| ObjectiveThe function of central executive system function include election, planning, switching, extraction, checking and so on. The neural mechanism of task switching remain to be clarified. It is an often replicated finding that switching between tasks is associated with longer latencies and higher error rates, which is known as the task-switch cost (Monsell,2003; Rogers,1995). A majority of theories explaining the task-switch cost are at least partly based on the observation that increasing the preparation time with advance knowledge of the upcoming task results in faster response latencies in general and in the reduction of the difference between task repetitions and task alternations. But at least as important for the theorizing about task switching is the finding that although advance task preparation reduces the switch cost, it never really eliminates the switch cost (Rogers,1995; Goffaux,2006; Monsell,2006; Sohn,2000), suggesting that advance task preparation, and endogenous control in general, is restricted in nature. This remaining residual switch cost received much attention in the task-switching literature and will be the focus of the present study. However, some researches suggested that residual switch cost could be eliminated just by short-time presentation for the task cue (Mayr,2001; Verbruggen,2007). But No study has been carried out on the neural mechanism of this phenomenon.Our research used a Chinese Stroop switching task paradigm. We aimed to investigate the neural mechanism of task switching cost in the task preparation phase and task execution stage by changing the cue presentation time. MethodsTwenty-two college students participated in the experiment. Data of3subjects was removed because of large artifacts. The rest19(10females) subjects aged between21and29years old (24.85±2.29years). All is right-handed without neurological or mental disorders.Our experiment used the Chinese Stroop switching task paradigm. The Chinese character "red" and "blue" were printed in red or blue color. Then generated four target stimulus:the red "red", the blue "red", the red "blue" and the blue "blue". All participants pressed the left key for "red" or red color and pressed right key for "blue" or blue color. The task cues were the Chinese character "word color" or "word meaning", which would be presented100/1000ms. The cue of "word color" indicated to response to the color of the word, while the "word meaning" indicated to the word meaning. The target stimulus presented1000ms. The interval between the cue and the target was1000ms. However, the interval between two cues was2500ms. The repeat trail involved the second cue when it was the same to the first trail. In the same way, the switch trail involved the second cue that was different. The repeat and switch trails emerged randomly. The repeat trails were controlled from2to4. There was no consecutive switch trail. The ratio between the total repeat (352trails) and the switch (248trails) was7to5.EEG was recorded using an ERP system developed in our laboratory. Event-related EEG epochs included two parts. The first included pre-stimulus activity of100ms and post-stimulus activity of900ms for the cues. While the second part came from-100ms before to1400ms after the onset of Stroop stimulus presentation.A two-way ANOVA was carried for behavioral performance and ERP data (the cue presentation time:100ms,1000ms)×(the type of succeeding task:repeat, switch).Statistical parametric mapping F-values [SPM(F)] was gained from interpolation calculated by each channel’s F-values. The significant level was0.05.ResultsBehavior performance The reaction time (RT, ms):the cue effect, the task effect and the interaction are not significant. While in the long cue presentation time, there is significant tendency (t(18)=-1.97, P=0.07) between the task switch group (620.8±107.8ms) and the task repeat group(605.7±104.4ms).The reaction time cost (RTC, ms, the switch subtracts the repeat):The difference between the long cue group (15.0±33.3ms) and the short group (0.80±40.7ms) had the significantly tendency:t(18)=-1.77, P=0.09.The accuracy(A,%):Both the interaction effect and the task effect are not significant, but not for the cue effect(F(1,18)=0.56, P=0.02). In the long cue presentation time, there is significant tendency (t(18)=2.35, P=0.04) between the task switch group [(90.8±6.8%)] and the task repeat group[(92.9±4.9%)]. In the short cue presentation time, there is significant tendency (t(18)=2.26, P=0.03) between the task switch group [(92.0±5.5%)] and the task repeat group[(93.5±4.6%)]The accuracy cost(AT,%, the repeat subtracts the switch):The difference between the long cue group (2.2±0.9%) and the short group (1.5±0.6ms) is not significant(t(18)=0.96, P=0.35).Spatiotemporal patterns of SPM (F)The cue processing phase:Significant effects of cue occurred in the frontal-occipital areas (100-200ms), frontoparietal and tempo-occipital regions (200-500ms), frontoparietal network and centro-occipital regions (500-900ms). There was no significant interaction effect of cue and task. In the same way, there was also no significant ERP effects of task.The task execution phase:The ERP effects of the cue were observed from the left fronto-occipital areas (300-400ms) to parieto-occipital areas (400-900ms). The ERP effects of the task started from the right hemisphere (500-600ms) and extended to bilateral prefrontal and centro-parietal area (600-900ms). There were also no significant interaction ERP effects between the cue and the task.ConclusionBehavior performance:In line with Verbruggen’s (2007) findings, the RTC results have significant tendency to disappear. While the AC has no significant difference. In the majority literatures, the switch cost points out RTC, having no relevant research about AC.The task preparation stage:Our experiment reveals the time change of cue presentation can significantly influence many encephalic regions. The early ERP differences of cue revealing in the frontal-occipital areas (100-200ms) and the P3b in frontoparietal and tempo-occipital regions (200-500ms) reflect the accentuation and prime of short cue processing. While the late EPR effects of persistent frontoparietal network (400-1000ms) reflects the short cue mobilizing more WM resources. Accordingly, we can conclude that the frontal cortex has an important regulation in task preparation and task switch. The parietal cortex may be involved in information storage and stimulus-response mapping.The task executive stage:(a) the cue effect:In our experiments, the ERP effects of the cue observed in the left fronto-occipital areas (300-400ms) may be reflect the extraction and control of speech task cue. While the persistent and lasting EPR effects in the parieto-occipital areas (400-900ms), which is called NSW, reveal the excitability heightened of sensory area, the retention of persistent attention and the effective storage of task cue.(b) the task effect:Our experiment observed the ERP effects of the task started from the right hemisphere (500-600ms) and extended to bilateral prefrontal and centro-parietal area (600-900ms), revealing the switch trials needing more executive function than repeat trials and the regulating function of central executive system (Sohn,2000; Collette,2002).In short, we can see that the short cue presentation time can reduce the RTC. It can not only early prime the preparation of frontoparietal network in the cue preparation stage, but also can enhance the left frontal to extract and use the cue in the task executive stage. |
PDF Full Text Request