Abstract—Anticipation increases the efficiency of daily tasks by partial advance activation of neural substrates involved in it. Previous off-line studies have shown the possibility of exploiting this activation for a Brain-Computer Interface (BCI) using electroencephalogram (EEG). In the current paper we report real-time and single trial recognition of this activation using a prototype of anticipation based BCI (aBCI). We report on-line classification accuracies with peak values of 85% and 80%, and with average values of 69.0±7.9% and 58.5±14.1% for subjects 1 and 2, respectively. Posterior off-line analysis showed improved accuracies for both subjects, with an average of 80.5±10.1% and 69.0 ± 10.5% with peak values of 95% and 85% respectively.