5.4. Neurosynth meta‐analyses

The main ALE analysis was based on a carefully selected group of studies and used a strict statistical threshold to control for false positives. Accordingly, it likely captures true regions underlying motivated cognitive control. On the other hand, it may overlook some regions that are below threshold, but nevertheless play a role in functions that contribute to motivated cognitive control. Thus, we also utilized an alternative, complementary approach: Neurosynth forward inference meta‐analyses (Yarkoni, Poldrack, Nichols, Van Essen, & Wager, 2011). Neurosynth makes it possible to examine consistent activation patterns across a massive database of studies, providing strong power to detect regions that are activated in the studies of interest. In this way, Neurosynth is less likely to miss relevant regions. However, Neurosynth performs an automated selection of studies based on certain key words (e.g., “cognitive control”) and therefore has less specificity in terms of delineating a select group of studies. In this way, Neurosynth analyses provide a complementary approach to ALE analyses. To perform such automated meta‐analyses, Neurosynth divides the entire database of coordinates into two sets: those that occur in articles containing a particular term and those that do not. A large‐scale meta‐analysis is then performed comparing the coordinates reported for studies with and without the term of interest. Forward inference maps reflect z‐scores corresponding to the likelihood that each voxel will activate if a study uses a particular term (P[Activation|Term]) and are corrected for multiple comparisons using a false discovery rate (FDR) of q = .01. Here, we conducted forward inference meta‐analyses using the terms “cognitive control” and “incentive” and looked for brain areas demonstrating overlapping recruitment across both domains. We reasoned that if a brain region is activated in studies of cognitive control and is activated in studies of incentive processing, then it is a good candidate for bridging cognitive and motivational functions. We used forward inference rather than reverse inference analyses because we were not looking for regions that are selective to incentive processing or cognitive control, but rather, are just involved in these domains. In other words, we were not looking to identify functional specialization in any regions. Our aim was to identify the constellation of regions that together support motivated cognitive control. Given this aim, forward inference analyses were ideal.