Covary

More specifically, to say that there is an informative criterion of diachronic identity is to say that facts of diachronic identity covary with facts about continuants’ instantaneous qualitative profiles […].

The participants certainly differ in how their practice is distributed (1, 2, or 3 days), but they also differ in how much total practice they get (3, 6, or 9 hours). This is a perfect example of a confound—it is impossible to tell if the results are due to one factor (distribution of practice) or the other (total practice hours); the two factors covary perfectly.