A common approach to measure and assess cortical dynamics focuses on the analysis of mass signals, such as the local field potential (LFP), as an indicator for the underlying network activity. To improve our understanding of how such field potentials and cortical spiking dynamics are related, we analyzed the phase and amplitude relationships between extracellular recordings from motor cortex of monkey in a delayed pointing task. We applied methods from phase synchronization analysis to extract the instantaneous phase of the LFP time series and to characterize the degree of phase coupling between the spike train and oscillation cycles in a frequency-independent manner. In particular, we investigated the dependence of observed phase preferences on the different periods of a behavioral trial. Furthermore, we present evidence to support the hypothesis that increased LFP oscillation amplitudes are related to a stronger degree of synchronization between the LFP and spike signals. However, neurons tend to keep a fixed phase relationship to the LFP independent of the amplitude or the choice of the electrode used to record the LFP.