FIGURE 2.

Schematic depiction of the general linear model to analyze the Pavlovian-instrumental transfer (PIT) data (Figure after Talmi et al., 2008; Geurts et al., 2013b). The main regressors (M) model the onset of a trial as a delta function. There is a main regressor for each of the six trial types. For all six main regressors, there are two parametric modulators (PM). The first parametric modulator (PM¹), the PIT regressor, consists of the number of button presses made per trial (0 for no-go). In the 7th main regressor (of no interest), every single button press is modeled by a delta function. For reasons of clarity, only two of the six trial types (approach neutral and withdrawal aversive) are depicted. The regressors of no interest are not shown (i.e., the movement nuisance regressors and the second parametric modulator).