“…A central assumption is that the representations used for effect-based control of both simple and complex actions are organized in a hierarchical fashion, with higher level (distal) action goals at the top, and the required sequence of more proximal, body-related effects at the bottom, each robustly linked to the efferent/motor activities that would bring these effects about (e.g., Hommel et al, 2001, Hommel, 2009; for a recent review, see Moeller & Pfister, 2022). During execution, imaginistic control can then access-and modify-each of these more proximal action components to fulfill task requirements, such as making a less forceful button press than usual (Cao et al, 2020), varying the musical notes one is playing in a longer piece (e.g., Keller & Koch, 2008), or forming a more pronounced circular trajectory than usual during gesturing (e.g., Bach et al, 2010a). Moreover, during learning more complex action hierarchies can be built form the ground up from simpler elements and assembled into larger chunks (e.g., Moeller & Pfister, 2019).…”