The ability to tell time, anticipate future events, and produce spatiotemporal motor behaviors, are among the most fundamental computations the brain performs. Precisely because of the importance of timing to brain function, we have proposed that timing is decentralized. I will present experimental and computational studies that reveal that on the scale of seconds the brain often uses its inherent neural dynamics in the form of population clocks—including neural sequences—to encode time. Interestingly both timing and working memory (WM) share the requirement of transiently storing information for future use: prospective information in the case of timing and retrospective information in the case of WM. And some of the same neural signatures that have been implicated in timing have also been implicated in working memory. I will also present computational and psychophysical results supporting the hypothesis that in some cases the encoding time and WM may be multiplexed in neural sequences.