Repetition Benefit In Task Switching:A ERP Study

In daily life, people often need to switch between different task, this flexibile in switching is controlled by the the brain,we call this execution control or congnitive control. Task-switching paradigm is a common way to the study of execution control, and the switch cost is an important indicator to evaluate the implementation control. Now most researchers have accepted that task-set reconfiguration is a primary cause of switch cost, but because of the source of switch cost is diversity, many other factors are still controversial.In this paper, we put different response-stimulus interval(RSI), cue-stimulus interval(CSI) and long sequences of repeat trials in an task-cuing switching paradigm with bilateral stimulation, as our paradigm. We focused on the long sequences of repeat trials effect in the task-switching paradigm, and its corresponding neural mechanisms. We simultaneously recorded the behavioral and eeg data of each participant, employed matlab and SPSS, Combined with behavioral, ERP and s LORETA analysis technology, and completed all the processing of data analysis.The behavior results showed that under the condition of short CSI, with the increase of repetitions, remarkably reduced reaction time, the pairwise contrasts for adjacent tasks repetitions were significant for the switching task VS the task repetition 1 and for task repetition 2 VS task repetition 3. But under the condition of long CSI, in a addition to the significant difference between the switching task VS the task repetition 1, there is nothing else difference. The results suggested that under the condition of short CSI, long sequences of repeat trials was also a cause of switch cost. The behavior results also found that the reaction time of long sequences of repeat trials decreased with the increase of CSI, and the compare of conflict response consistency found that the inconsistent response was significantly slower and lower accuracy than consistent response. The results indirectly indicated that the proactive interference existed in the task switching, and it would affect the completion of current task while it was unrelated. But under the condition of long CSI, there was engough time to selective inhibit the interference and improve behavioral responses. So the results supported the task-set inertia theory.Cue-related ERP showed that under the condition of short CSI, the amplitude of P300 of FCz and Cz decreased with the increase of repetitions, the pairwise contrasts for adjacent tasks repetitions were significant for task repetition 1 VS task repetition 2 and for task repetition 4 VS task repetition 5. And under the condition of long CSI, we had not found ERP components related to the long sequences of repeat trials. Stimulusrelated ERP showed that under the condition of short CSI, the P200 of FCz and Cz is similar to the cue-related P300, with the increase of repetitions,the amplitude decreased. The correlation analysis between P300 and P200 found the correlation coefficient is 0.99 or 1. So we believed that under the condition of short CSI, the stimulus-related P2 indicated the continuation of unfinished preparation activities. And under the condition of long CSI, we had not found ERP components related to the long sequences of repeat trials and the continuation of unfinished preparation activities. The results supported the task-set reconfiguration theory. Correlation analysis of ERP and behavior found ERP and behavior were relevant.sLORETA analysis found Inferior Frontal Gyrus, Middle Frontal Gyrus, Superior Frontal Gyrus, Precentral Gyrus and Postcentral Gyrus showed adaptation to the long sequences of repeat trials, and this suggested that in long sequences of repeat trials, the improvement of behavior and the decrease of the ERP amplitude were due to the adaptation process of abstract rule representation in these areas, but the adaptation process didn’t happen under the condition of long CSI.