There are two opposing views about how information is encoded in the spike trains, with many intermediate standpoint's. One view holds that it is only the average rate, the number of action potentials over some suitable interval (a few hundred msecs to 1 or more seconds), is relevant for information processing. The opposing view, known as temporal coding, argues that the seemingly random patterns of the spikes, even the exact temporal arrangement of interspike intervals both for the single cell as well as between multiple cells, encodes information. The broad idea of temporal coding is supported by evidence from a variety of sensory systems pointing toward the role of spike timing, in particular across ensembles of cells, in encoding various aspects of the stimulus.
A more balanced view would be somewhere in between of the two extremes. On the one hand we realize temporal patterns of the spike trains do encode information, on the other hand, we also acknowledge the random nature of the physiology of neuronal firing, such as the fluctuation of the injected current I (caused by, for example, the number and magnitudes of the synaptic inputs) and the threshold potential Vth.