Fig. 2. Baseline visual-motor connectivity predicts future learning rate.

(a) Visual module (yellow) and somato-motor module (purple), identified by time-resolved clustering methods applied to BOLD activity acquired during execution of motor sequences (Bassett et al., 2015). The modules were defined in a data-driven manner and correspond broadly to putative visual and somato-motor modules. (b) Functional connectivity between visual and somato-motor modules, estimated at rest and prior to learning, reliably predicts individual differences in future learning rate. We define the learning rate as the exponential drop-off parameter of the participant's movement time as a function of trials practiced, and we define functional connectivity as the average value of the correlation coefficient between activity in visual regions and activity in somato-motor regions. Note that we use the term “prediction” to imply that the value of one variable (at one point in time) can be used to predict the value of another variable (at a later point in time), without implying the use of out-of-sample generalization (Gabrieli et al., 2015).