Search

Switching to a dominant task incurs larger costs than switching to a non-dominant task. This study investigated whether this cost asymmetry derives from the inhibition of the dominant task rule that occurred during the previous trial. Participants were presented with a five-letter array consisting of (left) and (right), and asked to respond to the central target letter after being informed about the task rule with a pre-cue. The rule was switched between dominant (left-hand to, right-hand to) and non-dominant tasks (right-hand to, left-hand to) every two trials. Reaction times revealed the asymmetrical switch cost and the effect of target-flanker congruency. Stimulus-locked lateralized readiness potentials showed that the dominant task rule was inhibited during the non-dominant task, whereas this inhibition was not carried over to the dominant task trial.

Event-related potential (ERP) and reaction time (RT) were recorded to investigate the time course of processes involved in set switching. The cued set-switching paradigm required participants to switch stimulus task sets between male and female face-images memorized as targets prior to a trial block and response task sets between two stimulus-response mappings for each stimulus task. Replicating previous findings, an RT switch-cost was found when compared with set-repeat trials. The RT was also prolonged for a stimulus task requirement of memory comparison with two-face targets rather than one face. A similar prolongation with memory comparison was observed in P3b latency, which showed no switch effect. The switch effect was observed for the onset latency of stimulus-locked lateralized readiness potential (LRP), measured as an index of commencement of motor processes after response selection. The response-locked LRP indicated that the final process of motor execution itself was not modified by set switching. The processes producing the stimulus-locked LRP switch cost, associated with response task set, were discussed in terms of two hypotheses, exogenous reconfiguration and carryover.