Investigation of reward effects in overlapping dual-task situations

In dual-task (DT) situations, performance in reaction time and error rates decrease compared with single-task situations. These performance decrements are usually explained with the serial processing at the response selection stage constituting a bottleneck. Evidence for this assumption stems from the observation that response times for the second task (task 2; RT 2) increase with decreasing stimulus-onset asynchrony (SOA). In this study, we investigated the effect of reward on bottleneck processing in DTs. In Experiment 1, we addressed two questions. First, does reward provided for task 2 performance affect task 2 performance, or does it affect task 1 performance? To conclude whether reward affected task 2 or task 1 performance, we relied on the psychological refractory period paradigm (PRP) as a chronometric tool. Second, we asked for the locus of the reward effect within the DT stream. We demonstrated shorter RTs in task 1 in a rewarded compared with an un-rewarded condition indicating reward affected task 1 processing. Furthermore, this reward effect is propagated onto task 2 at short SOA suggesting that the locus of the reward effect can be pinpointed before or at the bottleneck of task 1. In Experiment 2, we tested for the locus of the effect propagation onto task 2. To this end, we implemented an additional difficulty manipulation of the response selection of task 2 and found that the reward effect is propagated from task 1 onto the response selection stage of task 2.