Concurrent Response Selection in Dual-Task Performance: Evidence for AdaptiveExecutive Control of Task Scheduling

摘要

Four experiments with the psychological refractory period (PRP) procedure arereported that investigate how people perform multiple tasks concurrently. In each experiment, a primary task was paired with a secondary task that had two levels of response-selection difficulty. Experiments 1 and 2 varied response-selection difficulty by manipulating the number of alternative stimulus-response (S-R) pairs in the secondary task. In both experiments, the effect of this factor on secondary-task reaction times (RTs) decreased reliably as the stimulus onset asynchrony (SOA) decreased. Experiments 3 and 4 varied response-selection difficulty by manipulating S-R compatibility for the secondary task. Again, the effect of this factor on secondary-task RTs decreased reliably as SOA decreased, regardless of whether or not the primary and secondary tasks involved the same response modality. Taken together, these results raise doubts about the existence of an immutable structural central bottleneck in response selection. Rather, it appears that response-selection processes for two concurrent tasks may temporally overlap. This outcome is consistent with dual-task performance models (Meyer Kieras, 1997a, 1997b, 1997c Meyer et al., 1995) under which people have adaptive executive control of their task-scheduling strategies.