In human crowds, interactions among individuals give rise to a variety of self-organized collective motions that help the group to effectively solve the problem of coordination. However, it is still not known how exactly are the humans adjusting their behavior locally, nor what are the direct consequences on the emergent organization. One of the underlying mechanisms of adjusting individual motions is the stepping dynamics. In this paper, we present first quantitative analysis on the stepping behavior in a one-dimensional pedestrian flow studied under controlled laboratory conditions. We find that the step length is proportional to the velocity of the pedestrian, and is directly related to the space available in front of him, while the variations of the step duration are much weaker. Furthermore, we study the phenomena of synchronization --walking in lock-steps-- and show its dependence on the flow densities. We show that the synchronization of steps is particularly important at high densities, which has direct impact on the studies of optimizing pedestrians flow in congested situations. However, some small synchronization and antisynchronization effects are found even at very low densities, for which no steric constraints exist between successive pedestrians, showing the natural tendency to synchronize according to the perceived visual signals.