Abstract
Time and numbers are interconnected in mental representation. Here we investigated whether this interaction between time and number could be modulated by working memory (WM). In experiment 1 participants first memorized a digit and then judged which of the two digits presented in succession were presented for longer (or shorter), with either one or neither of them being the same as the WM digit. The results showed that, when one of two digits was the same as the WM content, the numerical magnitude effect on temporal performance was abolished, whereas the interference effect occurred only when both the two digits were not the same as the WM content. In order to further investigate whether this effect was due to stimuli repetition, we instructed participants in experiment 2 to merely attend to the original digit but not memorize it during a time judgment task. The results did not reveal the modulation of WM content on time–number link. The WM-based interference effect, however, might reflect decision bias rather than perception illusion. In experiment 3 we eliminated response biases by using equality judgments, and found that WM content had no impact on time–number link. Finally, we conducted experiment 4 (comparative duration judgment) and experiment 5 (equality duration judgment) by manipulating WM load, and demonstrated that the WM-based interference effect was relevant to the duration of the WM content solely in comparative judgment. We concluded that the mental association between time and number representation could be modulated by WM at decision level.