loss; −1 for neutral), an A – B regressor of interest. This term was convolved with the HRF (Friston et al., 1997). Contrast vectors were also created for the A + B regressor (1 for reward or loss; 1 for neutral), as well as the condition trial not under study, a C regressor (ie., 1 for loss, 0 for reward, 0 for neutral when evaluating for reward connectivity change), to account for all task associated variance. The PPI model regressors consisted of the interaction term, the A-B contrast vector, the extracted time-series, the A+B regressor, the C regressor, and motion regressors from the original design (O’Reilly et al., 2012). Single subject contrasts for the first regressor (interaction term) were calculated which identified regions whose task-driven connectivity with the NAcc seed changes as a function of incentive anticipation, hereafter referred to as connectivity.
