The main objective of this thesis is to analyze the timing organization of the brain. The human brain is known to adjust its localized and also the reciprocal operations for each different cognitive task adaptively. This flexibility of the brain has attracted considerable interest in neuroscience. Elucidation of timing adaptation property of brain, however, remains as unresolved due to dynamically changing and nonlinear nature of the brain. In this thesis, we characterize the timing organization of the brain during motor imagery activity using electroencephalography signals. First, we propose a novel motor imagery activity recognition method that relies on the activity-specific time-lag between electroencephalography signals obtained from different brain regions. Next, we generalize this approach into three-parameter formulation to determine the timing profiles of activity-specific short-lived synchronization. The identification of activity-specific timing parameters was carried out using a heuristic approach that maximizes the average pairwise channel synchronizations during associated activity periods. Thereafter, we propose a novel BCI framework that find and use the timings of electroencephalography signals of localized brain regions that elicit localized activity-specific complexity features. We identify the timings for each different brain regions by adopting a heuristic-probabilistic method. Finally, we propose a novel autoregressive modeling framework that finds a representative model for each different cognitive activity. We demonstrated the efficacy of the proposed methods on publicly available brain-computer interfacing datasets on motor imagery. The performance results indicate that considering the timing organization of the brain is crucial for accurate characterization of cognitive activity. In addition, it may also account for the inconsistency of brain computer interfacing performance obtained from different subjects.