Everyday task sequences, such as cooking, contain overarching goals (completing the meal), sub-goals (preparing the vegetables), and motor actions (chopping). Such tasks can be considered hierarchical as superordinate levels (e.g., goals) affect performance at subordinate levels (e.g., sub-goals and motor actions). While we execute these hierarchically organized task sequences routinely, and the effects of practice on motor actions are well known, the specific effects of practice and the presence of simultaneous motor sequences on such task sequences remain unknown. To investigate these questions, we adapted a behavioral paradigm developed to investigate the hierarchical control of abstract, or non-motor, task sequences (Desrochers, Chatham, & Badre, 2015; Schneider & Logan, 2006). Using this paradigm, we dissociated effects at the level of the sequential goal, or abstract task sequence, at the level of the sub-goal or individual tasks, and at the motor sequence level. In three experiments, participants performed memorized abstract task sequences composed of simple tasks (e.g., color and shape judgements) that were practiced, novel, or contained embedded motor sequences. We found that abstract task sequence practice selectively reduced sequence initiation control costs, and that adding motor sequences further facilitated performance. Embedded motor sequences also reduced task level control costs at task switches. Manipulating which abstract task sequences were associated with embedded motor sequences provided support that the effects of embedded motor sequences are additive, and not the result of an integrated representation produced by practice. Together these experiments provide insight into the distinct roles of practice and motor processing in efficiently executing task sequences in daily life.

Abstract sequential task control is facilitated by practice and embedded motor sequences