Abstract: Working memory (WM) and cognitive flexibility are the two major components of executive function. Does working memory affect task shifting? This issue is still controversial. Some researchers had found that the switching cost was not affected by working memory, so working memory was not associated with cognitive flexibility. However, other researches had shown that working memory was closely related to task shifting and resource constraint of working memory was one of the main sources of switching cost. So far, most studies on the relationship between working memory and task switching have adopted the dual-task paradigm. The main reason of the contradictory results was that different investigators manipulated different components of working memory in the experiment. If the researchers increased the load on the central executive component of working memory in the experiment, it would have a significant impact on task switching.?On?the?contrary,?task?switching?cost?in?the?experiment would not be affected by working memory load when other components of WM are manipulated. To further test this hypothesis, a Digital Updating Task (DUT) was adapted into a Digital Alternate Comparison Task (DACT). The goal of the present study is to address the interaction between working memory and task shifting by using Digital Alternate Comparison Task (DACT). Fifty-six college students aged 18-24 were tested through the DACT. The task was divided into high working memory load and low working memory load. In the study phase, participants were presented with two or three initial items, one for each figure (hexagon, square or circle). Participants were instructed to memorize the number for each shape. Then, they would be displayed with a series of new items. For each new item participants had to compare the number for the same shape with initial item. Through the regulation of working memory load, observing the changing of switching cost under different working memory load level. According to previous researches on the correlation between working memory and task shifting. We hypothesized that the reaction time was longer under the switching condition and switching cost increased with the working memory load. The results showed that reaction time under switching condition was significantly higher than that under non-switching condition. The accuracy under switching condition was significantly lower than that under non-switching condition. This indicated that there was more significant accuracy and reaction time deficits under the switching condition. What's more, the reaction time in switching condition under high working memory load was significantly longer than that under low working memory load. The accuracy under high working memory load was significantly lower than that under low working memory load. And the switching cost under high working memory load was also significantly higher than that under low working memory load. This indicated that the switching cost increased along with working memory load. We discussed and analyzed the experimental?results?with?the?limited?theory?of?attention?resources?in?details. These?findings?suggested?that?switching?cost?existed?in?DACT.?The?bigger?the?working?memory load,?the?larger?the?shifting?cost?is.?That?is,?these?findings?further?confirmed?the?close?relationship between?working?memory?and?cognitive?flexibility.

Key words: Switching Cost, Working Memory Load, Cognitive Flexibility

摘要： 工作记忆与认知灵活性是执行功能的两个核心成分，但两者之间的关系并不清晰。本研究在一个数字大小交替比较任务中操纵工作记忆负荷，考察转换代价在不同记忆负荷下的差异。结果显示与非转换条件相比，转换条件下的反应时更长，正确率更低。另外，高记忆负荷下的转换代价显著高于低记忆负荷下的转换代价。这些结果表明工作记忆负荷越大，转换越难，代价越大。

关键词: 转换代价 , 工作记忆负荷, 认知灵活性