they're based on oxygen levels in the blood, instead of actual neural activity. This means that the data isn't very detailed; it’s an indirect measure of the activity of a very large number of neurons, says Detlef Wegener, a neuroscientist at the University of Bremen in Germany who didn't work on this study. Still, the work does hint that the human attention networks are characterised by particularities that "can't be fully deduced from monkey data", Wegener says.