Working memory capacity (WMC) is an important indicator that reflects the individual cognitive function of storage and control. There are different measurements for WMC. Automated reading span task and operation span task are usually applied for verbal WMC. Change detection task is widely applied for visual WMC. A great deal of research has clearly demonstrated that higher WMC is advantageous to a wide variety of complex cognitive and academic activities, such as reasoning, comprehension, and problem-solving. Meanwhile, creative thinking is an important part of high-level cognition of human beings. Remote associates test, Unusual Use Test, Insight Problem-solving and Task-unrelated Thoughts are usually applied for the test of creativity. So far, there are some arguments on whether WMC could predict creative thinking. To understand whether and how WMC predicts creativity thinking, the latest studies were summarized. Due to different experimental paradigms and measurements, the previous research provided inconsistent results. It undoubtedly agreed that higher WMC leads to better cognitive control in a cognitive process but conflicted that WMC could predict creative performance. Higher WMC individuals showed better creative performance in many studies which implied a positive prediction （De Dreu, Nijstad, & Baas, 2011; De Dreu, Nijstad, Baas, Wolsink, & Roskes, 2012; De Dreu, Dussel, & Ten Velden, 2015）. However, higher WMC individuals also showed worse creative performance in other studies which implied a negative prediction or unclear prediction (Ash & Wiley, 2006; Furley & Memmert, 2015; DeCaro, Rotar, Kendra, & Beilock, 2010). Higher WMC was considered to hinder creativity because it may lead individuals to employ complex and less-optimal strategies or miss insight solution. To explain the relationship between WMC and creativity, two theoretical models were compared, including the Dual Pathway to Creativity Model (DPCM) and the Amusement Park Model (APT). DPCM described two factors working in creative process: cognitive flexibility and cognitive persistence. It emphasized that they could work alone or work together for creativity. APT originally distinguished domain-generality and domain-specificity for creativity. It was most hypothesized and could roughly explain inconsistent prediction of WMC to creative thinking too. Generally, creative performance was attributed to different levels of cognitive control between higher and lower WMC individuals. Based on the discussion of the relationship between WMC and creative thinking, those potential issues in the previous researches were presented. They included at least five aspects as following: mismatches of WMC tasks and creative thinking measurements, low ecological validity in the experimental situation, ignoring the covariance of intelligence, ignoring the impact from personality and motivation, and lacking multiple evaluations of creative thinking linked to long term memory. Finally the corresponding suggestions of the future studies were discussed. Researchers need to not only considering verbal or visual WMC measurements but also matching corresponding creative tasks. It is important to explore the prediction of WMC to creativity in real life and pay more attention to normal people and patients. Addition, integrated research methods for individual differences are also necessary.