Commentary: The Role of the Parietal Cortex in the Representation of Task–Reward Associations

Broadly linked to domains such as attention (Corbetta and Shulman, 2002), episodic memory (Cabeza et al., 2008) and language (Price, 2010), profiling of the heterogeneous inferior parietal cortex (IPC) is challenging. Here we comment on a recent IPC paper by Wisniewski et al. (2015). Briefly, we believe that their main findings may be parsimoniously explained by existent theories on IPC, whereas another aspect of their study deserves more attention. Wisniewski and colleagues explored the brain areas underlying associations between tasks (parity/magnitude) and rewards (high/low). The information required for task execution gradually unfolded across a trial: Participants subsequently received a symbolic task-reward association cue (2-s) followed by a 6-s delay (phase 1), a fully informative task cue (1.5-s) followed by a 1.5s delay (phase 2), and the target stimulus followed by some time to respond (phase 3)1. Finally, successful performance was rewarded accordingly. Two major observations resulted from fMRIbased multi-voxel pattern analysis. First, during phase 1 the bilateral IPC coded for the instructed task-reward associations. Second, IPC coding was adaptive as IPC later in the trial also coded for reward prospect (∼phase 2/3) and (in left IPC) task-to-be-performed (∼phase 3). This exposes IPC as a critical link between cognitive control and motivational functions in the brain. Yet, below we critically discuss how these findings contribute to our knowledge about IPC. Phase 1 required the symbolically cued task-reward associations to be maintained for 6-s until task cue presentation. Robust coding was observed in bilateral IPC, covering parts of both the supramarginal gyrus (SMG) and the angular gyrus (AG). The current literature links these areas to semantic and phonological processing; for example, left SMG supports verbal working memory (Deschamps et al., 2014) including phonological maintenance (Sliwinska et al., 2012), and AG underlies semantic processing and the construction of meaning (Seghier, 2013). Indeed, right IPC is also often observed in working memory tasks (Owen et al., 2005)—possibly linked to spatial rehearsal (Smith and Jonides, 1998). With a symbolic task-reward association cue requiring maintenance (either in visuo-spatial or verbal format) and (at some point) semantic translation, the bilateral IPC coding during phase 1 may be directly linked to well-known roles in the context of a (verbal) working memory task.