It is fair to say that the raison d'être for science is to identify causations. This despite of the serious philosophical difficulties in defining causation and the scientific subtleties encountered when determining specific causal scenarios. Pragmatic information theoretic approaches were pioneered by Granger and have undergone much recent development in relation to studies of time series from complex systems such as the brain or finance. Here we first review the information theoretic and probabilistic approach to causality measures. We explain the difference between direct and indirect measures. Finally, we present an application of some of the measures to a real system. In particular, we discuss the findings on analysis of music performance where causation amongst the musicians and audience was studied by use of EEG time series.