On the temporal dynamics of reward utilization in dual-task situations.

In dual-task (DT) situations, performance typically deteriorates compared with single-tasking situations. These decrements can be explained by the serial scheduling of response selection stages constituting a central bottleneck as with decreasing stimulus-onset asynchrony (SOA) the reaction time for the second task (Task 2; RT2) increases. Prior studies indicated that the reaction time for the first task (Task 1; RT1) and RT2 are improved in reward compared with no-reward conditions for a block-wise reward prospect, which reflects reward-related optimization in DT processing. However, it remains unclear whether participants can flexibly utilize reward information in a trial-by-trial manner to achieve reward-related improvements. Additionally, it is unclear whether a potential reward-related optimization reflects optimized task preparation only or whether the prospect of reward can evoke an additional task optimization mechanism that extends beyond preparation-related processing improvements. For Experiment 1, we combined a trial-wise reward prospect for participants' Task 1 performance, which was signaled by a cue before Task 1 onset, with block-wise presented cue-target intervals (CTI) of either 200 ms or 700 ms, resulting in precise temporal predictability of Task 1 onset by participants. First, we observed a reduced RT1 in the reward compared with the no-reward condition. Furthermore, the reward effects increased on RT2 for short compared with long SOAs, reflecting effect propagation at short SOA from Task 1 onto Task 2. Second, RTs decreased with increasing CTI, while reward effects increased with increasing CTI. Consequently, preparation-related processing improvements of DT performance were additionally improved by reward utilization. For Experiment 2, temporal predictability of Task 1 onset was reduced compared with Experiment 1 by presenting CTIs randomized within blocks, which allowed replicating the result pattern of Experiment 1. Across both experiments, the results indicate that participants can flexibly utilize reward information in a trial-by-trial manner and that reward utilization additionally improves preparation-related processing improvements for DT conditions with predictable and less predictable Task 1 onset.